Compare commits

..

1 Commits

Author SHA1 Message Date
kelly
2e22b439e0 Merge branch 'master' into fix/analytics-v2-queries 2025-12-11 02:55:13 +00:00
133 changed files with 4301 additions and 10965 deletions

View File

@@ -1,3 +1,6 @@
when:
- event: [push, pull_request]
steps:
# ===========================================
# PR VALIDATION: Parallel type checks (PRs only)
@@ -69,7 +72,6 @@ steps:
# ===========================================
# MASTER DEPLOY: Parallel Docker builds
# NOTE: cache_from/cache_to removed due to plugin bug splitting on commas
# ===========================================
docker-backend:
image: woodpeckerci/plugin-docker-buildx
@@ -161,7 +163,7 @@ steps:
event: push
# ===========================================
# STAGE 3: Deploy and Run Migrations
# STAGE 3: Deploy (after Docker builds)
# ===========================================
deploy:
image: bitnami/kubectl:latest
@@ -172,17 +174,12 @@ steps:
- mkdir -p ~/.kube
- echo "$KUBECONFIG_CONTENT" | tr -d '[:space:]' | base64 -d > ~/.kube/config
- chmod 600 ~/.kube/config
# Deploy backend first
- kubectl set image deployment/scraper scraper=code.cannabrands.app/creationshop/dispensary-scraper:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl rollout status deployment/scraper -n dispensary-scraper --timeout=300s
# Note: Migrations run automatically at startup via auto-migrate
# Deploy remaining services
# Resilience: ensure workers are scaled up if at 0
- REPLICAS=$(kubectl get deployment scraper-worker -n dispensary-scraper -o jsonpath='{.spec.replicas}'); if [ "$REPLICAS" = "0" ]; then echo "Scaling workers from 0 to 5"; kubectl scale deployment/scraper-worker --replicas=5 -n dispensary-scraper; fi
- kubectl set image deployment/scraper-worker worker=code.cannabrands.app/creationshop/dispensary-scraper:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl set image deployment/cannaiq-frontend cannaiq-frontend=code.cannabrands.app/creationshop/cannaiq-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl set image deployment/findadispo-frontend findadispo-frontend=code.cannabrands.app/creationshop/findadispo-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl set image deployment/findagram-frontend findagram-frontend=code.cannabrands.app/creationshop/findagram-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl rollout status deployment/scraper -n dispensary-scraper --timeout=300s
- kubectl rollout status deployment/cannaiq-frontend -n dispensary-scraper --timeout=120s
depends_on:
- docker-backend

View File

@@ -1,191 +0,0 @@
steps:
# ===========================================
# PR VALIDATION: Only typecheck changed projects
# ===========================================
typecheck-backend:
image: code.cannabrands.app/creationshop/node:20
commands:
- npm config set cache /npm-cache/backend --global
- cd backend
- npm ci --prefer-offline
- npx tsc --noEmit
volumes:
- npm-cache:/npm-cache
depends_on: []
when:
event: pull_request
path:
include: ['backend/**']
typecheck-cannaiq:
image: code.cannabrands.app/creationshop/node:20
commands:
- npm config set cache /npm-cache/cannaiq --global
- cd cannaiq
- npm ci --prefer-offline
- npx tsc --noEmit
volumes:
- npm-cache:/npm-cache
depends_on: []
when:
event: pull_request
path:
include: ['cannaiq/**']
# findadispo/findagram typechecks skipped - they have || true anyway
# ===========================================
# AUTO-MERGE: Merge PR after all checks pass
# ===========================================
auto-merge:
image: alpine:latest
environment:
GITEA_TOKEN:
from_secret: gitea_token
commands:
- apk add --no-cache curl
- |
echo "Merging PR #${CI_COMMIT_PULL_REQUEST}..."
curl -s -X POST \
-H "Authorization: token $GITEA_TOKEN" \
-H "Content-Type: application/json" \
-d '{"Do":"merge"}' \
"https://code.cannabrands.app/api/v1/repos/Creationshop/dispensary-scraper/pulls/${CI_COMMIT_PULL_REQUEST}/merge"
depends_on:
- typecheck-backend
- typecheck-cannaiq
when:
event: pull_request
# ===========================================
# MASTER DEPLOY: Parallel Docker builds
# ===========================================
docker-backend:
image: woodpeckerci/plugin-docker-buildx
settings:
registry: code.cannabrands.app
repo: code.cannabrands.app/creationshop/dispensary-scraper
tags:
- latest
- ${CI_COMMIT_SHA:0:8}
dockerfile: backend/Dockerfile
context: backend
username:
from_secret: registry_username
password:
from_secret: registry_password
platforms: linux/amd64
provenance: false
cache_from: type=registry,ref=code.cannabrands.app/creationshop/dispensary-scraper:cache
cache_to: type=registry,ref=code.cannabrands.app/creationshop/dispensary-scraper:cache,mode=max
build_args:
APP_BUILD_VERSION: ${CI_COMMIT_SHA:0:8}
APP_GIT_SHA: ${CI_COMMIT_SHA}
APP_BUILD_TIME: ${CI_PIPELINE_CREATED}
CONTAINER_IMAGE_TAG: ${CI_COMMIT_SHA:0:8}
depends_on: []
when:
branch: master
event: push
docker-cannaiq:
image: woodpeckerci/plugin-docker-buildx
settings:
registry: code.cannabrands.app
repo: code.cannabrands.app/creationshop/cannaiq-frontend
tags:
- latest
- ${CI_COMMIT_SHA:0:8}
dockerfile: cannaiq/Dockerfile
context: cannaiq
username:
from_secret: registry_username
password:
from_secret: registry_password
platforms: linux/amd64
provenance: false
cache_from: type=registry,ref=code.cannabrands.app/creationshop/cannaiq-frontend:cache
cache_to: type=registry,ref=code.cannabrands.app/creationshop/cannaiq-frontend:cache,mode=max
depends_on: []
when:
branch: master
event: push
docker-findadispo:
image: woodpeckerci/plugin-docker-buildx
settings:
registry: code.cannabrands.app
repo: code.cannabrands.app/creationshop/findadispo-frontend
tags:
- latest
- ${CI_COMMIT_SHA:0:8}
dockerfile: findadispo/frontend/Dockerfile
context: findadispo/frontend
username:
from_secret: registry_username
password:
from_secret: registry_password
platforms: linux/amd64
provenance: false
cache_from: type=registry,ref=code.cannabrands.app/creationshop/findadispo-frontend:cache
cache_to: type=registry,ref=code.cannabrands.app/creationshop/findadispo-frontend:cache,mode=max
depends_on: []
when:
branch: master
event: push
docker-findagram:
image: woodpeckerci/plugin-docker-buildx
settings:
registry: code.cannabrands.app
repo: code.cannabrands.app/creationshop/findagram-frontend
tags:
- latest
- ${CI_COMMIT_SHA:0:8}
dockerfile: findagram/frontend/Dockerfile
context: findagram/frontend
username:
from_secret: registry_username
password:
from_secret: registry_password
platforms: linux/amd64
provenance: false
cache_from: type=registry,ref=code.cannabrands.app/creationshop/findagram-frontend:cache
cache_to: type=registry,ref=code.cannabrands.app/creationshop/findagram-frontend:cache,mode=max
depends_on: []
when:
branch: master
event: push
# ===========================================
# STAGE 3: Deploy and Run Migrations
# ===========================================
deploy:
image: bitnami/kubectl:latest
environment:
KUBECONFIG_CONTENT:
from_secret: kubeconfig_data
commands:
- mkdir -p ~/.kube
- echo "$KUBECONFIG_CONTENT" | tr -d '[:space:]' | base64 -d > ~/.kube/config
- chmod 600 ~/.kube/config
# Deploy backend first
- kubectl set image deployment/scraper scraper=code.cannabrands.app/creationshop/dispensary-scraper:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl rollout status deployment/scraper -n dispensary-scraper --timeout=300s
# Note: Migrations run automatically at startup via auto-migrate
# Deploy remaining services
# Resilience: ensure workers are scaled up if at 0
- REPLICAS=$(kubectl get deployment scraper-worker -n dispensary-scraper -o jsonpath='{.spec.replicas}'); if [ "$REPLICAS" = "0" ]; then echo "Scaling workers from 0 to 5"; kubectl scale deployment/scraper-worker --replicas=5 -n dispensary-scraper; fi
- kubectl set image deployment/scraper-worker worker=code.cannabrands.app/creationshop/dispensary-scraper:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl set image deployment/cannaiq-frontend cannaiq-frontend=code.cannabrands.app/creationshop/cannaiq-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl set image deployment/findadispo-frontend findadispo-frontend=code.cannabrands.app/creationshop/findadispo-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl set image deployment/findagram-frontend findagram-frontend=code.cannabrands.app/creationshop/findagram-frontend:${CI_COMMIT_SHA:0:8} -n dispensary-scraper
- kubectl rollout status deployment/cannaiq-frontend -n dispensary-scraper --timeout=120s
depends_on:
- docker-backend
- docker-cannaiq
- docker-findadispo
- docker-findagram
when:
branch: master
event: push

1444
CLAUDE.md

File diff suppressed because it is too large Load Diff

View File

@@ -25,9 +25,8 @@ ENV APP_GIT_SHA=${APP_GIT_SHA}
ENV APP_BUILD_TIME=${APP_BUILD_TIME}
ENV CONTAINER_IMAGE_TAG=${CONTAINER_IMAGE_TAG}
# Install Chromium dependencies and curl for HTTP requests
# Install Chromium dependencies
RUN apt-get update && apt-get install -y \
curl \
chromium \
fonts-liberation \
libnss3 \

View File

@@ -1,218 +0,0 @@
# CannaiQ Backend Codebase Map
**Last Updated:** 2025-12-12
**Purpose:** Help Claude and developers understand which code is current vs deprecated
---
## Quick Reference: What to Use
### For Crawling/Scraping
| Task | Use This | NOT This |
|------|----------|----------|
| Fetch products | `src/tasks/handlers/payload-fetch.ts` | `src/hydration/*` |
| Process products | `src/tasks/handlers/product-refresh.ts` | `src/scraper-v2/*` |
| GraphQL client | `src/platforms/dutchie/client.ts` | `src/dutchie-az/services/graphql-client.ts` |
| Worker system | `src/tasks/task-worker.ts` | `src/dutchie-az/services/worker.ts` |
### For Database
| Task | Use This | NOT This |
|------|----------|----------|
| Get DB pool | `src/db/pool.ts` | `src/dutchie-az/db/connection.ts` |
| Run migrations | `src/db/migrate.ts` (CLI only) | Never import at runtime |
| Query products | `store_products` table | `products`, `dutchie_products` |
| Query stores | `dispensaries` table | `stores` table |
### For Discovery
| Task | Use This |
|------|----------|
| Discover stores | `src/discovery/*.ts` |
| Run discovery | `npx tsx src/scripts/run-discovery.ts` |
---
## Directory Status
### ACTIVE DIRECTORIES (Use These)
```
src/
├── auth/ # JWT/session auth, middleware
├── db/ # Database pool, migrations
├── discovery/ # Dutchie store discovery pipeline
├── middleware/ # Express middleware
├── multi-state/ # Multi-state query support
├── platforms/ # Platform-specific clients (Dutchie, Jane, etc)
│ └── dutchie/ # THE Dutchie client - use this one
├── routes/ # Express API routes
├── services/ # Core services (logger, scheduler, etc)
├── tasks/ # Task system (workers, handlers, scheduler)
│ └── handlers/ # Task handlers (payload_fetch, product_refresh, etc)
├── types/ # TypeScript types
└── utils/ # Utilities (storage, image processing)
```
### DEPRECATED DIRECTORIES (DO NOT USE)
```
src/
├── hydration/ # DEPRECATED - Old pipeline approach
├── scraper-v2/ # DEPRECATED - Old scraper engine
├── canonical-hydration/# DEPRECATED - Merged into tasks/handlers
├── dutchie-az/ # PARTIAL - Some parts deprecated, some active
│ ├── db/ # DEPRECATED - Use src/db/pool.ts
│ └── services/ # PARTIAL - worker.ts still runs, graphql-client.ts deprecated
├── portals/ # FUTURE - Not yet implemented
├── seo/ # PARTIAL - Settings work, templates WIP
└── system/ # DEPRECATED - Old orchestration system
```
### DEPRECATED FILES (DO NOT USE)
```
src/dutchie-az/db/connection.ts # Use src/db/pool.ts instead
src/dutchie-az/services/graphql-client.ts # Use src/platforms/dutchie/client.ts
src/hydration/*.ts # Entire directory deprecated
src/scraper-v2/*.ts # Entire directory deprecated
```
---
## Key Files Reference
### Entry Points
| File | Purpose | Status |
|------|---------|--------|
| `src/index.ts` | Main Express server | ACTIVE |
| `src/dutchie-az/services/worker.ts` | Worker process entry | ACTIVE |
| `src/tasks/task-worker.ts` | Task worker (new system) | ACTIVE |
### Dutchie Integration
| File | Purpose | Status |
|------|---------|--------|
| `src/platforms/dutchie/client.ts` | GraphQL client, hashes, curl | **PRIMARY** |
| `src/platforms/dutchie/queries.ts` | High-level query functions | ACTIVE |
| `src/platforms/dutchie/index.ts` | Re-exports | ACTIVE |
### Task Handlers
| File | Purpose | Status |
|------|---------|--------|
| `src/tasks/handlers/payload-fetch.ts` | Fetch products from Dutchie | **PRIMARY** |
| `src/tasks/handlers/product-refresh.ts` | Process payload into DB | **PRIMARY** |
| `src/tasks/handlers/menu-detection.ts` | Detect menu type | ACTIVE |
| `src/tasks/handlers/id-resolution.ts` | Resolve platform IDs | ACTIVE |
| `src/tasks/handlers/image-download.ts` | Download product images | ACTIVE |
### Database
| File | Purpose | Status |
|------|---------|--------|
| `src/db/pool.ts` | Canonical DB pool | **PRIMARY** |
| `src/db/migrate.ts` | Migration runner (CLI only) | CLI ONLY |
| `src/db/auto-migrate.ts` | Auto-run migrations on startup | ACTIVE |
### Configuration
| File | Purpose | Status |
|------|---------|--------|
| `.env` | Environment variables | ACTIVE |
| `package.json` | Dependencies | ACTIVE |
| `tsconfig.json` | TypeScript config | ACTIVE |
---
## GraphQL Hashes (CRITICAL)
The correct hashes are in `src/platforms/dutchie/client.ts`:
```typescript
export const GRAPHQL_HASHES = {
FilteredProducts: 'ee29c060826dc41c527e470e9ae502c9b2c169720faa0a9f5d25e1b9a530a4a0',
GetAddressBasedDispensaryData: '13461f73abf7268770dfd05fe7e10c523084b2bb916a929c08efe3d87531977b',
ConsumerDispensaries: '0a5bfa6ca1d64ae47bcccb7c8077c87147cbc4e6982c17ceec97a2a4948b311b',
GetAllCitiesByState: 'ae547a0466ace5a48f91e55bf6699eacd87e3a42841560f0c0eabed5a0a920e6',
};
```
**ALWAYS** use `Status: 'Active'` for FilteredProducts (not `null` or `'All'`).
---
## Scripts Reference
### Useful Scripts (in `src/scripts/`)
| Script | Purpose |
|--------|---------|
| `run-discovery.ts` | Run Dutchie discovery |
| `crawl-single-store.ts` | Test crawl a single store |
| `test-dutchie-graphql.ts` | Test GraphQL queries |
### One-Off Scripts (probably don't need)
| Script | Purpose |
|--------|---------|
| `harmonize-az-dispensaries.ts` | One-time data cleanup |
| `bootstrap-stores-for-dispensaries.ts` | One-time migration |
| `backfill-*.ts` | Historical backfill scripts |
---
## API Routes
### Active Routes (in `src/routes/`)
| Route File | Mount Point | Purpose |
|------------|-------------|---------|
| `auth.ts` | `/api/auth` | Login/logout/session |
| `stores.ts` | `/api/stores` | Store CRUD |
| `dashboard.ts` | `/api/dashboard` | Dashboard stats |
| `workers.ts` | `/api/workers` | Worker monitoring |
| `pipeline.ts` | `/api/pipeline` | Crawl triggers |
| `discovery.ts` | `/api/discovery` | Discovery management |
| `analytics.ts` | `/api/analytics` | Analytics queries |
| `wordpress.ts` | `/api/v1/wordpress` | WordPress plugin API |
---
## Documentation Files
### Current Docs (in `backend/docs/`)
| Doc | Purpose | Currency |
|-----|---------|----------|
| `TASK_WORKFLOW_2024-12-10.md` | Task system architecture | CURRENT |
| `WORKER_TASK_ARCHITECTURE.md` | Worker/task design | CURRENT |
| `CRAWL_PIPELINE.md` | Crawl pipeline overview | CURRENT |
| `ORGANIC_SCRAPING_GUIDE.md` | Browser-based scraping | CURRENT |
| `CODEBASE_MAP.md` | This file | CURRENT |
| `ANALYTICS_V2_EXAMPLES.md` | Analytics API examples | CURRENT |
| `BRAND_INTELLIGENCE_API.md` | Brand API docs | CURRENT |
### Root Docs
| Doc | Purpose | Currency |
|-----|---------|----------|
| `CLAUDE.md` | Claude instructions | **PRIMARY** |
| `README.md` | Project overview | NEEDS UPDATE |
---
## Common Mistakes to Avoid
1. **Don't use `src/hydration/`** - It's an old approach that was superseded by the task system
2. **Don't use `src/dutchie-az/db/connection.ts`** - Use `src/db/pool.ts` instead
3. **Don't import `src/db/migrate.ts` at runtime** - It will crash. Only use for CLI migrations.
4. **Don't query `stores` table** - It's empty. Use `dispensaries`.
5. **Don't query `products` table** - It's empty. Use `store_products`.
6. **Don't use wrong GraphQL hash** - Always get hash from `GRAPHQL_HASHES` in client.ts
7. **Don't use `Status: null`** - It returns 0 products. Use `Status: 'Active'`.
---
## When in Doubt
1. Check if the file is imported in `src/index.ts` - if not, it may be deprecated
2. Check the last modified date - older files may be stale
3. Look for `DEPRECATED` comments in the code
4. Ask: "Is there a newer version of this in `src/tasks/` or `src/platforms/`?"
5. Read the relevant doc in `docs/` before modifying code

View File

@@ -500,18 +500,17 @@ CREATE TABLE proxies (
Proxies are mandatory. There is no environment variable to disable them. Workers will refuse to start without active proxies in the database.
### User-Agent Generation
### Fingerprints Available
See `workflow-12102025.md` for full specification.
The client includes 6 browser fingerprints:
- Chrome 131 on Windows
- Chrome 131 on macOS
- Chrome 120 on Windows
- Firefox 133 on Windows
- Safari 17.2 on macOS
- Edge 131 on Windows
**Summary:**
- Uses `intoli/user-agents` library (daily-updated market share data)
- Device distribution: Mobile 62%, Desktop 36%, Tablet 2%
- Browser whitelist: Chrome, Safari, Edge, Firefox only
- UA sticks until IP rotates (403 or manual rotation)
- Failure = alert admin + stop crawl (no fallback)
Each fingerprint includes proper `sec-ch-ua`, `sec-ch-ua-platform`, and `sec-ch-ua-mobile` headers.
Each includes proper `sec-ch-ua`, `sec-ch-ua-platform`, and `sec-ch-ua-mobile` headers.
---

View File

@@ -362,148 +362,6 @@ SET status = 'pending', retry_count = retry_count + 1
WHERE status = 'failed' AND retry_count < max_retries;
```
## Concurrent Task Processing (Added 2024-12)
Workers can now process multiple tasks concurrently within a single worker instance. This improves throughput by utilizing async I/O efficiently.
### Architecture
```
┌─────────────────────────────────────────────────────────────┐
│ Pod (K8s) │
│ │
│ ┌─────────────────────────────────────────────────────┐ │
│ │ TaskWorker │ │
│ │ │ │
│ │ ┌─────────┐ ┌─────────┐ ┌─────────┐ │ │
│ │ │ Task 1 │ │ Task 2 │ │ Task 3 │ (concurrent)│ │
│ │ └─────────┘ └─────────┘ └─────────┘ │ │
│ │ │ │
│ │ Resource Monitor │ │
│ │ ├── Memory: 65% (threshold: 85%) │ │
│ │ ├── CPU: 45% (threshold: 90%) │ │
│ │ └── Status: Normal │ │
│ └─────────────────────────────────────────────────────┘ │
└─────────────────────────────────────────────────────────────┘
```
### Environment Variables
| Variable | Default | Description |
|----------|---------|-------------|
| `MAX_CONCURRENT_TASKS` | 3 | Maximum tasks a worker will run concurrently |
| `MEMORY_BACKOFF_THRESHOLD` | 0.85 | Back off when heap memory exceeds 85% |
| `CPU_BACKOFF_THRESHOLD` | 0.90 | Back off when CPU exceeds 90% |
| `BACKOFF_DURATION_MS` | 10000 | How long to wait when backing off (10s) |
### How It Works
1. **Main Loop**: Worker continuously tries to fill up to `MAX_CONCURRENT_TASKS`
2. **Resource Monitoring**: Before claiming a new task, worker checks memory and CPU
3. **Backoff**: If resources exceed thresholds, worker pauses and stops claiming new tasks
4. **Concurrent Execution**: Tasks run in parallel using `Promise` - they don't block each other
5. **Graceful Shutdown**: On SIGTERM/decommission, worker stops claiming but waits for active tasks
### Resource Monitoring
```typescript
// ResourceStats interface
interface ResourceStats {
memoryPercent: number; // Current heap usage as decimal (0.0-1.0)
memoryMb: number; // Current heap used in MB
memoryTotalMb: number; // Total heap available in MB
cpuPercent: number; // CPU usage as percentage (0-100)
isBackingOff: boolean; // True if worker is in backoff state
backoffReason: string; // Why the worker is backing off
}
```
### Heartbeat Data
Workers report the following in their heartbeat:
```json
{
"worker_id": "worker-abc123",
"current_task_id": 456,
"current_task_ids": [456, 457, 458],
"active_task_count": 3,
"max_concurrent_tasks": 3,
"status": "active",
"resources": {
"memory_mb": 256,
"memory_total_mb": 512,
"memory_rss_mb": 320,
"memory_percent": 50,
"cpu_user_ms": 12500,
"cpu_system_ms": 3200,
"cpu_percent": 45,
"is_backing_off": false,
"backoff_reason": null
}
}
```
### Backoff Behavior
When resources exceed thresholds:
1. Worker logs the backoff reason:
```
[TaskWorker] MyWorker backing off: Memory at 87.3% (threshold: 85%)
```
2. Worker stops claiming new tasks but continues existing tasks
3. After `BACKOFF_DURATION_MS`, worker rechecks resources
4. When resources return to normal:
```
[TaskWorker] MyWorker resuming normal operation
```
### UI Display
The Workers Dashboard shows:
- **Tasks Column**: `2/3 tasks` (active/max concurrent)
- **Resources Column**: Memory % and CPU % with color coding
- Green: < 50%
- Yellow: 50-74%
- Amber: 75-89%
- Red: 90%+
- **Backing Off**: Orange warning badge when worker is in backoff state
### Task Count Badge Details
```
┌─────────────────────────────────────────────┐
│ Worker: "MyWorker" │
│ Tasks: 2/3 tasks #456, #457
│ Resources: 🧠 65% 💻 45% │
│ Status: ● Active │
└─────────────────────────────────────────────┘
```
### Best Practices
1. **Start Conservative**: Use `MAX_CONCURRENT_TASKS=3` initially
2. **Monitor Resources**: Watch for frequent backoffs in logs
3. **Tune Per Workload**: I/O-bound tasks benefit from higher concurrency
4. **Scale Horizontally**: Add more pods rather than cranking concurrency too high
### Code References
| File | Purpose |
|------|---------|
| `src/tasks/task-worker.ts:68-71` | Concurrency environment variables |
| `src/tasks/task-worker.ts:104-111` | ResourceStats interface |
| `src/tasks/task-worker.ts:149-179` | getResourceStats() method |
| `src/tasks/task-worker.ts:184-196` | shouldBackOff() method |
| `src/tasks/task-worker.ts:462-516` | mainLoop() with concurrent claiming |
| `src/routes/worker-registry.ts:148-195` | Heartbeat endpoint handling |
| `cannaiq/src/pages/WorkersDashboard.tsx:233-305` | UI components for resources |
## Monitoring
### Logs

View File

@@ -1,297 +0,0 @@
# Organic Browser-Based Scraping Guide
**Last Updated:** 2025-12-12
**Status:** Production-ready proof of concept
---
## Overview
This document describes the "organic" browser-based approach to scraping Dutchie dispensary menus. Unlike direct curl/axios requests, this method uses a real browser session to make API calls, making requests appear natural and reducing detection risk.
---
## Why Organic Scraping?
| Approach | Detection Risk | Speed | Complexity |
|----------|---------------|-------|------------|
| Direct curl | Higher | Fast | Low |
| curl-impersonate | Medium | Fast | Medium |
| **Browser-based (organic)** | **Lowest** | Slower | Higher |
Direct curl requests can be fingerprinted via:
- TLS fingerprint (cipher suites, extensions)
- Header order and values
- Missing cookies/session data
- Request patterns
Browser-based requests inherit:
- Real Chrome TLS fingerprint
- Session cookies from page visit
- Natural header order
- JavaScript execution environment
---
## Implementation
### Dependencies
```bash
npm install puppeteer puppeteer-extra puppeteer-extra-plugin-stealth
```
### Core Script: `test-intercept.js`
Located at: `backend/test-intercept.js`
```javascript
const puppeteer = require('puppeteer-extra');
const StealthPlugin = require('puppeteer-extra-plugin-stealth');
const fs = require('fs');
puppeteer.use(StealthPlugin());
async function capturePayload(config) {
const { dispensaryId, platformId, cName, outputPath } = config;
const browser = await puppeteer.launch({
headless: 'new',
args: ['--no-sandbox', '--disable-setuid-sandbox']
});
const page = await browser.newPage();
// STEP 1: Establish session by visiting the menu
const embedUrl = `https://dutchie.com/embedded-menu/${cName}?menuType=rec`;
await page.goto(embedUrl, { waitUntil: 'networkidle2', timeout: 60000 });
// STEP 2: Fetch ALL products using GraphQL from browser context
const result = await page.evaluate(async (platformId) => {
const allProducts = [];
let pageNum = 0;
const perPage = 100;
let totalCount = 0;
const sessionId = 'browser-session-' + Date.now();
while (pageNum < 30) {
const variables = {
includeEnterpriseSpecials: false,
productsFilter: {
dispensaryId: platformId,
pricingType: 'rec',
Status: 'Active', // CRITICAL: Must be 'Active', not null
types: [],
useCache: true,
isDefaultSort: true,
sortBy: 'popularSortIdx',
sortDirection: 1,
bypassOnlineThresholds: true,
isKioskMenu: false,
removeProductsBelowOptionThresholds: false,
},
page: pageNum,
perPage: perPage,
};
const extensions = {
persistedQuery: {
version: 1,
sha256Hash: 'ee29c060826dc41c527e470e9ae502c9b2c169720faa0a9f5d25e1b9a530a4a0'
}
};
const qs = new URLSearchParams({
operationName: 'FilteredProducts',
variables: JSON.stringify(variables),
extensions: JSON.stringify(extensions)
});
const response = await fetch(`https://dutchie.com/api-3/graphql?${qs}`, {
method: 'GET',
headers: {
'Accept': 'application/json',
'content-type': 'application/json',
'x-dutchie-session': sessionId,
'apollographql-client-name': 'Marketplace (production)',
},
credentials: 'include'
});
const json = await response.json();
const data = json?.data?.filteredProducts;
if (!data?.products) break;
allProducts.push(...data.products);
if (pageNum === 0) totalCount = data.queryInfo?.totalCount || 0;
if (allProducts.length >= totalCount) break;
pageNum++;
await new Promise(r => setTimeout(r, 200)); // Polite delay
}
return { products: allProducts, totalCount };
}, platformId);
await browser.close();
// STEP 3: Save payload
const payload = {
dispensaryId,
platformId,
cName,
fetchedAt: new Date().toISOString(),
productCount: result.products.length,
products: result.products,
};
fs.writeFileSync(outputPath, JSON.stringify(payload, null, 2));
return payload;
}
```
---
## Critical Parameters
### GraphQL Hash (FilteredProducts)
```
ee29c060826dc41c527e470e9ae502c9b2c169720faa0a9f5d25e1b9a530a4a0
```
**WARNING:** Using the wrong hash returns HTTP 400.
### Status Parameter
| Value | Result |
|-------|--------|
| `'Active'` | Returns in-stock products (1019 in test) |
| `null` | Returns 0 products |
| `'All'` | Returns HTTP 400 |
**ALWAYS use `Status: 'Active'`**
### Required Headers
```javascript
{
'Accept': 'application/json',
'content-type': 'application/json',
'x-dutchie-session': 'unique-session-id',
'apollographql-client-name': 'Marketplace (production)',
}
```
### Endpoint
```
https://dutchie.com/api-3/graphql
```
---
## Performance Benchmarks
Test store: AZ-Deeply-Rooted (1019 products)
| Metric | Value |
|--------|-------|
| Total products | 1019 |
| Time | 18.5 seconds |
| Payload size | 11.8 MB |
| Pages fetched | 11 (100 per page) |
| Success rate | 100% |
---
## Payload Format
The output matches the existing `payload-fetch.ts` handler format:
```json
{
"dispensaryId": 123,
"platformId": "6405ef617056e8014d79101b",
"cName": "AZ-Deeply-Rooted",
"fetchedAt": "2025-12-12T05:05:19.837Z",
"productCount": 1019,
"products": [
{
"id": "6927508db4851262f629a869",
"Name": "Product Name",
"brand": { "name": "Brand Name", ... },
"type": "Flower",
"THC": "25%",
"Prices": [...],
"Options": [...],
...
}
]
}
```
---
## Integration Points
### As a Task Handler
The organic approach can be integrated as an alternative to curl-based fetching:
```typescript
// In src/tasks/handlers/organic-payload-fetch.ts
export async function handleOrganicPayloadFetch(ctx: TaskContext): Promise<TaskResult> {
// Use puppeteer-based capture
// Save to same payload storage
// Queue product_refresh task
}
```
### Worker Configuration
Add to job_schedules:
```sql
INSERT INTO job_schedules (name, role, cron_expression)
VALUES ('organic_product_crawl', 'organic_payload_fetch', '0 */6 * * *');
```
---
## Troubleshooting
### HTTP 400 Bad Request
- Check hash is correct: `ee29c060...`
- Verify Status is `'Active'` (string, not null)
### 0 Products Returned
- Status was likely `null` or `'All'` - use `'Active'`
- Check platformId is valid MongoDB ObjectId
### Session Not Established
- Increase timeout on initial page.goto()
- Check cName is valid (matches embedded-menu URL)
### Detection/Blocking
- StealthPlugin should handle most cases
- Add random delays between pages
- Use headless: 'new' (not true/false)
---
## Files Reference
| File | Purpose |
|------|---------|
| `backend/test-intercept.js` | Proof of concept script |
| `backend/src/platforms/dutchie/client.ts` | GraphQL hashes, curl implementation |
| `backend/src/tasks/handlers/payload-fetch.ts` | Current curl-based handler |
| `backend/src/utils/payload-storage.ts` | Payload save/load utilities |
---
## See Also
- `DUTCHIE_CRAWL_WORKFLOW.md` - Full crawl pipeline documentation
- `TASK_WORKFLOW_2024-12-10.md` - Task system architecture
- `CLAUDE.md` - Project rules and constraints

View File

@@ -1,25 +0,0 @@
# ARCHIVED DOCUMENTATION
**WARNING: These docs may be outdated or inaccurate.**
The code has evolved significantly. These docs are kept for historical reference only.
## What to Use Instead
**The single source of truth is:**
- `CLAUDE.md` (root) - Essential rules and quick reference
- `docs/CODEBASE_MAP.md` - Current file/directory reference
## Why Archive?
These docs were written during development iterations and may reference:
- Old file paths that no longer exist
- Deprecated approaches (hydration, scraper-v2)
- APIs that have changed
- Database schemas that evolved
## If You Need Details
1. First check CODEBASE_MAP.md for current file locations
2. Then read the actual source code
3. Only use archive docs as a last resort for historical context

View File

@@ -1,584 +0,0 @@
# Task Workflow Documentation
**Date: 2024-12-10**
This document describes the complete task/job processing architecture after the 2024-12-10 rewrite.
---
## Complete Architecture
```
┌─────────────────────────────────────────────────────────────────────────────────┐
│ KUBERNETES CLUSTER │
├─────────────────────────────────────────────────────────────────────────────────┤
│ │
│ ┌─────────────────────────────────────────────────────────────────────────┐ │
│ │ API SERVER POD (scraper) │ │
│ │ │ │
│ │ ┌──────────────────┐ ┌────────────────────────────────────────┐ │ │
│ │ │ Express API │ │ TaskScheduler │ │ │
│ │ │ │ │ (src/services/task-scheduler.ts) │ │ │
│ │ │ /api/job-queue │ │ │ │ │
│ │ │ /api/tasks │ │ • Polls every 60s │ │ │
│ │ │ /api/schedules │ │ • Checks task_schedules table │ │ │
│ │ └────────┬─────────┘ │ • SELECT FOR UPDATE SKIP LOCKED │ │ │
│ │ │ │ • Generates tasks when due │ │ │
│ │ │ └──────────────────┬─────────────────────┘ │ │
│ │ │ │ │ │
│ └────────────┼──────────────────────────────────┼──────────────────────────┘ │
│ │ │ │
│ │ ┌────────────────────────┘ │
│ │ │ │
│ ▼ ▼ │
│ ┌─────────────────────────────────────────────────────────────────────────┐ │
│ │ POSTGRESQL DATABASE │ │
│ │ │ │
│ │ ┌─────────────────────┐ ┌─────────────────────┐ │ │
│ │ │ task_schedules │ │ worker_tasks │ │ │
│ │ │ │ │ │ │ │
│ │ │ • product_refresh │───────►│ • pending tasks │ │ │
│ │ │ • store_discovery │ create │ • claimed tasks │ │ │
│ │ │ • analytics_refresh │ tasks │ • running tasks │ │ │
│ │ │ │ │ • completed tasks │ │ │
│ │ │ next_run_at │ │ │ │ │
│ │ │ last_run_at │ │ role, dispensary_id │ │ │
│ │ │ interval_hours │ │ priority, status │ │ │
│ │ └─────────────────────┘ └──────────┬──────────┘ │ │
│ │ │ │ │
│ └─────────────────────────────────────────────┼────────────────────────────┘ │
│ │ │
│ ┌──────────────────────┘ │
│ │ Workers poll for tasks │
│ │ (SELECT FOR UPDATE SKIP LOCKED) │
│ ▼ │
│ ┌─────────────────────────────────────────────────────────────────────────┐ │
│ │ WORKER PODS (StatefulSet: scraper-worker) │ │
│ │ │ │
│ │ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ │ │
│ │ │ Worker 0 │ │ Worker 1 │ │ Worker 2 │ │ Worker N │ │ │
│ │ │ │ │ │ │ │ │ │ │ │
│ │ │ task-worker │ │ task-worker │ │ task-worker │ │ task-worker │ │ │
│ │ │ .ts │ │ .ts │ │ .ts │ │ .ts │ │ │
│ │ └─────────────┘ └─────────────┘ └─────────────┘ └─────────────┘ │ │
│ │ │ │
│ └──────────────────────────────────────────────────────────────────────────┘ │
│ │
└──────────────────────────────────────────────────────────────────────────────────┘
```
---
## Startup Sequence
```
┌─────────────────────────────────────────────────────────────────────────────┐
│ API SERVER STARTUP │
├─────────────────────────────────────────────────────────────────────────────┤
│ │
│ 1. Express app initializes │
│ │ │
│ ▼ │
│ 2. runAutoMigrations() │
│ • Runs pending migrations (including 079_task_schedules.sql) │
│ │ │
│ ▼ │
│ 3. initializeMinio() / initializeImageStorage() │
│ │ │
│ ▼ │
│ 4. cleanupOrphanedJobs() │
│ │ │
│ ▼ │
│ 5. taskScheduler.start() ◄─── NEW (per TASK_WORKFLOW_2024-12-10.md) │
│ │ │
│ ├── Recover stale tasks (workers that died) │
│ ├── Ensure default schedules exist in task_schedules │
│ ├── Check and run any due schedules immediately │
│ └── Start 60-second poll interval │
│ │ │
│ ▼ │
│ 6. app.listen(PORT) │
│ │
└─────────────────────────────────────────────────────────────────────────────┘
┌─────────────────────────────────────────────────────────────────────────────┐
│ WORKER POD STARTUP │
├─────────────────────────────────────────────────────────────────────────────┤
│ │
│ 1. K8s starts pod from StatefulSet │
│ │ │
│ ▼ │
│ 2. TaskWorker.constructor() │
│ • Create DB pool │
│ • Create CrawlRotator │
│ │ │
│ ▼ │
│ 3. initializeStealth() │
│ • Load proxies from DB (REQUIRED - fails if none) │
│ • Wire rotator to Dutchie client │
│ │ │
│ ▼ │
│ 4. register() with API │
│ • Optional - continues if fails │
│ │ │
│ ▼ │
│ 5. startRegistryHeartbeat() every 30s │
│ │ │
│ ▼ │
│ 6. processNextTask() loop │
│ │ │
│ ├── Poll for pending task (FOR UPDATE SKIP LOCKED) │
│ ├── Claim task atomically │
│ ├── Execute handler (product_refresh, store_discovery, etc.) │
│ ├── Mark complete/failed │
│ ├── Chain next task if applicable │
│ └── Loop │
│ │
└─────────────────────────────────────────────────────────────────────────────┘
```
---
## Schedule Flow
```
┌─────────────────────────────────────────────────────────────────────────────┐
│ SCHEDULER POLL (every 60 seconds) │
├─────────────────────────────────────────────────────────────────────────────┤
│ │
│ BEGIN TRANSACTION │
│ │ │
│ ▼ │
│ SELECT * FROM task_schedules │
│ WHERE enabled = true AND next_run_at <= NOW() │
│ FOR UPDATE SKIP LOCKED ◄─── Prevents duplicate execution across replicas │
│ │ │
│ ▼ │
│ For each due schedule: │
│ │ │
│ ├── product_refresh_all │
│ │ └─► Query dispensaries needing crawl │
│ │ └─► Create product_refresh tasks in worker_tasks │
│ │ │
│ ├── store_discovery_dutchie │
│ │ └─► Create single store_discovery task │
│ │ │
│ └── analytics_refresh │
│ └─► Create single analytics_refresh task │
│ │ │
│ ▼ │
│ UPDATE task_schedules SET │
│ last_run_at = NOW(), │
│ next_run_at = NOW() + interval_hours │
│ │ │
│ ▼ │
│ COMMIT │
│ │
└─────────────────────────────────────────────────────────────────────────────┘
```
---
## Task Lifecycle
```
┌──────────┐
│ SCHEDULE │
│ DUE │
└────┬─────┘
┌──────────────┐ claim ┌──────────────┐ start ┌──────────────┐
│ PENDING │────────────►│ CLAIMED │────────────►│ RUNNING │
└──────────────┘ └──────────────┘ └──────┬───────┘
▲ │
│ ┌──────────────┼──────────────┐
│ retry │ │ │
│ (if retries < max) ▼ ▼ ▼
│ ┌──────────┐ ┌──────────┐ ┌──────────┐
└──────────────────────────────────│ FAILED │ │ COMPLETED│ │ STALE │
└──────────┘ └──────────┘ └────┬─────┘
recover_stale_tasks()
┌──────────┐
│ PENDING │
└──────────┘
```
---
## Database Tables
### task_schedules (NEW - migration 079)
Stores schedule definitions. Survives restarts.
```sql
CREATE TABLE task_schedules (
id SERIAL PRIMARY KEY,
name VARCHAR(100) NOT NULL UNIQUE,
role VARCHAR(50) NOT NULL, -- product_refresh, store_discovery, etc.
enabled BOOLEAN DEFAULT TRUE,
interval_hours INTEGER NOT NULL, -- How often to run
priority INTEGER DEFAULT 0, -- Task priority when created
state_code VARCHAR(2), -- Optional filter
last_run_at TIMESTAMPTZ, -- When it last ran
next_run_at TIMESTAMPTZ, -- When it's due next
last_task_count INTEGER, -- Tasks created last run
last_error TEXT -- Error message if failed
);
```
### worker_tasks (migration 074)
The task queue. Workers pull from here.
```sql
CREATE TABLE worker_tasks (
id SERIAL PRIMARY KEY,
role task_role NOT NULL, -- What type of work
dispensary_id INTEGER, -- Which store (if applicable)
platform VARCHAR(50), -- Which platform
status task_status DEFAULT 'pending',
priority INTEGER DEFAULT 0, -- Higher = process first
scheduled_for TIMESTAMP, -- Don't process before this time
worker_id VARCHAR(100), -- Which worker claimed it
claimed_at TIMESTAMP,
started_at TIMESTAMP,
completed_at TIMESTAMP,
last_heartbeat_at TIMESTAMP, -- For stale detection
result JSONB,
error_message TEXT,
retry_count INTEGER DEFAULT 0,
max_retries INTEGER DEFAULT 3
);
```
---
## Default Schedules
| Name | Role | Interval | Priority | Description |
|------|------|----------|----------|-------------|
| `payload_fetch_all` | payload_fetch | 4 hours | 0 | Fetch payloads from Dutchie API (chains to product_refresh) |
| `store_discovery_dutchie` | store_discovery | 24 hours | 5 | Find new Dutchie stores |
| `analytics_refresh` | analytics_refresh | 6 hours | 0 | Refresh MVs |
---
## Task Roles
| Role | Description | Creates Tasks For |
|------|-------------|-------------------|
| `payload_fetch` | **NEW** - Fetch from Dutchie API, save to disk | Each dispensary needing crawl |
| `product_refresh` | **CHANGED** - Read local payload, normalize, upsert to DB | Chained from payload_fetch |
| `store_discovery` | Find new dispensaries, returns newStoreIds[] | Single task per platform |
| `entry_point_discovery` | **DEPRECATED** - Resolve platform IDs | No longer used |
| `product_discovery` | Initial product fetch for new stores | Chained from store_discovery |
| `analytics_refresh` | Refresh MVs | Single global task |
### Payload/Refresh Separation (2024-12-10)
The crawl workflow is now split into two phases:
```
payload_fetch (scheduled every 4h)
└─► Hit Dutchie GraphQL API
└─► Save raw JSON to /storage/payloads/{year}/{month}/{day}/store_{id}_{ts}.json.gz
└─► Record metadata in raw_crawl_payloads table
└─► Queue product_refresh task with payload_id
product_refresh (chained from payload_fetch)
└─► Load payload from filesystem (NOT from API)
└─► Normalize via DutchieNormalizer
└─► Upsert to store_products
└─► Create snapshots
└─► Track missing products
└─► Download images
```
**Benefits:**
- **Retry-friendly**: If normalize fails, re-run product_refresh without re-crawling
- **Replay-able**: Run product_refresh against any historical payload
- **Faster refreshes**: Local file read vs network call
- **Historical diffs**: Compare payloads to see what changed between crawls
- **Less API pressure**: Only payload_fetch hits Dutchie
---
## Task Chaining
Tasks automatically queue follow-up tasks upon successful completion. This creates two main flows:
### Discovery Flow (New Stores)
When `store_discovery` finds new dispensaries, they automatically get their initial product data:
```
store_discovery
└─► Discovers new locations via Dutchie GraphQL
└─► Auto-promotes valid locations to dispensaries table
└─► Collects newDispensaryIds[] from promotions
└─► Returns { newStoreIds: [...] } in result
chainNextTask() detects newStoreIds
└─► Creates product_discovery task for each new store
product_discovery
└─► Calls handlePayloadFetch() internally
└─► payload_fetch hits Dutchie API
└─► Saves raw JSON to /storage/payloads/
└─► Queues product_refresh task with payload_id
product_refresh
└─► Loads payload from filesystem
└─► Normalizes and upserts to store_products
└─► Creates snapshots, downloads images
```
**Complete Discovery Chain:**
```
store_discovery → product_discovery → payload_fetch → product_refresh
(internal call) (queues next)
```
### Scheduled Flow (Existing Stores)
For existing stores, `payload_fetch_all` schedule runs every 4 hours:
```
TaskScheduler (every 60s)
└─► Checks task_schedules for due schedules
└─► payload_fetch_all is due
└─► Generates payload_fetch task for each dispensary
payload_fetch
└─► Hits Dutchie GraphQL API
└─► Saves raw JSON to /storage/payloads/
└─► Queues product_refresh task with payload_id
product_refresh
└─► Loads payload from filesystem (NOT API)
└─► Normalizes via DutchieNormalizer
└─► Upserts to store_products
└─► Creates snapshots
```
**Complete Scheduled Chain:**
```
payload_fetch → product_refresh
(queues) (reads local)
```
### Chaining Implementation
Task chaining is handled in two places:
1. **Internal chaining (handler calls handler):**
- `product_discovery` calls `handlePayloadFetch()` directly
2. **External chaining (chainNextTask() in task-service.ts):**
- Called after task completion
- `store_discovery` → queues `product_discovery` for each newStoreId
3. **Queue-based chaining (taskService.createTask):**
- `payload_fetch` queues `product_refresh` with `payload: { payload_id }`
---
## Payload API Endpoints
Raw crawl payloads can be accessed via the Payloads API:
| Endpoint | Method | Description |
|----------|--------|-------------|
| `GET /api/payloads` | GET | List payload metadata (paginated) |
| `GET /api/payloads/:id` | GET | Get payload metadata by ID |
| `GET /api/payloads/:id/data` | GET | Get full payload JSON (decompressed) |
| `GET /api/payloads/store/:dispensaryId` | GET | List payloads for a store |
| `GET /api/payloads/store/:dispensaryId/latest` | GET | Get latest payload for a store |
| `GET /api/payloads/store/:dispensaryId/diff` | GET | Diff two payloads for changes |
### Payload Diff Response
The diff endpoint returns:
```json
{
"success": true,
"from": { "id": 123, "fetchedAt": "...", "productCount": 100 },
"to": { "id": 456, "fetchedAt": "...", "productCount": 105 },
"diff": {
"added": 10,
"removed": 5,
"priceChanges": 8,
"stockChanges": 12
},
"details": {
"added": [...],
"removed": [...],
"priceChanges": [...],
"stockChanges": [...]
}
}
```
---
## API Endpoints
### Schedules (NEW)
| Endpoint | Method | Description |
|----------|--------|-------------|
| `GET /api/schedules` | GET | List all schedules |
| `PUT /api/schedules/:id` | PUT | Update schedule |
| `POST /api/schedules/:id/trigger` | POST | Run schedule immediately |
### Task Creation (rewired 2024-12-10)
| Endpoint | Method | Description |
|----------|--------|-------------|
| `POST /api/job-queue/enqueue` | POST | Create single task |
| `POST /api/job-queue/enqueue-batch` | POST | Create batch tasks |
| `POST /api/job-queue/enqueue-state` | POST | Create tasks for state |
| `POST /api/tasks` | POST | Direct task creation |
### Task Management
| Endpoint | Method | Description |
|----------|--------|-------------|
| `GET /api/tasks` | GET | List tasks |
| `GET /api/tasks/:id` | GET | Get single task |
| `GET /api/tasks/counts` | GET | Task counts by status |
| `POST /api/tasks/recover-stale` | POST | Recover stale tasks |
---
## Key Files
| File | Purpose |
|------|---------|
| `src/services/task-scheduler.ts` | **NEW** - DB-driven scheduler |
| `src/tasks/task-worker.ts` | Worker that processes tasks |
| `src/tasks/task-service.ts` | Task CRUD operations |
| `src/tasks/handlers/payload-fetch.ts` | **NEW** - Fetches from API, saves to disk |
| `src/tasks/handlers/product-refresh.ts` | **CHANGED** - Reads from disk, processes to DB |
| `src/utils/payload-storage.ts` | **NEW** - Payload save/load utilities |
| `src/routes/tasks.ts` | Task API endpoints |
| `src/routes/job-queue.ts` | Job Queue UI endpoints (rewired) |
| `migrations/079_task_schedules.sql` | Schedule table |
| `migrations/080_raw_crawl_payloads.sql` | Payload metadata table |
| `migrations/081_payload_fetch_columns.sql` | payload, last_fetch_at columns |
| `migrations/074_worker_task_queue.sql` | Task queue table |
---
## Legacy Code (DEPRECATED)
| File | Status | Replacement |
|------|--------|-------------|
| `src/services/scheduler.ts` | DEPRECATED | `task-scheduler.ts` |
| `dispensary_crawl_jobs` table | ORPHANED | `worker_tasks` |
| `job_schedules` table | LEGACY | `task_schedules` |
---
## Dashboard Integration
Both pages remain wired to the dashboard:
| Page | Data Source | Actions |
|------|-------------|---------|
| **Job Queue** | `worker_tasks`, `task_schedules` | Create tasks, view schedules |
| **Task Queue** | `worker_tasks` | View tasks, recover stale |
---
## Multi-Replica Safety
The scheduler uses `SELECT FOR UPDATE SKIP LOCKED` to ensure:
1. **Only one replica** executes a schedule at a time
2. **No duplicate tasks** created
3. **Survives pod restarts** - state in DB, not memory
4. **Self-healing** - recovers stale tasks on startup
```sql
-- This query is atomic across all API server replicas
SELECT * FROM task_schedules
WHERE enabled = true AND next_run_at <= NOW()
FOR UPDATE SKIP LOCKED
```
---
## Worker Scaling (K8s)
Workers run as a StatefulSet in Kubernetes. You can scale from the admin UI or CLI.
### From Admin UI
The Workers page (`/admin/workers`) provides:
- Current replica count display
- Scale up/down buttons
- Target replica input
### API Endpoints
| Endpoint | Method | Description |
|----------|--------|-------------|
| `GET /api/workers/k8s/replicas` | GET | Get current/desired replica counts |
| `POST /api/workers/k8s/scale` | POST | Scale to N replicas (body: `{ replicas: N }`) |
### From CLI
```bash
# View current replicas
kubectl get statefulset scraper-worker -n dispensary-scraper
# Scale to 10 workers
kubectl scale statefulset scraper-worker -n dispensary-scraper --replicas=10
# Scale down to 3 workers
kubectl scale statefulset scraper-worker -n dispensary-scraper --replicas=3
```
### Configuration
Environment variables for the API server:
| Variable | Default | Description |
|----------|---------|-------------|
| `K8S_NAMESPACE` | `dispensary-scraper` | Kubernetes namespace |
| `K8S_WORKER_STATEFULSET` | `scraper-worker` | StatefulSet name |
### RBAC Requirements
The API server pod needs these K8s permissions:
```yaml
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
name: worker-scaler
namespace: dispensary-scraper
rules:
- apiGroups: ["apps"]
resources: ["statefulsets"]
verbs: ["get", "patch"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
name: scraper-worker-scaler
namespace: dispensary-scraper
subjects:
- kind: ServiceAccount
name: default
namespace: dispensary-scraper
roleRef:
kind: Role
name: worker-scaler
apiGroup: rbac.authorization.k8s.io
```

View File

@@ -1,27 +0,0 @@
-- Migration: Worker Commands Table
-- Purpose: Store commands for workers (decommission, etc.)
-- Workers poll this table after each task to check for commands
CREATE TABLE IF NOT EXISTS worker_commands (
id SERIAL PRIMARY KEY,
worker_id TEXT NOT NULL,
command TEXT NOT NULL, -- 'decommission', 'pause', 'resume'
reason TEXT,
issued_by TEXT,
issued_at TIMESTAMPTZ DEFAULT NOW(),
acknowledged_at TIMESTAMPTZ,
executed_at TIMESTAMPTZ,
status TEXT DEFAULT 'pending' -- 'pending', 'acknowledged', 'executed', 'cancelled'
);
-- Index for worker lookups
CREATE INDEX IF NOT EXISTS idx_worker_commands_worker_id ON worker_commands(worker_id);
CREATE INDEX IF NOT EXISTS idx_worker_commands_pending ON worker_commands(worker_id, status) WHERE status = 'pending';
-- Add decommission_requested column to worker_registry for quick checks
ALTER TABLE worker_registry ADD COLUMN IF NOT EXISTS decommission_requested BOOLEAN DEFAULT FALSE;
ALTER TABLE worker_registry ADD COLUMN IF NOT EXISTS decommission_reason TEXT;
ALTER TABLE worker_registry ADD COLUMN IF NOT EXISTS decommission_requested_at TIMESTAMPTZ;
-- Comment
COMMENT ON TABLE worker_commands IS 'Commands issued to workers (decommission after task, pause, etc.)';

View File

@@ -1,8 +0,0 @@
-- Migration 078: Add consecutive_403_count to proxies table
-- Per workflow-12102025.md: Track consecutive 403s per proxy
-- After 3 consecutive 403s with different fingerprints → disable proxy
ALTER TABLE proxies ADD COLUMN IF NOT EXISTS consecutive_403_count INTEGER DEFAULT 0;
-- Add comment explaining the column
COMMENT ON COLUMN proxies.consecutive_403_count IS 'Tracks consecutive 403 blocks. Reset to 0 on success. Proxy disabled at 3.';

View File

@@ -1,49 +0,0 @@
-- Migration 079: Task Schedules for Database-Driven Scheduler
-- Per TASK_WORKFLOW_2024-12-10.md: Replaces node-cron with DB-driven scheduling
--
-- 2024-12-10: Created for reliable, multi-replica-safe task scheduling
-- task_schedules: Stores schedule definitions and state
CREATE TABLE IF NOT EXISTS task_schedules (
id SERIAL PRIMARY KEY,
name VARCHAR(100) NOT NULL UNIQUE,
role VARCHAR(50) NOT NULL, -- TaskRole: product_refresh, store_discovery, etc.
description TEXT,
-- Schedule configuration
enabled BOOLEAN DEFAULT TRUE,
interval_hours INTEGER NOT NULL DEFAULT 4,
priority INTEGER DEFAULT 0,
-- Optional scope filters
state_code VARCHAR(2), -- NULL = all states
platform VARCHAR(50), -- NULL = all platforms
-- Execution state (updated by scheduler)
last_run_at TIMESTAMPTZ,
next_run_at TIMESTAMPTZ,
last_task_count INTEGER DEFAULT 0,
last_error TEXT,
created_at TIMESTAMPTZ DEFAULT NOW(),
updated_at TIMESTAMPTZ DEFAULT NOW()
);
-- Indexes for scheduler queries
CREATE INDEX IF NOT EXISTS idx_task_schedules_enabled ON task_schedules(enabled) WHERE enabled = TRUE;
CREATE INDEX IF NOT EXISTS idx_task_schedules_next_run ON task_schedules(next_run_at) WHERE enabled = TRUE;
-- Insert default schedules
INSERT INTO task_schedules (name, role, interval_hours, priority, description, next_run_at)
VALUES
('product_refresh_all', 'product_refresh', 4, 0, 'Generate product refresh tasks for all crawl-enabled stores every 4 hours', NOW()),
('store_discovery_dutchie', 'store_discovery', 24, 5, 'Discover new Dutchie stores daily', NOW()),
('analytics_refresh', 'analytics_refresh', 6, 0, 'Refresh analytics materialized views every 6 hours', NOW())
ON CONFLICT (name) DO NOTHING;
-- Comment for documentation
COMMENT ON TABLE task_schedules IS 'Database-driven task scheduler configuration. Per TASK_WORKFLOW_2024-12-10.md:
- Schedules persist in DB (survive restarts)
- Uses SELECT FOR UPDATE SKIP LOCKED for multi-replica safety
- Scheduler polls every 60s and executes due schedules
- Creates tasks in worker_tasks for task-worker.ts to process';

View File

@@ -1,58 +0,0 @@
-- Migration 080: Raw Crawl Payloads Metadata Table
-- Per TASK_WORKFLOW_2024-12-10.md: Store full GraphQL payloads for historical analysis
--
-- Design Pattern: Metadata/Payload Separation
-- - Metadata (this table): Small, indexed, queryable
-- - Payload (filesystem): Gzipped JSON at storage_path
--
-- Benefits:
-- - Compare any two crawls to see what changed
-- - Replay/re-normalize historical data if logic changes
-- - Debug issues by seeing exactly what the API returned
-- - DB stays small, backups stay fast
--
-- Storage location: /storage/payloads/{year}/{month}/{day}/store_{id}_{timestamp}.json.gz
-- Compression: ~90% reduction (1.5MB -> 150KB per crawl)
CREATE TABLE IF NOT EXISTS raw_crawl_payloads (
id SERIAL PRIMARY KEY,
-- Links to crawl tracking
crawl_run_id INTEGER REFERENCES crawl_runs(id) ON DELETE SET NULL,
dispensary_id INTEGER NOT NULL REFERENCES dispensaries(id) ON DELETE CASCADE,
-- File location (gzipped JSON)
storage_path TEXT NOT NULL,
-- Metadata for quick queries without loading file
product_count INTEGER NOT NULL DEFAULT 0,
size_bytes INTEGER, -- Compressed size
size_bytes_raw INTEGER, -- Uncompressed size
-- Timestamps
fetched_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
-- Optional: checksum for integrity verification
checksum_sha256 VARCHAR(64)
);
-- Indexes for common queries
CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_dispensary
ON raw_crawl_payloads(dispensary_id);
CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_dispensary_fetched
ON raw_crawl_payloads(dispensary_id, fetched_at DESC);
CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_fetched
ON raw_crawl_payloads(fetched_at DESC);
CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_crawl_run
ON raw_crawl_payloads(crawl_run_id)
WHERE crawl_run_id IS NOT NULL;
-- Comments
COMMENT ON TABLE raw_crawl_payloads IS 'Metadata for raw GraphQL payloads stored on filesystem. Per TASK_WORKFLOW_2024-12-10.md: Full payloads enable historical diffs and replay.';
COMMENT ON COLUMN raw_crawl_payloads.storage_path IS 'Path to gzipped JSON file, e.g. /storage/payloads/2024/12/10/store_123_1702234567.json.gz';
COMMENT ON COLUMN raw_crawl_payloads.size_bytes IS 'Compressed file size in bytes';
COMMENT ON COLUMN raw_crawl_payloads.size_bytes_raw IS 'Uncompressed payload size in bytes';

View File

@@ -1,37 +0,0 @@
-- Migration 081: Payload Fetch Columns
-- Per TASK_WORKFLOW_2024-12-10.md: Separates API fetch from data processing
--
-- New architecture:
-- - payload_fetch: Hits Dutchie API, saves raw payload to disk
-- - product_refresh: Reads local payload, normalizes, upserts to DB
--
-- This migration adds:
-- 1. payload column to worker_tasks (for task chaining data)
-- 2. processed_at column to raw_crawl_payloads (track when payload was processed)
-- 3. last_fetch_at column to dispensaries (track when last payload was fetched)
-- Add payload column to worker_tasks for task chaining
-- Used by payload_fetch to pass payload_id to product_refresh
ALTER TABLE worker_tasks
ADD COLUMN IF NOT EXISTS payload JSONB DEFAULT NULL;
COMMENT ON COLUMN worker_tasks.payload IS 'Per TASK_WORKFLOW_2024-12-10.md: Task chaining data (e.g., payload_id from payload_fetch to product_refresh)';
-- Add processed_at to raw_crawl_payloads
-- Tracks when the payload was processed by product_refresh
ALTER TABLE raw_crawl_payloads
ADD COLUMN IF NOT EXISTS processed_at TIMESTAMPTZ DEFAULT NULL;
COMMENT ON COLUMN raw_crawl_payloads.processed_at IS 'When this payload was processed by product_refresh handler';
-- Index for finding unprocessed payloads
CREATE INDEX IF NOT EXISTS idx_raw_crawl_payloads_unprocessed
ON raw_crawl_payloads(dispensary_id, fetched_at DESC)
WHERE processed_at IS NULL;
-- Add last_fetch_at to dispensaries
-- Tracks when the last payload was fetched (separate from last_crawl_at which is when processing completed)
ALTER TABLE dispensaries
ADD COLUMN IF NOT EXISTS last_fetch_at TIMESTAMPTZ DEFAULT NULL;
COMMENT ON COLUMN dispensaries.last_fetch_at IS 'Per TASK_WORKFLOW_2024-12-10.md: When last payload was fetched from API (separate from last_crawl_at which is when processing completed)';

View File

@@ -1,27 +0,0 @@
-- Migration: 082_proxy_notification_trigger
-- Date: 2024-12-11
-- Description: Add PostgreSQL NOTIFY trigger to alert workers when proxies are added
-- Create function to notify workers when active proxy is added/activated
CREATE OR REPLACE FUNCTION notify_proxy_added()
RETURNS TRIGGER AS $$
BEGIN
-- Only notify if proxy is active
IF NEW.active = true THEN
PERFORM pg_notify('proxy_added', NEW.id::text);
END IF;
RETURN NEW;
END;
$$ LANGUAGE plpgsql;
-- Drop existing trigger if any
DROP TRIGGER IF EXISTS proxy_added_trigger ON proxies;
-- Create trigger on insert and update of active column
CREATE TRIGGER proxy_added_trigger
AFTER INSERT OR UPDATE OF active ON proxies
FOR EACH ROW
EXECUTE FUNCTION notify_proxy_added();
COMMENT ON FUNCTION notify_proxy_added() IS
'Sends PostgreSQL NOTIFY to proxy_added channel when an active proxy is added or activated. Workers LISTEN on this channel to wake up immediately.';

View File

@@ -1,88 +0,0 @@
-- Migration 083: Discovery Run Tracking
-- Tracks progress of store discovery runs step-by-step
-- Main discovery runs table
CREATE TABLE IF NOT EXISTS discovery_runs (
id SERIAL PRIMARY KEY,
platform VARCHAR(50) NOT NULL DEFAULT 'dutchie',
status VARCHAR(20) NOT NULL DEFAULT 'running', -- running, completed, failed
started_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
finished_at TIMESTAMPTZ,
task_id INTEGER REFERENCES worker_task_queue(id),
-- Totals
states_total INTEGER DEFAULT 0,
states_completed INTEGER DEFAULT 0,
locations_discovered INTEGER DEFAULT 0,
locations_promoted INTEGER DEFAULT 0,
new_store_ids INTEGER[] DEFAULT '{}',
-- Error info
error_message TEXT,
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
);
-- Per-state progress within a run
CREATE TABLE IF NOT EXISTS discovery_run_states (
id SERIAL PRIMARY KEY,
run_id INTEGER NOT NULL REFERENCES discovery_runs(id) ON DELETE CASCADE,
state_code VARCHAR(2) NOT NULL,
status VARCHAR(20) NOT NULL DEFAULT 'pending', -- pending, running, completed, failed
started_at TIMESTAMPTZ,
finished_at TIMESTAMPTZ,
-- Results
cities_found INTEGER DEFAULT 0,
locations_found INTEGER DEFAULT 0,
locations_upserted INTEGER DEFAULT 0,
new_dispensary_ids INTEGER[] DEFAULT '{}',
-- Error info
error_message TEXT,
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
UNIQUE(run_id, state_code)
);
-- Step-by-step log for detailed progress tracking
CREATE TABLE IF NOT EXISTS discovery_run_steps (
id SERIAL PRIMARY KEY,
run_id INTEGER NOT NULL REFERENCES discovery_runs(id) ON DELETE CASCADE,
state_code VARCHAR(2),
step_name VARCHAR(100) NOT NULL,
status VARCHAR(20) NOT NULL DEFAULT 'started', -- started, completed, failed
started_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
finished_at TIMESTAMPTZ,
-- Details (JSON for flexibility)
details JSONB DEFAULT '{}',
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
);
-- Indexes for querying
CREATE INDEX IF NOT EXISTS idx_discovery_runs_status ON discovery_runs(status);
CREATE INDEX IF NOT EXISTS idx_discovery_runs_platform ON discovery_runs(platform);
CREATE INDEX IF NOT EXISTS idx_discovery_runs_started_at ON discovery_runs(started_at DESC);
CREATE INDEX IF NOT EXISTS idx_discovery_run_states_run_id ON discovery_run_states(run_id);
CREATE INDEX IF NOT EXISTS idx_discovery_run_steps_run_id ON discovery_run_steps(run_id);
-- View for latest run status per platform
CREATE OR REPLACE VIEW v_latest_discovery_runs AS
SELECT DISTINCT ON (platform)
id,
platform,
status,
started_at,
finished_at,
states_total,
states_completed,
locations_discovered,
locations_promoted,
array_length(new_store_ids, 1) as new_stores_count,
error_message,
EXTRACT(EPOCH FROM (COALESCE(finished_at, NOW()) - started_at)) as duration_seconds
FROM discovery_runs
ORDER BY platform, started_at DESC;

View File

@@ -1,253 +0,0 @@
-- Migration 084: Dual Transport Preflight System
-- Workers run both curl and http (Puppeteer) preflights on startup
-- Tasks can require a specific transport method
-- ===================================================================
-- PART 1: Add preflight columns to worker_registry
-- ===================================================================
-- Preflight status for curl/axios transport (proxy-based)
ALTER TABLE worker_registry
ADD COLUMN IF NOT EXISTS preflight_curl_status VARCHAR(20) DEFAULT 'pending';
-- Preflight status for http/Puppeteer transport (browser-based)
ALTER TABLE worker_registry
ADD COLUMN IF NOT EXISTS preflight_http_status VARCHAR(20) DEFAULT 'pending';
-- Timestamps for when each preflight completed
ALTER TABLE worker_registry
ADD COLUMN IF NOT EXISTS preflight_curl_at TIMESTAMPTZ;
ALTER TABLE worker_registry
ADD COLUMN IF NOT EXISTS preflight_http_at TIMESTAMPTZ;
-- Error messages for failed preflights
ALTER TABLE worker_registry
ADD COLUMN IF NOT EXISTS preflight_curl_error TEXT;
ALTER TABLE worker_registry
ADD COLUMN IF NOT EXISTS preflight_http_error TEXT;
-- Response time for successful preflights (ms)
ALTER TABLE worker_registry
ADD COLUMN IF NOT EXISTS preflight_curl_ms INTEGER;
ALTER TABLE worker_registry
ADD COLUMN IF NOT EXISTS preflight_http_ms INTEGER;
-- Constraints for preflight status values
ALTER TABLE worker_registry
DROP CONSTRAINT IF EXISTS valid_preflight_curl_status;
ALTER TABLE worker_registry
ADD CONSTRAINT valid_preflight_curl_status
CHECK (preflight_curl_status IN ('pending', 'passed', 'failed', 'skipped'));
ALTER TABLE worker_registry
DROP CONSTRAINT IF EXISTS valid_preflight_http_status;
ALTER TABLE worker_registry
ADD CONSTRAINT valid_preflight_http_status
CHECK (preflight_http_status IN ('pending', 'passed', 'failed', 'skipped'));
-- ===================================================================
-- PART 2: Add method column to worker_tasks
-- ===================================================================
-- Transport method requirement for the task
-- NULL = no preference (any worker can claim)
-- 'curl' = requires curl/axios transport (proxy-based, fast)
-- 'http' = requires http/Puppeteer transport (browser-based, anti-detect)
ALTER TABLE worker_tasks
ADD COLUMN IF NOT EXISTS method VARCHAR(10);
-- Constraint for valid method values
ALTER TABLE worker_tasks
DROP CONSTRAINT IF EXISTS valid_task_method;
ALTER TABLE worker_tasks
ADD CONSTRAINT valid_task_method
CHECK (method IS NULL OR method IN ('curl', 'http'));
-- Index for method-based task claiming
CREATE INDEX IF NOT EXISTS idx_worker_tasks_method
ON worker_tasks(method)
WHERE status = 'pending';
-- Set default method for all existing pending tasks to 'http'
-- ALL current tasks require Puppeteer/browser-based transport
UPDATE worker_tasks
SET method = 'http'
WHERE method IS NULL;
-- ===================================================================
-- PART 3: Update claim_task function for method compatibility
-- ===================================================================
CREATE OR REPLACE FUNCTION claim_task(
p_role VARCHAR(50),
p_worker_id VARCHAR(100),
p_curl_passed BOOLEAN DEFAULT TRUE,
p_http_passed BOOLEAN DEFAULT FALSE
) RETURNS worker_tasks AS $$
DECLARE
claimed_task worker_tasks;
BEGIN
UPDATE worker_tasks
SET
status = 'claimed',
worker_id = p_worker_id,
claimed_at = NOW(),
updated_at = NOW()
WHERE id = (
SELECT id FROM worker_tasks
WHERE role = p_role
AND status = 'pending'
AND (scheduled_for IS NULL OR scheduled_for <= NOW())
-- Method compatibility: worker must have passed the required preflight
AND (
method IS NULL -- No preference, any worker can claim
OR (method = 'curl' AND p_curl_passed = TRUE)
OR (method = 'http' AND p_http_passed = TRUE)
)
-- Exclude stores that already have an active task
AND (dispensary_id IS NULL OR dispensary_id NOT IN (
SELECT dispensary_id FROM worker_tasks
WHERE status IN ('claimed', 'running')
AND dispensary_id IS NOT NULL
))
ORDER BY priority DESC, created_at ASC
LIMIT 1
FOR UPDATE SKIP LOCKED
)
RETURNING * INTO claimed_task;
RETURN claimed_task;
END;
$$ LANGUAGE plpgsql;
-- ===================================================================
-- PART 4: Update v_active_workers view
-- ===================================================================
DROP VIEW IF EXISTS v_active_workers;
CREATE VIEW v_active_workers AS
SELECT
wr.id,
wr.worker_id,
wr.friendly_name,
wr.role,
wr.status,
wr.pod_name,
wr.hostname,
wr.started_at,
wr.last_heartbeat_at,
wr.last_task_at,
wr.tasks_completed,
wr.tasks_failed,
wr.current_task_id,
-- Preflight status
wr.preflight_curl_status,
wr.preflight_http_status,
wr.preflight_curl_at,
wr.preflight_http_at,
wr.preflight_curl_error,
wr.preflight_http_error,
wr.preflight_curl_ms,
wr.preflight_http_ms,
-- Computed fields
EXTRACT(EPOCH FROM (NOW() - wr.last_heartbeat_at)) as seconds_since_heartbeat,
CASE
WHEN wr.status = 'offline' THEN 'offline'
WHEN wr.last_heartbeat_at < NOW() - INTERVAL '2 minutes' THEN 'stale'
WHEN wr.current_task_id IS NOT NULL THEN 'busy'
ELSE 'ready'
END as health_status,
-- Capability flags (can this worker handle curl/http tasks?)
(wr.preflight_curl_status = 'passed') as can_curl,
(wr.preflight_http_status = 'passed') as can_http
FROM worker_registry wr
WHERE wr.status != 'terminated'
ORDER BY wr.status = 'active' DESC, wr.last_heartbeat_at DESC;
-- ===================================================================
-- PART 5: View for task queue with method info
-- ===================================================================
DROP VIEW IF EXISTS v_task_history;
CREATE VIEW v_task_history AS
SELECT
t.id,
t.role,
t.dispensary_id,
d.name as dispensary_name,
t.platform,
t.status,
t.priority,
t.method,
t.worker_id,
t.scheduled_for,
t.claimed_at,
t.started_at,
t.completed_at,
t.error_message,
t.retry_count,
t.created_at,
EXTRACT(EPOCH FROM (t.completed_at - t.started_at)) as duration_sec
FROM worker_tasks t
LEFT JOIN dispensaries d ON d.id = t.dispensary_id
ORDER BY t.created_at DESC;
-- ===================================================================
-- PART 6: Helper function to update worker preflight status
-- ===================================================================
CREATE OR REPLACE FUNCTION update_worker_preflight(
p_worker_id VARCHAR(100),
p_transport VARCHAR(10), -- 'curl' or 'http'
p_status VARCHAR(20), -- 'passed', 'failed', 'skipped'
p_response_ms INTEGER DEFAULT NULL,
p_error TEXT DEFAULT NULL
) RETURNS VOID AS $$
BEGIN
IF p_transport = 'curl' THEN
UPDATE worker_registry
SET
preflight_curl_status = p_status,
preflight_curl_at = NOW(),
preflight_curl_ms = p_response_ms,
preflight_curl_error = p_error,
updated_at = NOW()
WHERE worker_id = p_worker_id;
ELSIF p_transport = 'http' THEN
UPDATE worker_registry
SET
preflight_http_status = p_status,
preflight_http_at = NOW(),
preflight_http_ms = p_response_ms,
preflight_http_error = p_error,
updated_at = NOW()
WHERE worker_id = p_worker_id;
END IF;
END;
$$ LANGUAGE plpgsql;
-- ===================================================================
-- Comments
-- ===================================================================
COMMENT ON COLUMN worker_registry.preflight_curl_status IS 'Status of curl/axios preflight: pending, passed, failed, skipped';
COMMENT ON COLUMN worker_registry.preflight_http_status IS 'Status of http/Puppeteer preflight: pending, passed, failed, skipped';
COMMENT ON COLUMN worker_registry.preflight_curl_at IS 'When curl preflight completed';
COMMENT ON COLUMN worker_registry.preflight_http_at IS 'When http preflight completed';
COMMENT ON COLUMN worker_registry.preflight_curl_error IS 'Error message if curl preflight failed';
COMMENT ON COLUMN worker_registry.preflight_http_error IS 'Error message if http preflight failed';
COMMENT ON COLUMN worker_registry.preflight_curl_ms IS 'Response time of successful curl preflight (ms)';
COMMENT ON COLUMN worker_registry.preflight_http_ms IS 'Response time of successful http preflight (ms)';
COMMENT ON COLUMN worker_tasks.method IS 'Transport method required: NULL=any, curl=proxy-based, http=browser-based';
COMMENT ON FUNCTION claim_task IS 'Atomically claim a task, respecting method requirements and per-store locking';
COMMENT ON FUNCTION update_worker_preflight IS 'Update a workers preflight status for a given transport';

View File

@@ -1,6 +1,6 @@
{
"name": "dutchie-menus-backend",
"version": "1.6.0",
"version": "1.5.1",
"lockfileVersion": 3,
"requires": true,
"packages": {
@@ -46,97 +46,6 @@
"resolved": "https://registry.npmjs.org/@ioredis/commands/-/commands-1.4.0.tgz",
"integrity": "sha512-aFT2yemJJo+TZCmieA7qnYGQooOS7QfNmYrzGtsYd3g9j5iDP8AimYYAesf79ohjbLG12XxC4nG5DyEnC88AsQ=="
},
"node_modules/@jsep-plugin/assignment": {
"version": "1.3.0",
"resolved": "https://registry.npmjs.org/@jsep-plugin/assignment/-/assignment-1.3.0.tgz",
"integrity": "sha512-VVgV+CXrhbMI3aSusQyclHkenWSAm95WaiKrMxRFam3JSUiIaQjoMIw2sEs/OX4XifnqeQUN4DYbJjlA8EfktQ==",
"engines": {
"node": ">= 10.16.0"
},
"peerDependencies": {
"jsep": "^0.4.0||^1.0.0"
}
},
"node_modules/@jsep-plugin/regex": {
"version": "1.0.4",
"resolved": "https://registry.npmjs.org/@jsep-plugin/regex/-/regex-1.0.4.tgz",
"integrity": "sha512-q7qL4Mgjs1vByCaTnDFcBnV9HS7GVPJX5vyVoCgZHNSC9rjwIlmbXG5sUuorR5ndfHAIlJ8pVStxvjXHbNvtUg==",
"engines": {
"node": ">= 10.16.0"
},
"peerDependencies": {
"jsep": "^0.4.0||^1.0.0"
}
},
"node_modules/@kubernetes/client-node": {
"version": "1.4.0",
"resolved": "https://registry.npmjs.org/@kubernetes/client-node/-/client-node-1.4.0.tgz",
"integrity": "sha512-Zge3YvF7DJi264dU1b3wb/GmzR99JhUpqTvp+VGHfwZT+g7EOOYNScDJNZwXy9cszyIGPIs0VHr+kk8e95qqrA==",
"dependencies": {
"@types/js-yaml": "^4.0.1",
"@types/node": "^24.0.0",
"@types/node-fetch": "^2.6.13",
"@types/stream-buffers": "^3.0.3",
"form-data": "^4.0.0",
"hpagent": "^1.2.0",
"isomorphic-ws": "^5.0.0",
"js-yaml": "^4.1.0",
"jsonpath-plus": "^10.3.0",
"node-fetch": "^2.7.0",
"openid-client": "^6.1.3",
"rfc4648": "^1.3.0",
"socks-proxy-agent": "^8.0.4",
"stream-buffers": "^3.0.2",
"tar-fs": "^3.0.9",
"ws": "^8.18.2"
}
},
"node_modules/@kubernetes/client-node/node_modules/@types/node": {
"version": "24.10.3",
"resolved": "https://registry.npmjs.org/@types/node/-/node-24.10.3.tgz",
"integrity": "sha512-gqkrWUsS8hcm0r44yn7/xZeV1ERva/nLgrLxFRUGb7aoNMIJfZJ3AC261zDQuOAKC7MiXai1WCpYc48jAHoShQ==",
"dependencies": {
"undici-types": "~7.16.0"
}
},
"node_modules/@kubernetes/client-node/node_modules/tar-fs": {
"version": "3.1.1",
"resolved": "https://registry.npmjs.org/tar-fs/-/tar-fs-3.1.1.tgz",
"integrity": "sha512-LZA0oaPOc2fVo82Txf3gw+AkEd38szODlptMYejQUhndHMLQ9M059uXR+AfS7DNo0NpINvSqDsvyaCrBVkptWg==",
"dependencies": {
"pump": "^3.0.0",
"tar-stream": "^3.1.5"
},
"optionalDependencies": {
"bare-fs": "^4.0.1",
"bare-path": "^3.0.0"
}
},
"node_modules/@kubernetes/client-node/node_modules/undici-types": {
"version": "7.16.0",
"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-7.16.0.tgz",
"integrity": "sha512-Zz+aZWSj8LE6zoxD+xrjh4VfkIG8Ya6LvYkZqtUQGJPZjYl53ypCaUwWqo7eI0x66KBGeRo+mlBEkMSeSZ38Nw=="
},
"node_modules/@kubernetes/client-node/node_modules/ws": {
"version": "8.18.3",
"resolved": "https://registry.npmjs.org/ws/-/ws-8.18.3.tgz",
"integrity": "sha512-PEIGCY5tSlUt50cqyMXfCzX+oOPqN0vuGqWzbcJ2xvnkzkq46oOpz7dQaTDBdfICb4N14+GARUDw2XV2N4tvzg==",
"engines": {
"node": ">=10.0.0"
},
"peerDependencies": {
"bufferutil": "^4.0.1",
"utf-8-validate": ">=5.0.2"
},
"peerDependenciesMeta": {
"bufferutil": {
"optional": true
},
"utf-8-validate": {
"optional": true
}
}
},
"node_modules/@mapbox/node-pre-gyp": {
"version": "1.0.11",
"resolved": "https://registry.npmjs.org/@mapbox/node-pre-gyp/-/node-pre-gyp-1.0.11.tgz",
@@ -342,11 +251,6 @@
"integrity": "sha512-r8Tayk8HJnX0FztbZN7oVqGccWgw98T/0neJphO91KkmOzug1KkofZURD4UaD5uH8AqcFLfdPErnBod0u71/qg==",
"dev": true
},
"node_modules/@types/js-yaml": {
"version": "4.0.9",
"resolved": "https://registry.npmjs.org/@types/js-yaml/-/js-yaml-4.0.9.tgz",
"integrity": "sha512-k4MGaQl5TGo/iipqb2UDG2UwjXziSWkh0uysQelTlJpX1qGlpUZYm8PnO4DxG1qBomtJUdYJ6qR6xdIah10JLg=="
},
"node_modules/@types/jsonwebtoken": {
"version": "9.0.10",
"resolved": "https://registry.npmjs.org/@types/jsonwebtoken/-/jsonwebtoken-9.0.10.tgz",
@@ -372,6 +276,7 @@
"version": "20.19.25",
"resolved": "https://registry.npmjs.org/@types/node/-/node-20.19.25.tgz",
"integrity": "sha512-ZsJzA5thDQMSQO788d7IocwwQbI8B5OPzmqNvpf3NY/+MHDAS759Wo0gd2WQeXYt5AAAQjzcrTVC6SKCuYgoCQ==",
"devOptional": true,
"dependencies": {
"undici-types": "~6.21.0"
}
@@ -382,15 +287,6 @@
"integrity": "sha512-0ikrnug3/IyneSHqCBeslAhlK2aBfYek1fGo4bP4QnZPmiqSGRK+Oy7ZMisLWkesffJvQ1cqAcBnJC+8+nxIAg==",
"dev": true
},
"node_modules/@types/node-fetch": {
"version": "2.6.13",
"resolved": "https://registry.npmjs.org/@types/node-fetch/-/node-fetch-2.6.13.tgz",
"integrity": "sha512-QGpRVpzSaUs30JBSGPjOg4Uveu384erbHBoT1zeONvyCfwQxIkUshLAOqN/k9EjGviPRmWTTe6aH2qySWKTVSw==",
"dependencies": {
"@types/node": "*",
"form-data": "^4.0.4"
}
},
"node_modules/@types/pg": {
"version": "8.15.6",
"resolved": "https://registry.npmjs.org/@types/pg/-/pg-8.15.6.tgz",
@@ -444,14 +340,6 @@
"@types/node": "*"
}
},
"node_modules/@types/stream-buffers": {
"version": "3.0.8",
"resolved": "https://registry.npmjs.org/@types/stream-buffers/-/stream-buffers-3.0.8.tgz",
"integrity": "sha512-J+7VaHKNvlNPJPEJXX/fKa9DZtR/xPMwuIbe+yNOwp1YB+ApUOBv2aUpEoBJEi8nJgbgs1x8e73ttg0r1rSUdw==",
"dependencies": {
"@types/node": "*"
}
},
"node_modules/@types/uuid": {
"version": "9.0.8",
"resolved": "https://registry.npmjs.org/@types/uuid/-/uuid-9.0.8.tgz",
@@ -632,78 +520,6 @@
}
}
},
"node_modules/bare-fs": {
"version": "4.5.2",
"resolved": "https://registry.npmjs.org/bare-fs/-/bare-fs-4.5.2.tgz",
"integrity": "sha512-veTnRzkb6aPHOvSKIOy60KzURfBdUflr5VReI+NSaPL6xf+XLdONQgZgpYvUuZLVQ8dCqxpBAudaOM1+KpAUxw==",
"optional": true,
"dependencies": {
"bare-events": "^2.5.4",
"bare-path": "^3.0.0",
"bare-stream": "^2.6.4",
"bare-url": "^2.2.2",
"fast-fifo": "^1.3.2"
},
"engines": {
"bare": ">=1.16.0"
},
"peerDependencies": {
"bare-buffer": "*"
},
"peerDependenciesMeta": {
"bare-buffer": {
"optional": true
}
}
},
"node_modules/bare-os": {
"version": "3.6.2",
"resolved": "https://registry.npmjs.org/bare-os/-/bare-os-3.6.2.tgz",
"integrity": "sha512-T+V1+1srU2qYNBmJCXZkUY5vQ0B4FSlL3QDROnKQYOqeiQR8UbjNHlPa+TIbM4cuidiN9GaTaOZgSEgsvPbh5A==",
"optional": true,
"engines": {
"bare": ">=1.14.0"
}
},
"node_modules/bare-path": {
"version": "3.0.0",
"resolved": "https://registry.npmjs.org/bare-path/-/bare-path-3.0.0.tgz",
"integrity": "sha512-tyfW2cQcB5NN8Saijrhqn0Zh7AnFNsnczRcuWODH0eYAXBsJ5gVxAUuNr7tsHSC6IZ77cA0SitzT+s47kot8Mw==",
"optional": true,
"dependencies": {
"bare-os": "^3.0.1"
}
},
"node_modules/bare-stream": {
"version": "2.7.0",
"resolved": "https://registry.npmjs.org/bare-stream/-/bare-stream-2.7.0.tgz",
"integrity": "sha512-oyXQNicV1y8nc2aKffH+BUHFRXmx6VrPzlnaEvMhram0nPBrKcEdcyBg5r08D0i8VxngHFAiVyn1QKXpSG0B8A==",
"optional": true,
"dependencies": {
"streamx": "^2.21.0"
},
"peerDependencies": {
"bare-buffer": "*",
"bare-events": "*"
},
"peerDependenciesMeta": {
"bare-buffer": {
"optional": true
},
"bare-events": {
"optional": true
}
}
},
"node_modules/bare-url": {
"version": "2.3.2",
"resolved": "https://registry.npmjs.org/bare-url/-/bare-url-2.3.2.tgz",
"integrity": "sha512-ZMq4gd9ngV5aTMa5p9+UfY0b3skwhHELaDkhEHetMdX0LRkW9kzaym4oo/Eh+Ghm0CCDuMTsRIGM/ytUc1ZYmw==",
"optional": true,
"dependencies": {
"bare-path": "^3.0.0"
}
},
"node_modules/base64-js": {
"version": "1.5.1",
"resolved": "https://registry.npmjs.org/base64-js/-/base64-js-1.5.1.tgz",
@@ -2203,14 +2019,6 @@
"node": ">=16.0.0"
}
},
"node_modules/hpagent": {
"version": "1.2.0",
"resolved": "https://registry.npmjs.org/hpagent/-/hpagent-1.2.0.tgz",
"integrity": "sha512-A91dYTeIB6NoXG+PxTQpCCDDnfHsW9kc06Lvpu1TEe9gnd6ZFeiBoRO9JvzEv6xK7EX97/dUE8g/vBMTqTS3CA==",
"engines": {
"node": ">=14"
}
},
"node_modules/htmlparser2": {
"version": "10.0.0",
"resolved": "https://registry.npmjs.org/htmlparser2/-/htmlparser2-10.0.0.tgz",
@@ -2574,22 +2382,6 @@
"node": ">=0.10.0"
}
},
"node_modules/isomorphic-ws": {
"version": "5.0.0",
"resolved": "https://registry.npmjs.org/isomorphic-ws/-/isomorphic-ws-5.0.0.tgz",
"integrity": "sha512-muId7Zzn9ywDsyXgTIafTry2sV3nySZeUDe6YedVd1Hvuuep5AsIlqK+XefWpYTyJG5e503F2xIuT2lcU6rCSw==",
"peerDependencies": {
"ws": "*"
}
},
"node_modules/jose": {
"version": "6.1.3",
"resolved": "https://registry.npmjs.org/jose/-/jose-6.1.3.tgz",
"integrity": "sha512-0TpaTfihd4QMNwrz/ob2Bp7X04yuxJkjRGi4aKmOqwhov54i6u79oCv7T+C7lo70MKH6BesI3vscD1yb/yzKXQ==",
"funding": {
"url": "https://github.com/sponsors/panva"
}
},
"node_modules/js-tokens": {
"version": "4.0.0",
"resolved": "https://registry.npmjs.org/js-tokens/-/js-tokens-4.0.0.tgz",
@@ -2606,14 +2398,6 @@
"js-yaml": "bin/js-yaml.js"
}
},
"node_modules/jsep": {
"version": "1.4.0",
"resolved": "https://registry.npmjs.org/jsep/-/jsep-1.4.0.tgz",
"integrity": "sha512-B7qPcEVE3NVkmSJbaYxvv4cHkVW7DQsZz13pUMrfS8z8Q/BuShN+gcTXrUlPiGqM2/t/EEaI030bpxMqY8gMlw==",
"engines": {
"node": ">= 10.16.0"
}
},
"node_modules/json-parse-even-better-errors": {
"version": "2.3.1",
"resolved": "https://registry.npmjs.org/json-parse-even-better-errors/-/json-parse-even-better-errors-2.3.1.tgz",
@@ -2635,23 +2419,6 @@
"graceful-fs": "^4.1.6"
}
},
"node_modules/jsonpath-plus": {
"version": "10.3.0",
"resolved": "https://registry.npmjs.org/jsonpath-plus/-/jsonpath-plus-10.3.0.tgz",
"integrity": "sha512-8TNmfeTCk2Le33A3vRRwtuworG/L5RrgMvdjhKZxvyShO+mBu2fP50OWUjRLNtvw344DdDarFh9buFAZs5ujeA==",
"dependencies": {
"@jsep-plugin/assignment": "^1.3.0",
"@jsep-plugin/regex": "^1.0.4",
"jsep": "^1.4.0"
},
"bin": {
"jsonpath": "bin/jsonpath-cli.js",
"jsonpath-plus": "bin/jsonpath-cli.js"
},
"engines": {
"node": ">=18.0.0"
}
},
"node_modules/jsonwebtoken": {
"version": "9.0.2",
"resolved": "https://registry.npmjs.org/jsonwebtoken/-/jsonwebtoken-9.0.2.tgz",
@@ -2726,11 +2493,6 @@
"resolved": "https://registry.npmjs.org/lodash/-/lodash-4.17.21.tgz",
"integrity": "sha512-v2kDEe57lecTulaDIuNTPy3Ry4gLGJ6Z1O3vE1krgXZNrsQ+LFTGHVxVjcXPs17LhbZVGedAJv8XZ1tvj5FvSg=="
},
"node_modules/lodash.clonedeep": {
"version": "4.5.0",
"resolved": "https://registry.npmjs.org/lodash.clonedeep/-/lodash.clonedeep-4.5.0.tgz",
"integrity": "sha512-H5ZhCF25riFd9uB5UCkVKo61m3S/xZk1x4wA6yp/L3RFP6Z/eHH1ymQcGLo7J3GMPfm0V/7m1tryHuGVxpqEBQ=="
},
"node_modules/lodash.defaults": {
"version": "4.2.0",
"resolved": "https://registry.npmjs.org/lodash.defaults/-/lodash.defaults-4.2.0.tgz",
@@ -3180,14 +2942,6 @@
"url": "https://github.com/fb55/nth-check?sponsor=1"
}
},
"node_modules/oauth4webapi": {
"version": "3.8.3",
"resolved": "https://registry.npmjs.org/oauth4webapi/-/oauth4webapi-3.8.3.tgz",
"integrity": "sha512-pQ5BsX3QRTgnt5HxgHwgunIRaDXBdkT23tf8dfzmtTIL2LTpdmxgbpbBm0VgFWAIDlezQvQCTgnVIUmHupXHxw==",
"funding": {
"url": "https://github.com/sponsors/panva"
}
},
"node_modules/object-assign": {
"version": "4.1.1",
"resolved": "https://registry.npmjs.org/object-assign/-/object-assign-4.1.1.tgz",
@@ -3226,18 +2980,6 @@
"wrappy": "1"
}
},
"node_modules/openid-client": {
"version": "6.8.1",
"resolved": "https://registry.npmjs.org/openid-client/-/openid-client-6.8.1.tgz",
"integrity": "sha512-VoYT6enBo6Vj2j3Q5Ec0AezS+9YGzQo1f5Xc42lreMGlfP4ljiXPKVDvCADh+XHCV/bqPu/wWSiCVXbJKvrODw==",
"dependencies": {
"jose": "^6.1.0",
"oauth4webapi": "^3.8.2"
},
"funding": {
"url": "https://github.com/sponsors/panva"
}
},
"node_modules/pac-proxy-agent": {
"version": "7.2.0",
"resolved": "https://registry.npmjs.org/pac-proxy-agent/-/pac-proxy-agent-7.2.0.tgz",
@@ -4141,11 +3883,6 @@
"url": "https://github.com/privatenumber/resolve-pkg-maps?sponsor=1"
}
},
"node_modules/rfc4648": {
"version": "1.5.4",
"resolved": "https://registry.npmjs.org/rfc4648/-/rfc4648-1.5.4.tgz",
"integrity": "sha512-rRg/6Lb+IGfJqO05HZkN50UtY7K/JhxJag1kP23+zyMfrvoB0B7RWv06MbOzoc79RgCdNTiUaNsTT1AJZ7Z+cg=="
},
"node_modules/rimraf": {
"version": "3.0.2",
"resolved": "https://registry.npmjs.org/rimraf/-/rimraf-3.0.2.tgz",
@@ -4576,14 +4313,6 @@
"node": ">= 0.8"
}
},
"node_modules/stream-buffers": {
"version": "3.0.3",
"resolved": "https://registry.npmjs.org/stream-buffers/-/stream-buffers-3.0.3.tgz",
"integrity": "sha512-pqMqwQCso0PBJt2PQmDO0cFj0lyqmiwOMiMSkVtRokl7e+ZTRYgDHKnuZNbqjiJXgsg4nuqtD/zxuo9KqTp0Yw==",
"engines": {
"node": ">= 0.10.0"
}
},
"node_modules/streamx": {
"version": "2.23.0",
"resolved": "https://registry.npmjs.org/streamx/-/streamx-2.23.0.tgz",
@@ -4803,7 +4532,8 @@
"node_modules/undici-types": {
"version": "6.21.0",
"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-6.21.0.tgz",
"integrity": "sha512-iwDZqg0QAGrg9Rav5H4n0M64c3mkR59cJ6wQp+7C4nI0gsmExaedaYLNO44eT4AtBBwjbTiGPMlt2Md0T9H9JQ=="
"integrity": "sha512-iwDZqg0QAGrg9Rav5H4n0M64c3mkR59cJ6wQp+7C4nI0gsmExaedaYLNO44eT4AtBBwjbTiGPMlt2Md0T9H9JQ==",
"devOptional": true
},
"node_modules/universalify": {
"version": "2.0.1",
@@ -4826,14 +4556,6 @@
"resolved": "https://registry.npmjs.org/urlpattern-polyfill/-/urlpattern-polyfill-10.0.0.tgz",
"integrity": "sha512-H/A06tKD7sS1O1X2SshBVeA5FLycRpjqiBeqGKmBwBDBy28EnRjORxTNe269KSSr5un5qyWi1iL61wLxpd+ZOg=="
},
"node_modules/user-agents": {
"version": "1.1.669",
"resolved": "https://registry.npmjs.org/user-agents/-/user-agents-1.1.669.tgz",
"integrity": "sha512-pbIzG+AOqCaIpySKJ4IAm1l0VyE4jMnK4y1thV8lm8PYxI+7X5uWcppOK7zY79TCKKTAnJH3/4gaVIZHsjrmJA==",
"dependencies": {
"lodash.clonedeep": "^4.5.0"
}
},
"node_modules/util": {
"version": "0.12.5",
"resolved": "https://registry.npmjs.org/util/-/util-0.12.5.tgz",

View File

@@ -1,14 +1,13 @@
{
"name": "dutchie-menus-backend",
"version": "1.6.0",
"version": "1.5.1",
"lockfileVersion": 3,
"requires": true,
"packages": {
"": {
"name": "dutchie-menus-backend",
"version": "1.6.0",
"version": "1.5.1",
"dependencies": {
"@kubernetes/client-node": "^1.4.0",
"@types/bcryptjs": "^3.0.0",
"axios": "^1.6.2",
"bcrypt": "^5.1.1",
@@ -35,7 +34,6 @@
"puppeteer-extra-plugin-stealth": "^2.11.2",
"sharp": "^0.32.0",
"socks-proxy-agent": "^8.0.2",
"user-agents": "^1.1.669",
"uuid": "^9.0.1",
"zod": "^3.22.4"
},
@@ -494,97 +492,6 @@
"resolved": "https://registry.npmjs.org/@ioredis/commands/-/commands-1.4.0.tgz",
"integrity": "sha512-aFT2yemJJo+TZCmieA7qnYGQooOS7QfNmYrzGtsYd3g9j5iDP8AimYYAesf79ohjbLG12XxC4nG5DyEnC88AsQ=="
},
"node_modules/@jsep-plugin/assignment": {
"version": "1.3.0",
"resolved": "https://registry.npmjs.org/@jsep-plugin/assignment/-/assignment-1.3.0.tgz",
"integrity": "sha512-VVgV+CXrhbMI3aSusQyclHkenWSAm95WaiKrMxRFam3JSUiIaQjoMIw2sEs/OX4XifnqeQUN4DYbJjlA8EfktQ==",
"engines": {
"node": ">= 10.16.0"
},
"peerDependencies": {
"jsep": "^0.4.0||^1.0.0"
}
},
"node_modules/@jsep-plugin/regex": {
"version": "1.0.4",
"resolved": "https://registry.npmjs.org/@jsep-plugin/regex/-/regex-1.0.4.tgz",
"integrity": "sha512-q7qL4Mgjs1vByCaTnDFcBnV9HS7GVPJX5vyVoCgZHNSC9rjwIlmbXG5sUuorR5ndfHAIlJ8pVStxvjXHbNvtUg==",
"engines": {
"node": ">= 10.16.0"
},
"peerDependencies": {
"jsep": "^0.4.0||^1.0.0"
}
},
"node_modules/@kubernetes/client-node": {
"version": "1.4.0",
"resolved": "https://registry.npmjs.org/@kubernetes/client-node/-/client-node-1.4.0.tgz",
"integrity": "sha512-Zge3YvF7DJi264dU1b3wb/GmzR99JhUpqTvp+VGHfwZT+g7EOOYNScDJNZwXy9cszyIGPIs0VHr+kk8e95qqrA==",
"dependencies": {
"@types/js-yaml": "^4.0.1",
"@types/node": "^24.0.0",
"@types/node-fetch": "^2.6.13",
"@types/stream-buffers": "^3.0.3",
"form-data": "^4.0.0",
"hpagent": "^1.2.0",
"isomorphic-ws": "^5.0.0",
"js-yaml": "^4.1.0",
"jsonpath-plus": "^10.3.0",
"node-fetch": "^2.7.0",
"openid-client": "^6.1.3",
"rfc4648": "^1.3.0",
"socks-proxy-agent": "^8.0.4",
"stream-buffers": "^3.0.2",
"tar-fs": "^3.0.9",
"ws": "^8.18.2"
}
},
"node_modules/@kubernetes/client-node/node_modules/@types/node": {
"version": "24.10.3",
"resolved": "https://registry.npmjs.org/@types/node/-/node-24.10.3.tgz",
"integrity": "sha512-gqkrWUsS8hcm0r44yn7/xZeV1ERva/nLgrLxFRUGb7aoNMIJfZJ3AC261zDQuOAKC7MiXai1WCpYc48jAHoShQ==",
"dependencies": {
"undici-types": "~7.16.0"
}
},
"node_modules/@kubernetes/client-node/node_modules/tar-fs": {
"version": "3.1.1",
"resolved": "https://registry.npmjs.org/tar-fs/-/tar-fs-3.1.1.tgz",
"integrity": "sha512-LZA0oaPOc2fVo82Txf3gw+AkEd38szODlptMYejQUhndHMLQ9M059uXR+AfS7DNo0NpINvSqDsvyaCrBVkptWg==",
"dependencies": {
"pump": "^3.0.0",
"tar-stream": "^3.1.5"
},
"optionalDependencies": {
"bare-fs": "^4.0.1",
"bare-path": "^3.0.0"
}
},
"node_modules/@kubernetes/client-node/node_modules/undici-types": {
"version": "7.16.0",
"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-7.16.0.tgz",
"integrity": "sha512-Zz+aZWSj8LE6zoxD+xrjh4VfkIG8Ya6LvYkZqtUQGJPZjYl53ypCaUwWqo7eI0x66KBGeRo+mlBEkMSeSZ38Nw=="
},
"node_modules/@kubernetes/client-node/node_modules/ws": {
"version": "8.18.3",
"resolved": "https://registry.npmjs.org/ws/-/ws-8.18.3.tgz",
"integrity": "sha512-PEIGCY5tSlUt50cqyMXfCzX+oOPqN0vuGqWzbcJ2xvnkzkq46oOpz7dQaTDBdfICb4N14+GARUDw2XV2N4tvzg==",
"engines": {
"node": ">=10.0.0"
},
"peerDependencies": {
"bufferutil": "^4.0.1",
"utf-8-validate": ">=5.0.2"
},
"peerDependenciesMeta": {
"bufferutil": {
"optional": true
},
"utf-8-validate": {
"optional": true
}
}
},
"node_modules/@mapbox/node-pre-gyp": {
"version": "1.0.11",
"resolved": "https://registry.npmjs.org/@mapbox/node-pre-gyp/-/node-pre-gyp-1.0.11.tgz",
@@ -850,11 +757,6 @@
"integrity": "sha512-r8Tayk8HJnX0FztbZN7oVqGccWgw98T/0neJphO91KkmOzug1KkofZURD4UaD5uH8AqcFLfdPErnBod0u71/qg==",
"dev": true
},
"node_modules/@types/js-yaml": {
"version": "4.0.9",
"resolved": "https://registry.npmjs.org/@types/js-yaml/-/js-yaml-4.0.9.tgz",
"integrity": "sha512-k4MGaQl5TGo/iipqb2UDG2UwjXziSWkh0uysQelTlJpX1qGlpUZYm8PnO4DxG1qBomtJUdYJ6qR6xdIah10JLg=="
},
"node_modules/@types/jsonwebtoken": {
"version": "9.0.10",
"resolved": "https://registry.npmjs.org/@types/jsonwebtoken/-/jsonwebtoken-9.0.10.tgz",
@@ -880,6 +782,7 @@
"version": "20.19.25",
"resolved": "https://registry.npmjs.org/@types/node/-/node-20.19.25.tgz",
"integrity": "sha512-ZsJzA5thDQMSQO788d7IocwwQbI8B5OPzmqNvpf3NY/+MHDAS759Wo0gd2WQeXYt5AAAQjzcrTVC6SKCuYgoCQ==",
"devOptional": true,
"dependencies": {
"undici-types": "~6.21.0"
}
@@ -890,15 +793,6 @@
"integrity": "sha512-0ikrnug3/IyneSHqCBeslAhlK2aBfYek1fGo4bP4QnZPmiqSGRK+Oy7ZMisLWkesffJvQ1cqAcBnJC+8+nxIAg==",
"dev": true
},
"node_modules/@types/node-fetch": {
"version": "2.6.13",
"resolved": "https://registry.npmjs.org/@types/node-fetch/-/node-fetch-2.6.13.tgz",
"integrity": "sha512-QGpRVpzSaUs30JBSGPjOg4Uveu384erbHBoT1zeONvyCfwQxIkUshLAOqN/k9EjGviPRmWTTe6aH2qySWKTVSw==",
"dependencies": {
"@types/node": "*",
"form-data": "^4.0.4"
}
},
"node_modules/@types/pg": {
"version": "8.15.6",
"resolved": "https://registry.npmjs.org/@types/pg/-/pg-8.15.6.tgz",
@@ -952,14 +846,6 @@
"@types/node": "*"
}
},
"node_modules/@types/stream-buffers": {
"version": "3.0.8",
"resolved": "https://registry.npmjs.org/@types/stream-buffers/-/stream-buffers-3.0.8.tgz",
"integrity": "sha512-J+7VaHKNvlNPJPEJXX/fKa9DZtR/xPMwuIbe+yNOwp1YB+ApUOBv2aUpEoBJEi8nJgbgs1x8e73ttg0r1rSUdw==",
"dependencies": {
"@types/node": "*"
}
},
"node_modules/@types/uuid": {
"version": "9.0.8",
"resolved": "https://registry.npmjs.org/@types/uuid/-/uuid-9.0.8.tgz",
@@ -1140,78 +1026,6 @@
}
}
},
"node_modules/bare-fs": {
"version": "4.5.2",
"resolved": "https://registry.npmjs.org/bare-fs/-/bare-fs-4.5.2.tgz",
"integrity": "sha512-veTnRzkb6aPHOvSKIOy60KzURfBdUflr5VReI+NSaPL6xf+XLdONQgZgpYvUuZLVQ8dCqxpBAudaOM1+KpAUxw==",
"optional": true,
"dependencies": {
"bare-events": "^2.5.4",
"bare-path": "^3.0.0",
"bare-stream": "^2.6.4",
"bare-url": "^2.2.2",
"fast-fifo": "^1.3.2"
},
"engines": {
"bare": ">=1.16.0"
},
"peerDependencies": {
"bare-buffer": "*"
},
"peerDependenciesMeta": {
"bare-buffer": {
"optional": true
}
}
},
"node_modules/bare-os": {
"version": "3.6.2",
"resolved": "https://registry.npmjs.org/bare-os/-/bare-os-3.6.2.tgz",
"integrity": "sha512-T+V1+1srU2qYNBmJCXZkUY5vQ0B4FSlL3QDROnKQYOqeiQR8UbjNHlPa+TIbM4cuidiN9GaTaOZgSEgsvPbh5A==",
"optional": true,
"engines": {
"bare": ">=1.14.0"
}
},
"node_modules/bare-path": {
"version": "3.0.0",
"resolved": "https://registry.npmjs.org/bare-path/-/bare-path-3.0.0.tgz",
"integrity": "sha512-tyfW2cQcB5NN8Saijrhqn0Zh7AnFNsnczRcuWODH0eYAXBsJ5gVxAUuNr7tsHSC6IZ77cA0SitzT+s47kot8Mw==",
"optional": true,
"dependencies": {
"bare-os": "^3.0.1"
}
},
"node_modules/bare-stream": {
"version": "2.7.0",
"resolved": "https://registry.npmjs.org/bare-stream/-/bare-stream-2.7.0.tgz",
"integrity": "sha512-oyXQNicV1y8nc2aKffH+BUHFRXmx6VrPzlnaEvMhram0nPBrKcEdcyBg5r08D0i8VxngHFAiVyn1QKXpSG0B8A==",
"optional": true,
"dependencies": {
"streamx": "^2.21.0"
},
"peerDependencies": {
"bare-buffer": "*",
"bare-events": "*"
},
"peerDependenciesMeta": {
"bare-buffer": {
"optional": true
},
"bare-events": {
"optional": true
}
}
},
"node_modules/bare-url": {
"version": "2.3.2",
"resolved": "https://registry.npmjs.org/bare-url/-/bare-url-2.3.2.tgz",
"integrity": "sha512-ZMq4gd9ngV5aTMa5p9+UfY0b3skwhHELaDkhEHetMdX0LRkW9kzaym4oo/Eh+Ghm0CCDuMTsRIGM/ytUc1ZYmw==",
"optional": true,
"dependencies": {
"bare-path": "^3.0.0"
}
},
"node_modules/base64-js": {
"version": "1.5.1",
"resolved": "https://registry.npmjs.org/base64-js/-/base64-js-1.5.1.tgz",
@@ -2725,14 +2539,6 @@
"node": ">=16.0.0"
}
},
"node_modules/hpagent": {
"version": "1.2.0",
"resolved": "https://registry.npmjs.org/hpagent/-/hpagent-1.2.0.tgz",
"integrity": "sha512-A91dYTeIB6NoXG+PxTQpCCDDnfHsW9kc06Lvpu1TEe9gnd6ZFeiBoRO9JvzEv6xK7EX97/dUE8g/vBMTqTS3CA==",
"engines": {
"node": ">=14"
}
},
"node_modules/htmlparser2": {
"version": "10.0.0",
"resolved": "https://registry.npmjs.org/htmlparser2/-/htmlparser2-10.0.0.tgz",
@@ -3096,22 +2902,6 @@
"node": ">=0.10.0"
}
},
"node_modules/isomorphic-ws": {
"version": "5.0.0",
"resolved": "https://registry.npmjs.org/isomorphic-ws/-/isomorphic-ws-5.0.0.tgz",
"integrity": "sha512-muId7Zzn9ywDsyXgTIafTry2sV3nySZeUDe6YedVd1Hvuuep5AsIlqK+XefWpYTyJG5e503F2xIuT2lcU6rCSw==",
"peerDependencies": {
"ws": "*"
}
},
"node_modules/jose": {
"version": "6.1.3",
"resolved": "https://registry.npmjs.org/jose/-/jose-6.1.3.tgz",
"integrity": "sha512-0TpaTfihd4QMNwrz/ob2Bp7X04yuxJkjRGi4aKmOqwhov54i6u79oCv7T+C7lo70MKH6BesI3vscD1yb/yzKXQ==",
"funding": {
"url": "https://github.com/sponsors/panva"
}
},
"node_modules/js-tokens": {
"version": "4.0.0",
"resolved": "https://registry.npmjs.org/js-tokens/-/js-tokens-4.0.0.tgz",
@@ -3128,14 +2918,6 @@
"js-yaml": "bin/js-yaml.js"
}
},
"node_modules/jsep": {
"version": "1.4.0",
"resolved": "https://registry.npmjs.org/jsep/-/jsep-1.4.0.tgz",
"integrity": "sha512-B7qPcEVE3NVkmSJbaYxvv4cHkVW7DQsZz13pUMrfS8z8Q/BuShN+gcTXrUlPiGqM2/t/EEaI030bpxMqY8gMlw==",
"engines": {
"node": ">= 10.16.0"
}
},
"node_modules/json-parse-even-better-errors": {
"version": "2.3.1",
"resolved": "https://registry.npmjs.org/json-parse-even-better-errors/-/json-parse-even-better-errors-2.3.1.tgz",
@@ -3157,23 +2939,6 @@
"graceful-fs": "^4.1.6"
}
},
"node_modules/jsonpath-plus": {
"version": "10.3.0",
"resolved": "https://registry.npmjs.org/jsonpath-plus/-/jsonpath-plus-10.3.0.tgz",
"integrity": "sha512-8TNmfeTCk2Le33A3vRRwtuworG/L5RrgMvdjhKZxvyShO+mBu2fP50OWUjRLNtvw344DdDarFh9buFAZs5ujeA==",
"dependencies": {
"@jsep-plugin/assignment": "^1.3.0",
"@jsep-plugin/regex": "^1.0.4",
"jsep": "^1.4.0"
},
"bin": {
"jsonpath": "bin/jsonpath-cli.js",
"jsonpath-plus": "bin/jsonpath-cli.js"
},
"engines": {
"node": ">=18.0.0"
}
},
"node_modules/jsonwebtoken": {
"version": "9.0.2",
"resolved": "https://registry.npmjs.org/jsonwebtoken/-/jsonwebtoken-9.0.2.tgz",
@@ -3248,11 +3013,6 @@
"resolved": "https://registry.npmjs.org/lodash/-/lodash-4.17.21.tgz",
"integrity": "sha512-v2kDEe57lecTulaDIuNTPy3Ry4gLGJ6Z1O3vE1krgXZNrsQ+LFTGHVxVjcXPs17LhbZVGedAJv8XZ1tvj5FvSg=="
},
"node_modules/lodash.clonedeep": {
"version": "4.5.0",
"resolved": "https://registry.npmjs.org/lodash.clonedeep/-/lodash.clonedeep-4.5.0.tgz",
"integrity": "sha512-H5ZhCF25riFd9uB5UCkVKo61m3S/xZk1x4wA6yp/L3RFP6Z/eHH1ymQcGLo7J3GMPfm0V/7m1tryHuGVxpqEBQ=="
},
"node_modules/lodash.defaults": {
"version": "4.2.0",
"resolved": "https://registry.npmjs.org/lodash.defaults/-/lodash.defaults-4.2.0.tgz",
@@ -3702,14 +3462,6 @@
"url": "https://github.com/fb55/nth-check?sponsor=1"
}
},
"node_modules/oauth4webapi": {
"version": "3.8.3",
"resolved": "https://registry.npmjs.org/oauth4webapi/-/oauth4webapi-3.8.3.tgz",
"integrity": "sha512-pQ5BsX3QRTgnt5HxgHwgunIRaDXBdkT23tf8dfzmtTIL2LTpdmxgbpbBm0VgFWAIDlezQvQCTgnVIUmHupXHxw==",
"funding": {
"url": "https://github.com/sponsors/panva"
}
},
"node_modules/object-assign": {
"version": "4.1.1",
"resolved": "https://registry.npmjs.org/object-assign/-/object-assign-4.1.1.tgz",
@@ -3748,18 +3500,6 @@
"wrappy": "1"
}
},
"node_modules/openid-client": {
"version": "6.8.1",
"resolved": "https://registry.npmjs.org/openid-client/-/openid-client-6.8.1.tgz",
"integrity": "sha512-VoYT6enBo6Vj2j3Q5Ec0AezS+9YGzQo1f5Xc42lreMGlfP4ljiXPKVDvCADh+XHCV/bqPu/wWSiCVXbJKvrODw==",
"dependencies": {
"jose": "^6.1.0",
"oauth4webapi": "^3.8.2"
},
"funding": {
"url": "https://github.com/sponsors/panva"
}
},
"node_modules/pac-proxy-agent": {
"version": "7.2.0",
"resolved": "https://registry.npmjs.org/pac-proxy-agent/-/pac-proxy-agent-7.2.0.tgz",
@@ -4676,11 +4416,6 @@
"url": "https://github.com/privatenumber/resolve-pkg-maps?sponsor=1"
}
},
"node_modules/rfc4648": {
"version": "1.5.4",
"resolved": "https://registry.npmjs.org/rfc4648/-/rfc4648-1.5.4.tgz",
"integrity": "sha512-rRg/6Lb+IGfJqO05HZkN50UtY7K/JhxJag1kP23+zyMfrvoB0B7RWv06MbOzoc79RgCdNTiUaNsTT1AJZ7Z+cg=="
},
"node_modules/rimraf": {
"version": "3.0.2",
"resolved": "https://registry.npmjs.org/rimraf/-/rimraf-3.0.2.tgz",
@@ -5111,14 +4846,6 @@
"node": ">= 0.8"
}
},
"node_modules/stream-buffers": {
"version": "3.0.3",
"resolved": "https://registry.npmjs.org/stream-buffers/-/stream-buffers-3.0.3.tgz",
"integrity": "sha512-pqMqwQCso0PBJt2PQmDO0cFj0lyqmiwOMiMSkVtRokl7e+ZTRYgDHKnuZNbqjiJXgsg4nuqtD/zxuo9KqTp0Yw==",
"engines": {
"node": ">= 0.10.0"
}
},
"node_modules/streamx": {
"version": "2.23.0",
"resolved": "https://registry.npmjs.org/streamx/-/streamx-2.23.0.tgz",
@@ -5338,7 +5065,8 @@
"node_modules/undici-types": {
"version": "6.21.0",
"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-6.21.0.tgz",
"integrity": "sha512-iwDZqg0QAGrg9Rav5H4n0M64c3mkR59cJ6wQp+7C4nI0gsmExaedaYLNO44eT4AtBBwjbTiGPMlt2Md0T9H9JQ=="
"integrity": "sha512-iwDZqg0QAGrg9Rav5H4n0M64c3mkR59cJ6wQp+7C4nI0gsmExaedaYLNO44eT4AtBBwjbTiGPMlt2Md0T9H9JQ==",
"devOptional": true
},
"node_modules/universalify": {
"version": "2.0.1",
@@ -5361,14 +5089,6 @@
"resolved": "https://registry.npmjs.org/urlpattern-polyfill/-/urlpattern-polyfill-10.0.0.tgz",
"integrity": "sha512-H/A06tKD7sS1O1X2SshBVeA5FLycRpjqiBeqGKmBwBDBy28EnRjORxTNe269KSSr5un5qyWi1iL61wLxpd+ZOg=="
},
"node_modules/user-agents": {
"version": "1.1.669",
"resolved": "https://registry.npmjs.org/user-agents/-/user-agents-1.1.669.tgz",
"integrity": "sha512-pbIzG+AOqCaIpySKJ4IAm1l0VyE4jMnK4y1thV8lm8PYxI+7X5uWcppOK7zY79TCKKTAnJH3/4gaVIZHsjrmJA==",
"dependencies": {
"lodash.clonedeep": "^4.5.0"
}
},
"node_modules/util": {
"version": "0.12.5",
"resolved": "https://registry.npmjs.org/util/-/util-0.12.5.tgz",

View File

@@ -22,7 +22,6 @@
"seed:dt:cities:bulk": "tsx src/scripts/seed-dt-cities-bulk.ts"
},
"dependencies": {
"@kubernetes/client-node": "^1.4.0",
"@types/bcryptjs": "^3.0.0",
"axios": "^1.6.2",
"bcrypt": "^5.1.1",
@@ -49,7 +48,6 @@
"puppeteer-extra-plugin-stealth": "^2.11.2",
"sharp": "^0.32.0",
"socks-proxy-agent": "^8.0.2",
"user-agents": "^1.1.669",
"uuid": "^9.0.1",
"zod": "^3.22.4"
},

View File

@@ -1,46 +0,0 @@
# DEPRECATED CODE - DO NOT USE
**These directories contain OLD, ABANDONED code.**
## What's Here
| Directory | What It Was | Why Deprecated |
|-----------|-------------|----------------|
| `hydration/` | Old pipeline for processing crawl data | Replaced by `src/tasks/handlers/` |
| `scraper-v2/` | Old Puppeteer-based scraper engine | Replaced by curl-based `src/platforms/dutchie/client.ts` |
| `canonical-hydration/` | Intermediate step toward canonical schema | Merged into task handlers |
## What to Use Instead
| Old (DONT USE) | New (USE THIS) |
|----------------|----------------|
| `hydration/normalizers/dutchie.ts` | `src/tasks/handlers/product-refresh.ts` |
| `hydration/producer.ts` | `src/tasks/handlers/payload-fetch.ts` |
| `scraper-v2/engine.ts` | `src/platforms/dutchie/client.ts` |
| `scraper-v2/scheduler.ts` | `src/services/task-scheduler.ts` |
## Why Keep This Code?
- Historical reference only
- Some patterns may be useful for debugging
- Will be deleted once confirmed not needed
## Claude Instructions
**IF YOU ARE CLAUDE:**
1. NEVER import from `src/_deprecated/`
2. NEVER reference these files as examples
3. NEVER try to "fix" or "update" code in here
4. If you see imports from these directories, suggest replacing them
**Correct imports:**
```typescript
// GOOD
import { executeGraphQL } from '../platforms/dutchie/client';
import { pool } from '../db/pool';
// BAD - DO NOT USE
import { something } from '../_deprecated/hydration/...';
import { something } from '../_deprecated/scraper-v2/...';
```

View File

@@ -1,584 +0,0 @@
/**
* System API Routes
*
* Provides REST API endpoints for system monitoring and control:
* - /api/system/sync/* - Sync orchestrator
* - /api/system/dlq/* - Dead-letter queue
* - /api/system/integrity/* - Integrity checks
* - /api/system/fix/* - Auto-fix routines
* - /api/system/alerts/* - System alerts
* - /metrics - Prometheus metrics
*
* Phase 5: Full Production Sync + Monitoring
*/
import { Router, Request, Response } from 'express';
import { Pool } from 'pg';
import {
SyncOrchestrator,
MetricsService,
DLQService,
AlertService,
IntegrityService,
AutoFixService,
} from '../services';
export function createSystemRouter(pool: Pool): Router {
const router = Router();
// Initialize services
const metrics = new MetricsService(pool);
const dlq = new DLQService(pool);
const alerts = new AlertService(pool);
const integrity = new IntegrityService(pool, alerts);
const autoFix = new AutoFixService(pool, alerts);
const orchestrator = new SyncOrchestrator(pool, metrics, dlq, alerts);
// ============================================================
// SYNC ORCHESTRATOR ENDPOINTS
// ============================================================
/**
* GET /api/system/sync/status
* Get current sync status
*/
router.get('/sync/status', async (_req: Request, res: Response) => {
try {
const status = await orchestrator.getStatus();
res.json(status);
} catch (error) {
console.error('[System] Sync status error:', error);
res.status(500).json({ error: 'Failed to get sync status' });
}
});
/**
* POST /api/system/sync/run
* Trigger a sync run
*/
router.post('/sync/run', async (req: Request, res: Response) => {
try {
const triggeredBy = req.body.triggeredBy || 'api';
const result = await orchestrator.runSync();
res.json({
success: true,
triggeredBy,
metrics: result,
});
} catch (error) {
console.error('[System] Sync run error:', error);
res.status(500).json({
success: false,
error: error instanceof Error ? error.message : 'Sync run failed',
});
}
});
/**
* GET /api/system/sync/queue-depth
* Get queue depth information
*/
router.get('/sync/queue-depth', async (_req: Request, res: Response) => {
try {
const depth = await orchestrator.getQueueDepth();
res.json(depth);
} catch (error) {
console.error('[System] Queue depth error:', error);
res.status(500).json({ error: 'Failed to get queue depth' });
}
});
/**
* GET /api/system/sync/health
* Get sync health status
*/
router.get('/sync/health', async (_req: Request, res: Response) => {
try {
const health = await orchestrator.getHealth();
res.status(health.healthy ? 200 : 503).json(health);
} catch (error) {
console.error('[System] Health check error:', error);
res.status(500).json({ healthy: false, error: 'Health check failed' });
}
});
/**
* POST /api/system/sync/pause
* Pause the orchestrator
*/
router.post('/sync/pause', async (req: Request, res: Response) => {
try {
const reason = req.body.reason || 'Manual pause';
await orchestrator.pause(reason);
res.json({ success: true, message: 'Orchestrator paused' });
} catch (error) {
console.error('[System] Pause error:', error);
res.status(500).json({ error: 'Failed to pause orchestrator' });
}
});
/**
* POST /api/system/sync/resume
* Resume the orchestrator
*/
router.post('/sync/resume', async (_req: Request, res: Response) => {
try {
await orchestrator.resume();
res.json({ success: true, message: 'Orchestrator resumed' });
} catch (error) {
console.error('[System] Resume error:', error);
res.status(500).json({ error: 'Failed to resume orchestrator' });
}
});
// ============================================================
// DLQ ENDPOINTS
// ============================================================
/**
* GET /api/system/dlq
* List DLQ payloads
*/
router.get('/dlq', async (req: Request, res: Response) => {
try {
const options = {
status: req.query.status as string,
errorType: req.query.errorType as string,
dispensaryId: req.query.dispensaryId ? parseInt(req.query.dispensaryId as string) : undefined,
limit: req.query.limit ? parseInt(req.query.limit as string) : 50,
offset: req.query.offset ? parseInt(req.query.offset as string) : 0,
};
const result = await dlq.listPayloads(options);
res.json(result);
} catch (error) {
console.error('[System] DLQ list error:', error);
res.status(500).json({ error: 'Failed to list DLQ payloads' });
}
});
/**
* GET /api/system/dlq/stats
* Get DLQ statistics
*/
router.get('/dlq/stats', async (_req: Request, res: Response) => {
try {
const stats = await dlq.getStats();
res.json(stats);
} catch (error) {
console.error('[System] DLQ stats error:', error);
res.status(500).json({ error: 'Failed to get DLQ stats' });
}
});
/**
* GET /api/system/dlq/summary
* Get DLQ summary by error type
*/
router.get('/dlq/summary', async (_req: Request, res: Response) => {
try {
const summary = await dlq.getSummary();
res.json(summary);
} catch (error) {
console.error('[System] DLQ summary error:', error);
res.status(500).json({ error: 'Failed to get DLQ summary' });
}
});
/**
* GET /api/system/dlq/:id
* Get a specific DLQ payload
*/
router.get('/dlq/:id', async (req: Request, res: Response) => {
try {
const payload = await dlq.getPayload(req.params.id);
if (!payload) {
return res.status(404).json({ error: 'Payload not found' });
}
res.json(payload);
} catch (error) {
console.error('[System] DLQ get error:', error);
res.status(500).json({ error: 'Failed to get DLQ payload' });
}
});
/**
* POST /api/system/dlq/:id/retry
* Retry a DLQ payload
*/
router.post('/dlq/:id/retry', async (req: Request, res: Response) => {
try {
const result = await dlq.retryPayload(req.params.id);
if (result.success) {
res.json(result);
} else {
res.status(400).json(result);
}
} catch (error) {
console.error('[System] DLQ retry error:', error);
res.status(500).json({ error: 'Failed to retry payload' });
}
});
/**
* POST /api/system/dlq/:id/abandon
* Abandon a DLQ payload
*/
router.post('/dlq/:id/abandon', async (req: Request, res: Response) => {
try {
const reason = req.body.reason || 'Manually abandoned';
const abandonedBy = req.body.abandonedBy || 'api';
const success = await dlq.abandonPayload(req.params.id, reason, abandonedBy);
res.json({ success });
} catch (error) {
console.error('[System] DLQ abandon error:', error);
res.status(500).json({ error: 'Failed to abandon payload' });
}
});
/**
* POST /api/system/dlq/bulk-retry
* Bulk retry payloads by error type
*/
router.post('/dlq/bulk-retry', async (req: Request, res: Response) => {
try {
const { errorType } = req.body;
if (!errorType) {
return res.status(400).json({ error: 'errorType is required' });
}
const result = await dlq.bulkRetryByErrorType(errorType);
res.json(result);
} catch (error) {
console.error('[System] DLQ bulk retry error:', error);
res.status(500).json({ error: 'Failed to bulk retry' });
}
});
// ============================================================
// INTEGRITY CHECK ENDPOINTS
// ============================================================
/**
* POST /api/system/integrity/run
* Run all integrity checks
*/
router.post('/integrity/run', async (req: Request, res: Response) => {
try {
const triggeredBy = req.body.triggeredBy || 'api';
const result = await integrity.runAllChecks(triggeredBy);
res.json(result);
} catch (error) {
console.error('[System] Integrity run error:', error);
res.status(500).json({ error: 'Failed to run integrity checks' });
}
});
/**
* GET /api/system/integrity/runs
* Get recent integrity check runs
*/
router.get('/integrity/runs', async (req: Request, res: Response) => {
try {
const limit = req.query.limit ? parseInt(req.query.limit as string) : 10;
const runs = await integrity.getRecentRuns(limit);
res.json(runs);
} catch (error) {
console.error('[System] Integrity runs error:', error);
res.status(500).json({ error: 'Failed to get integrity runs' });
}
});
/**
* GET /api/system/integrity/runs/:runId
* Get results for a specific integrity run
*/
router.get('/integrity/runs/:runId', async (req: Request, res: Response) => {
try {
const results = await integrity.getRunResults(req.params.runId);
res.json(results);
} catch (error) {
console.error('[System] Integrity run results error:', error);
res.status(500).json({ error: 'Failed to get run results' });
}
});
// ============================================================
// AUTO-FIX ENDPOINTS
// ============================================================
/**
* GET /api/system/fix/routines
* Get available fix routines
*/
router.get('/fix/routines', (_req: Request, res: Response) => {
try {
const routines = autoFix.getAvailableRoutines();
res.json(routines);
} catch (error) {
console.error('[System] Get routines error:', error);
res.status(500).json({ error: 'Failed to get routines' });
}
});
/**
* POST /api/system/fix/:routine
* Run a fix routine
*/
router.post('/fix/:routine', async (req: Request, res: Response) => {
try {
const routineName = req.params.routine;
const dryRun = req.body.dryRun === true;
const triggeredBy = req.body.triggeredBy || 'api';
const result = await autoFix.runRoutine(routineName as any, triggeredBy, { dryRun });
res.json(result);
} catch (error) {
console.error('[System] Fix routine error:', error);
res.status(500).json({ error: 'Failed to run fix routine' });
}
});
/**
* GET /api/system/fix/runs
* Get recent fix runs
*/
router.get('/fix/runs', async (req: Request, res: Response) => {
try {
const limit = req.query.limit ? parseInt(req.query.limit as string) : 20;
const runs = await autoFix.getRecentRuns(limit);
res.json(runs);
} catch (error) {
console.error('[System] Fix runs error:', error);
res.status(500).json({ error: 'Failed to get fix runs' });
}
});
// ============================================================
// ALERTS ENDPOINTS
// ============================================================
/**
* GET /api/system/alerts
* List alerts
*/
router.get('/alerts', async (req: Request, res: Response) => {
try {
const options = {
status: req.query.status as any,
severity: req.query.severity as any,
type: req.query.type as string,
limit: req.query.limit ? parseInt(req.query.limit as string) : 50,
offset: req.query.offset ? parseInt(req.query.offset as string) : 0,
};
const result = await alerts.listAlerts(options);
res.json(result);
} catch (error) {
console.error('[System] Alerts list error:', error);
res.status(500).json({ error: 'Failed to list alerts' });
}
});
/**
* GET /api/system/alerts/active
* Get active alerts
*/
router.get('/alerts/active', async (_req: Request, res: Response) => {
try {
const activeAlerts = await alerts.getActiveAlerts();
res.json(activeAlerts);
} catch (error) {
console.error('[System] Active alerts error:', error);
res.status(500).json({ error: 'Failed to get active alerts' });
}
});
/**
* GET /api/system/alerts/summary
* Get alert summary
*/
router.get('/alerts/summary', async (_req: Request, res: Response) => {
try {
const summary = await alerts.getSummary();
res.json(summary);
} catch (error) {
console.error('[System] Alerts summary error:', error);
res.status(500).json({ error: 'Failed to get alerts summary' });
}
});
/**
* POST /api/system/alerts/:id/acknowledge
* Acknowledge an alert
*/
router.post('/alerts/:id/acknowledge', async (req: Request, res: Response) => {
try {
const alertId = parseInt(req.params.id);
const acknowledgedBy = req.body.acknowledgedBy || 'api';
const success = await alerts.acknowledgeAlert(alertId, acknowledgedBy);
res.json({ success });
} catch (error) {
console.error('[System] Acknowledge alert error:', error);
res.status(500).json({ error: 'Failed to acknowledge alert' });
}
});
/**
* POST /api/system/alerts/:id/resolve
* Resolve an alert
*/
router.post('/alerts/:id/resolve', async (req: Request, res: Response) => {
try {
const alertId = parseInt(req.params.id);
const resolvedBy = req.body.resolvedBy || 'api';
const success = await alerts.resolveAlert(alertId, resolvedBy);
res.json({ success });
} catch (error) {
console.error('[System] Resolve alert error:', error);
res.status(500).json({ error: 'Failed to resolve alert' });
}
});
/**
* POST /api/system/alerts/bulk-acknowledge
* Bulk acknowledge alerts
*/
router.post('/alerts/bulk-acknowledge', async (req: Request, res: Response) => {
try {
const { ids, acknowledgedBy } = req.body;
if (!ids || !Array.isArray(ids)) {
return res.status(400).json({ error: 'ids array is required' });
}
const count = await alerts.bulkAcknowledge(ids, acknowledgedBy || 'api');
res.json({ acknowledged: count });
} catch (error) {
console.error('[System] Bulk acknowledge error:', error);
res.status(500).json({ error: 'Failed to bulk acknowledge' });
}
});
// ============================================================
// METRICS ENDPOINTS
// ============================================================
/**
* GET /api/system/metrics
* Get all current metrics
*/
router.get('/metrics', async (_req: Request, res: Response) => {
try {
const allMetrics = await metrics.getAllMetrics();
res.json(allMetrics);
} catch (error) {
console.error('[System] Metrics error:', error);
res.status(500).json({ error: 'Failed to get metrics' });
}
});
/**
* GET /api/system/metrics/:name
* Get a specific metric
*/
router.get('/metrics/:name', async (req: Request, res: Response) => {
try {
const metric = await metrics.getMetric(req.params.name);
if (!metric) {
return res.status(404).json({ error: 'Metric not found' });
}
res.json(metric);
} catch (error) {
console.error('[System] Metric error:', error);
res.status(500).json({ error: 'Failed to get metric' });
}
});
/**
* GET /api/system/metrics/:name/history
* Get metric time series
*/
router.get('/metrics/:name/history', async (req: Request, res: Response) => {
try {
const hours = req.query.hours ? parseInt(req.query.hours as string) : 24;
const history = await metrics.getMetricHistory(req.params.name, hours);
res.json(history);
} catch (error) {
console.error('[System] Metric history error:', error);
res.status(500).json({ error: 'Failed to get metric history' });
}
});
/**
* GET /api/system/errors
* Get error summary
*/
router.get('/errors', async (_req: Request, res: Response) => {
try {
const summary = await metrics.getErrorSummary();
res.json(summary);
} catch (error) {
console.error('[System] Error summary error:', error);
res.status(500).json({ error: 'Failed to get error summary' });
}
});
/**
* GET /api/system/errors/recent
* Get recent errors
*/
router.get('/errors/recent', async (req: Request, res: Response) => {
try {
const limit = req.query.limit ? parseInt(req.query.limit as string) : 50;
const errorType = req.query.type as string;
const errors = await metrics.getRecentErrors(limit, errorType);
res.json(errors);
} catch (error) {
console.error('[System] Recent errors error:', error);
res.status(500).json({ error: 'Failed to get recent errors' });
}
});
/**
* POST /api/system/errors/acknowledge
* Acknowledge errors
*/
router.post('/errors/acknowledge', async (req: Request, res: Response) => {
try {
const { ids, acknowledgedBy } = req.body;
if (!ids || !Array.isArray(ids)) {
return res.status(400).json({ error: 'ids array is required' });
}
const count = await metrics.acknowledgeErrors(ids, acknowledgedBy || 'api');
res.json({ acknowledged: count });
} catch (error) {
console.error('[System] Acknowledge errors error:', error);
res.status(500).json({ error: 'Failed to acknowledge errors' });
}
});
return router;
}
/**
* Create Prometheus metrics endpoint (standalone)
*/
export function createPrometheusRouter(pool: Pool): Router {
const router = Router();
const metrics = new MetricsService(pool);
/**
* GET /metrics
* Prometheus-compatible metrics endpoint
*/
router.get('/', async (_req: Request, res: Response) => {
try {
const prometheusOutput = await metrics.getPrometheusMetrics();
res.set('Content-Type', 'text/plain; version=0.0.4');
res.send(prometheusOutput);
} catch (error) {
console.error('[Prometheus] Metrics error:', error);
res.status(500).send('# Error generating metrics');
}
});
return router;
}

View File

@@ -172,9 +172,6 @@ export async function runFullDiscovery(
console.log(`Errors: ${totalErrors}`);
}
// Per TASK_WORKFLOW_2024-12-10.md: Track new dispensary IDs for task chaining
let newDispensaryIds: number[] = [];
// Step 4: Auto-validate and promote discovered locations
if (!dryRun && totalLocationsUpserted > 0) {
console.log('\n[Discovery] Step 4: Auto-promoting discovered locations...');
@@ -183,13 +180,6 @@ export async function runFullDiscovery(
console.log(` Created: ${promotionResult.created} new dispensaries`);
console.log(` Updated: ${promotionResult.updated} existing dispensaries`);
console.log(` Rejected: ${promotionResult.rejected} (validation failed)`);
// Per TASK_WORKFLOW_2024-12-10.md: Capture new IDs for task chaining
newDispensaryIds = promotionResult.newDispensaryIds;
if (newDispensaryIds.length > 0) {
console.log(` New store IDs for crawl: [${newDispensaryIds.join(', ')}]`);
}
if (promotionResult.rejectedRecords.length > 0) {
console.log(` Rejection reasons:`);
promotionResult.rejectedRecords.slice(0, 5).forEach(r => {
@@ -224,8 +214,6 @@ export async function runFullDiscovery(
totalLocationsFound,
totalLocationsUpserted,
durationMs,
// Per TASK_WORKFLOW_2024-12-10.md: Return new IDs for task chaining
newDispensaryIds,
};
}

View File

@@ -127,8 +127,6 @@ export interface PromotionSummary {
errors: string[];
}>;
durationMs: number;
// Per TASK_WORKFLOW_2024-12-10.md: Track new dispensary IDs for task chaining
newDispensaryIds: number[];
}
/**
@@ -471,8 +469,6 @@ export async function promoteDiscoveredLocations(
const results: PromotionResult[] = [];
const rejectedRecords: PromotionSummary['rejectedRecords'] = [];
// Per TASK_WORKFLOW_2024-12-10.md: Track new dispensary IDs for task chaining
const newDispensaryIds: number[] = [];
let created = 0;
let updated = 0;
let skipped = 0;
@@ -529,8 +525,6 @@ export async function promoteDiscoveredLocations(
if (promotionResult.action === 'created') {
created++;
// Per TASK_WORKFLOW_2024-12-10.md: Track new IDs for task chaining
newDispensaryIds.push(promotionResult.dispensaryId);
} else {
updated++;
}
@@ -554,8 +548,6 @@ export async function promoteDiscoveredLocations(
results,
rejectedRecords,
durationMs: Date.now() - startTime,
// Per TASK_WORKFLOW_2024-12-10.md: Return new IDs for task chaining
newDispensaryIds,
};
}

View File

@@ -211,8 +211,6 @@ export interface FullDiscoveryResult {
totalLocationsFound: number;
totalLocationsUpserted: number;
durationMs: number;
// Per TASK_WORKFLOW_2024-12-10.md: Track new dispensary IDs for task chaining
newDispensaryIds?: number[];
}
// ============================================================

View File

@@ -6,8 +6,6 @@ import { initializeMinio, isMinioEnabled } from './utils/minio';
import { initializeImageStorage } from './utils/image-storage';
import { logger } from './services/logger';
import { cleanupOrphanedJobs } from './services/proxyTestQueue';
// Per TASK_WORKFLOW_2024-12-10.md: Database-driven task scheduler
import { taskScheduler } from './services/task-scheduler';
import { runAutoMigrations } from './db/auto-migrate';
import { getPool } from './db/pool';
import healthRoutes from './routes/health';
@@ -109,7 +107,7 @@ import scraperMonitorRoutes from './routes/scraper-monitor';
import apiTokensRoutes from './routes/api-tokens';
import apiPermissionsRoutes from './routes/api-permissions';
import parallelScrapeRoutes from './routes/parallel-scrape';
// crawler-sandbox moved to _deprecated
import crawlerSandboxRoutes from './routes/crawler-sandbox';
import versionRoutes from './routes/version';
import deployStatusRoutes from './routes/deploy-status';
import publicApiRoutes from './routes/public-api';
@@ -144,9 +142,6 @@ import seoRoutes from './routes/seo';
import priceAnalyticsRoutes from './routes/price-analytics';
import tasksRoutes from './routes/tasks';
import workerRegistryRoutes from './routes/worker-registry';
// Per TASK_WORKFLOW_2024-12-10.md: Raw payload access API
import payloadsRoutes from './routes/payloads';
import k8sRoutes from './routes/k8s';
// Mark requests from trusted domains (cannaiq.co, findagram.co, findadispo.com)
// These domains can access the API without authentication
@@ -187,7 +182,7 @@ app.use('/api/scraper-monitor', scraperMonitorRoutes);
app.use('/api/api-tokens', apiTokensRoutes);
app.use('/api/api-permissions', apiPermissionsRoutes);
app.use('/api/parallel-scrape', parallelScrapeRoutes);
// crawler-sandbox moved to _deprecated
app.use('/api/crawler-sandbox', crawlerSandboxRoutes);
app.use('/api/version', versionRoutes);
app.use('/api/admin/deploy-status', deployStatusRoutes);
console.log('[DeployStatus] Routes registered at /api/admin/deploy-status');
@@ -227,14 +222,6 @@ console.log('[Tasks] Routes registered at /api/tasks');
app.use('/api/worker-registry', workerRegistryRoutes);
console.log('[WorkerRegistry] Routes registered at /api/worker-registry');
// Per TASK_WORKFLOW_2024-12-10.md: Raw payload access API
app.use('/api/payloads', payloadsRoutes);
console.log('[Payloads] Routes registered at /api/payloads');
// K8s control routes - worker scaling from admin UI
app.use('/api/k8s', k8sRoutes);
console.log('[K8s] Routes registered at /api/k8s');
// Phase 3: Analytics V2 - Enhanced analytics with rec/med state segmentation
try {
const analyticsV2Router = createAnalyticsV2Router(getPool());
@@ -339,17 +326,6 @@ async function startServer() {
// Clean up any orphaned proxy test jobs from previous server runs
await cleanupOrphanedJobs();
// Per TASK_WORKFLOW_2024-12-10.md: Start database-driven task scheduler
// This replaces node-cron - schedules are stored in DB and survive restarts
// Uses SELECT FOR UPDATE SKIP LOCKED for multi-replica safety
try {
await taskScheduler.start();
logger.info('system', 'Task scheduler started');
} catch (err: any) {
// Non-fatal - scheduler can recover on next poll
logger.warn('system', `Task scheduler startup warning: ${err.message}`);
}
app.listen(PORT, () => {
logger.info('system', `Server running on port ${PORT}`);
console.log(`🚀 Server running on port ${PORT}`);

View File

@@ -702,10 +702,12 @@ export class StateQueryService {
async getNationalSummary(): Promise<NationalSummary> {
const stateMetrics = await this.getAllStateMetrics();
// Get all states count and aggregate metrics
const result = await this.pool.query(`
SELECT
COUNT(DISTINCT s.code) AS total_states,
COUNT(DISTINCT CASE WHEN EXISTS (
SELECT 1 FROM dispensaries d WHERE d.state = s.code AND d.menu_type IS NOT NULL
) THEN s.code END) AS active_states,
(SELECT COUNT(*) FROM dispensaries WHERE state IS NOT NULL) AS total_stores,
(SELECT COUNT(*) FROM store_products sp
JOIN dispensaries d ON sp.dispensary_id = d.id
@@ -723,7 +725,7 @@ export class StateQueryService {
return {
totalStates: parseInt(data.total_states),
activeStates: parseInt(data.total_states), // Same as totalStates - all states shown
activeStates: parseInt(data.active_states),
totalStores: parseInt(data.total_stores),
totalProducts: parseInt(data.total_products),
totalBrands: parseInt(data.total_brands),

View File

@@ -5,35 +5,22 @@
*
* DO NOT MODIFY THIS FILE WITHOUT EXPLICIT AUTHORIZATION.
*
* Updated: 2025-12-10 per workflow-12102025.md
*
* KEY BEHAVIORS (per workflow-12102025.md):
* 1. startSession() gets identity from PROXY LOCATION, not task params
* 2. On 403: immediately get new IP + new fingerprint, then retry
* 3. After 3 consecutive 403s on same proxy → disable it (burned)
* 4. Language is always English (en-US)
* This is the canonical HTTP client for all Dutchie communication.
* All Dutchie workers (Alice, Bella, etc.) MUST use this client.
*
* IMPLEMENTATION:
* - Uses curl via child_process.execSync (bypasses TLS fingerprinting)
* - NO Puppeteer, NO axios, NO fetch
* - Uses intoli/user-agents via CrawlRotator for realistic fingerprints
* - Fingerprint rotation on 403
* - Residential IP compatible
*
* USAGE:
* import { curlPost, curlGet, executeGraphQL, startSession } from '@dutchie/client';
* import { curlPost, curlGet, executeGraphQL } from '@dutchie/client';
*
* ============================================================
*/
import { execSync } from 'child_process';
import {
buildOrderedHeaders,
buildRefererFromMenuUrl,
getCurlBinary,
isCurlImpersonateAvailable,
HeaderContext,
BrowserType,
} from '../../services/http-fingerprint';
// ============================================================
// TYPES
@@ -45,8 +32,6 @@ export interface CurlResponse {
error?: string;
}
// Per workflow-12102025.md: fingerprint comes from CrawlRotator's BrowserFingerprint
// We keep a simplified interface here for header building
export interface Fingerprint {
userAgent: string;
acceptLanguage: string;
@@ -72,13 +57,15 @@ export const DUTCHIE_CONFIG = {
// ============================================================
// PROXY SUPPORT
// Per workflow-12102025.md:
// - On 403: recordBlock() → increment consecutive_403_count
// - After 3 consecutive 403s → proxy disabled
// - Immediately rotate to new IP + new fingerprint on 403
// ============================================================
// Integrates with the CrawlRotator system from proxy-rotator.ts
// On 403 errors:
// 1. Record failure on current proxy
// 2. Rotate to next proxy
// 3. Retry with new proxy
// ============================================================
import type { CrawlRotator, BrowserFingerprint } from '../../services/crawl-rotator';
import type { CrawlRotator, Proxy } from '../../services/crawl-rotator';
let currentProxy: string | null = null;
let crawlRotator: CrawlRotator | null = null;
@@ -105,12 +92,13 @@ export function getProxy(): string | null {
/**
* Set CrawlRotator for proxy rotation on 403s
* Per workflow-12102025.md: enables automatic rotation when blocked
* This enables automatic proxy rotation when blocked
*/
export function setCrawlRotator(rotator: CrawlRotator | null): void {
crawlRotator = rotator;
if (rotator) {
console.log('[Dutchie Client] CrawlRotator attached - proxy rotation enabled');
// Set initial proxy from rotator
const proxy = rotator.proxy.getCurrent();
if (proxy) {
currentProxy = rotator.proxy.getProxyUrl(proxy);
@@ -127,41 +115,30 @@ export function getCrawlRotator(): CrawlRotator | null {
}
/**
* Handle 403 block - per workflow-12102025.md:
* 1. Record block on current proxy (increments consecutive_403_count)
* 2. Immediately rotate to new proxy (new IP)
* 3. Rotate fingerprint
* Returns false if no more proxies available
* Rotate to next proxy (called on 403)
*/
async function handle403Block(): Promise<boolean> {
async function rotateProxyOn403(error?: string): Promise<boolean> {
if (!crawlRotator) {
console.warn('[Dutchie Client] No CrawlRotator - cannot handle 403');
return false;
}
// Per workflow-12102025.md: record block (tracks consecutive 403s)
const wasDisabled = await crawlRotator.recordBlock();
if (wasDisabled) {
console.log('[Dutchie Client] Current proxy was disabled (3 consecutive 403s)');
}
// Per workflow-12102025.md: immediately get new IP + new fingerprint
const { proxy: nextProxy, fingerprint } = crawlRotator.rotateBoth();
// Record failure on current proxy
await crawlRotator.recordFailure(error || '403 Forbidden');
// Rotate to next proxy
const nextProxy = crawlRotator.rotateProxy();
if (nextProxy) {
currentProxy = crawlRotator.proxy.getProxyUrl(nextProxy);
console.log(`[Dutchie Client] Rotated to new proxy: ${currentProxy.replace(/:[^:@]+@/, ':***@')}`);
console.log(`[Dutchie Client] New fingerprint: ${fingerprint.userAgent.slice(0, 50)}...`);
console.log(`[Dutchie Client] Rotated proxy: ${currentProxy.replace(/:[^:@]+@/, ':***@')}`);
return true;
}
console.error('[Dutchie Client] No more proxies available!');
console.warn('[Dutchie Client] No more proxies available');
return false;
}
/**
* Record success on current proxy
* Per workflow-12102025.md: resets consecutive_403_count
*/
async function recordProxySuccess(responseTimeMs?: number): Promise<void> {
if (crawlRotator) {
@@ -185,69 +162,163 @@ export const GRAPHQL_HASHES = {
GetAllCitiesByState: 'ae547a0466ace5a48f91e55bf6699eacd87e3a42841560f0c0eabed5a0a920e6',
};
// ============================================================
// FINGERPRINTS - Browser profiles for anti-detect
// ============================================================
const FINGERPRINTS: Fingerprint[] = [
// Chrome Windows (latest) - typical residential user, use first
{
userAgent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36',
acceptLanguage: 'en-US,en;q=0.9',
secChUa: '"Google Chrome";v="131", "Chromium";v="131", "Not_A Brand";v="24"',
secChUaPlatform: '"Windows"',
secChUaMobile: '?0',
},
// Chrome Mac (latest)
{
userAgent: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36',
acceptLanguage: 'en-US,en;q=0.9',
secChUa: '"Google Chrome";v="131", "Chromium";v="131", "Not_A Brand";v="24"',
secChUaPlatform: '"macOS"',
secChUaMobile: '?0',
},
// Chrome Windows (120)
{
userAgent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
acceptLanguage: 'en-US,en;q=0.9',
secChUa: '"Chromium";v="120", "Google Chrome";v="120", "Not-A.Brand";v="99"',
secChUaPlatform: '"Windows"',
secChUaMobile: '?0',
},
// Firefox Windows
{
userAgent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:133.0) Gecko/20100101 Firefox/133.0',
acceptLanguage: 'en-US,en;q=0.5',
},
// Safari Mac
{
userAgent: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 14_2) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.2 Safari/605.1.15',
acceptLanguage: 'en-US,en;q=0.9',
},
// Edge Windows
{
userAgent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36 Edg/131.0.0.0',
acceptLanguage: 'en-US,en;q=0.9',
secChUa: '"Microsoft Edge";v="131", "Chromium";v="131", "Not_A Brand";v="24"',
secChUaPlatform: '"Windows"',
secChUaMobile: '?0',
},
];
let currentFingerprintIndex = 0;
// Forward declaration for session (actual CrawlSession interface defined later)
let currentSession: {
sessionId: string;
fingerprint: Fingerprint;
proxyUrl: string | null;
stateCode?: string;
timezone?: string;
startedAt: Date;
} | null = null;
/**
* Get current fingerprint - returns session fingerprint if active, otherwise default
*/
export function getFingerprint(): Fingerprint {
// Use session fingerprint if a session is active
if (currentSession) {
return currentSession.fingerprint;
}
return FINGERPRINTS[currentFingerprintIndex];
}
export function rotateFingerprint(): Fingerprint {
currentFingerprintIndex = (currentFingerprintIndex + 1) % FINGERPRINTS.length;
const fp = FINGERPRINTS[currentFingerprintIndex];
console.log(`[Dutchie Client] Rotated to fingerprint: ${fp.userAgent.slice(0, 50)}...`);
return fp;
}
export function resetFingerprint(): void {
currentFingerprintIndex = 0;
}
/**
* Get a random fingerprint from the pool
*/
export function getRandomFingerprint(): Fingerprint {
const index = Math.floor(Math.random() * FINGERPRINTS.length);
return FINGERPRINTS[index];
}
// ============================================================
// SESSION MANAGEMENT
// Per workflow-12102025.md:
// - Session identity comes from PROXY LOCATION
// - NOT from task params (no stateCode/timezone params)
// - Language is always English
// Per-session fingerprint rotation for stealth
// ============================================================
export interface CrawlSession {
sessionId: string;
fingerprint: BrowserFingerprint;
fingerprint: Fingerprint;
proxyUrl: string | null;
proxyTimezone?: string;
proxyState?: string;
stateCode?: string;
timezone?: string;
startedAt: Date;
// Per workflow-12102025.md: Dynamic Referer per dispensary
menuUrl?: string;
referer: string;
}
let currentSession: CrawlSession | null = null;
// Note: currentSession variable declared earlier in file for proper scoping
/**
* Start a new crawl session
*
* Per workflow-12102025.md:
* - NO state/timezone params - identity comes from proxy location
* - Gets fingerprint from CrawlRotator (uses intoli/user-agents)
* - Language is always English (en-US)
* - Dynamic Referer per dispensary (from menuUrl)
*
* @param menuUrl - The dispensary's menu URL for dynamic Referer header
* Timezone to Accept-Language mapping
* US timezones all use en-US but this can be extended for international
*/
export function startSession(menuUrl?: string): CrawlSession {
if (!crawlRotator) {
throw new Error('[Dutchie Client] Cannot start session without CrawlRotator');
}
const TIMEZONE_TO_LOCALE: Record<string, string> = {
'America/Phoenix': 'en-US,en;q=0.9',
'America/Los_Angeles': 'en-US,en;q=0.9',
'America/Denver': 'en-US,en;q=0.9',
'America/Chicago': 'en-US,en;q=0.9',
'America/New_York': 'en-US,en;q=0.9',
'America/Detroit': 'en-US,en;q=0.9',
'America/Anchorage': 'en-US,en;q=0.9',
'Pacific/Honolulu': 'en-US,en;q=0.9',
};
// Per workflow-12102025.md: get identity from proxy location
const proxyLocation = crawlRotator.getProxyLocation();
const fingerprint = crawlRotator.userAgent.getCurrent();
/**
* Get Accept-Language header for a given timezone
*/
export function getLocaleForTimezone(timezone?: string): string {
if (!timezone) return 'en-US,en;q=0.9';
return TIMEZONE_TO_LOCALE[timezone] || 'en-US,en;q=0.9';
}
// Per workflow-12102025.md: Dynamic Referer per dispensary
const referer = buildRefererFromMenuUrl(menuUrl);
/**
* Start a new crawl session with a random fingerprint
* Call this before crawling a store to get a fresh identity
*/
export function startSession(stateCode?: string, timezone?: string): CrawlSession {
const baseFp = getRandomFingerprint();
// Override Accept-Language based on timezone for geographic consistency
const fingerprint: Fingerprint = {
...baseFp,
acceptLanguage: getLocaleForTimezone(timezone),
};
currentSession = {
sessionId: `session_${Date.now()}_${Math.random().toString(36).slice(2, 8)}`,
fingerprint,
proxyUrl: currentProxy,
proxyTimezone: proxyLocation?.timezone,
proxyState: proxyLocation?.state,
stateCode,
timezone,
startedAt: new Date(),
menuUrl,
referer,
};
console.log(`[Dutchie Client] Started session ${currentSession.sessionId}`);
console.log(`[Dutchie Client] Browser: ${fingerprint.browserName} (${fingerprint.deviceCategory})`);
console.log(`[Dutchie Client] DNT: ${fingerprint.httpFingerprint.hasDNT ? 'enabled' : 'disabled'}`);
console.log(`[Dutchie Client] TLS: ${fingerprint.httpFingerprint.curlImpersonateBinary}`);
console.log(`[Dutchie Client] Referer: ${referer}`);
if (proxyLocation?.timezone) {
console.log(`[Dutchie Client] Proxy: ${proxyLocation.state || 'unknown'} (${proxyLocation.timezone})`);
console.log(`[Dutchie Client] Fingerprint: ${fingerprint.userAgent.slice(0, 50)}...`);
console.log(`[Dutchie Client] Accept-Language: ${fingerprint.acceptLanguage}`);
if (timezone) {
console.log(`[Dutchie Client] Timezone: ${timezone}`);
}
return currentSession;
@@ -276,80 +347,48 @@ export function getCurrentSession(): CrawlSession | null {
// ============================================================
/**
* Per workflow-12102025.md: Build headers using HTTP fingerprint system
* Returns headers in browser-specific order with all natural variations
* Build headers for Dutchie requests
*/
export function buildHeaders(isPost: boolean, contentLength?: number): { headers: Record<string, string>; orderedHeaders: string[] } {
if (!currentSession || !crawlRotator) {
throw new Error('[Dutchie Client] Cannot build headers without active session');
}
export function buildHeaders(refererPath: string, fingerprint?: Fingerprint): Record<string, string> {
const fp = fingerprint || getFingerprint();
const refererUrl = `https://dutchie.com${refererPath}`;
const fp = currentSession.fingerprint;
const httpFp = fp.httpFingerprint;
// Per workflow-12102025.md: Build context for ordered headers
const context: HeaderContext = {
userAgent: fp.userAgent,
secChUa: fp.secChUa,
secChUaPlatform: fp.secChUaPlatform,
secChUaMobile: fp.secChUaMobile,
referer: currentSession.referer,
isPost,
contentLength,
const headers: Record<string, string> = {
'accept': 'application/json, text/plain, */*',
'accept-language': fp.acceptLanguage,
'content-type': 'application/json',
'origin': 'https://dutchie.com',
'referer': refererUrl,
'user-agent': fp.userAgent,
'apollographql-client-name': 'Marketplace (production)',
};
// Per workflow-12102025.md: Get ordered headers from HTTP fingerprint service
return buildOrderedHeaders(httpFp, context);
if (fp.secChUa) {
headers['sec-ch-ua'] = fp.secChUa;
headers['sec-ch-ua-mobile'] = fp.secChUaMobile || '?0';
headers['sec-ch-ua-platform'] = fp.secChUaPlatform || '"Windows"';
headers['sec-fetch-dest'] = 'empty';
headers['sec-fetch-mode'] = 'cors';
headers['sec-fetch-site'] = 'same-site';
}
return headers;
}
/**
* Per workflow-12102025.md: Get curl binary for current session's browser
* Uses curl-impersonate for TLS fingerprint matching
* Execute HTTP POST using curl (bypasses TLS fingerprinting)
*/
function getCurlBinaryForSession(): string {
if (!currentSession) {
return 'curl'; // Fallback to standard curl
}
const browserType = currentSession.fingerprint.browserName as BrowserType;
// Per workflow-12102025.md: Check if curl-impersonate is available
if (isCurlImpersonateAvailable(browserType)) {
return getCurlBinary(browserType);
}
// Fallback to standard curl with warning
console.warn(`[Dutchie Client] curl-impersonate not available for ${browserType}, using standard curl`);
return 'curl';
}
/**
* Per workflow-12102025.md: Execute HTTP POST using curl/curl-impersonate
* - Uses browser-specific TLS fingerprint via curl-impersonate
* - Headers sent in browser-specific order
* - Dynamic Referer per dispensary
*/
export function curlPost(url: string, body: any, timeout = 30000): CurlResponse {
const bodyJson = JSON.stringify(body);
// Per workflow-12102025.md: Build ordered headers for POST request
const { headers, orderedHeaders } = buildHeaders(true, bodyJson.length);
// Per workflow-12102025.md: Build header args in browser-specific order
const headerArgs = orderedHeaders
.filter(h => h !== 'Host' && h !== 'Content-Length') // curl handles these
.map(h => `-H '${h}: ${headers[h]}'`)
export function curlPost(url: string, body: any, headers: Record<string, string>, timeout = 30000): CurlResponse {
const filteredHeaders = Object.entries(headers)
.filter(([k]) => k.toLowerCase() !== 'accept-encoding')
.map(([k, v]) => `-H '${k}: ${v}'`)
.join(' ');
const bodyEscaped = bodyJson.replace(/'/g, "'\\''");
const bodyJson = JSON.stringify(body).replace(/'/g, "'\\''");
const timeoutSec = Math.ceil(timeout / 1000);
const separator = '___HTTP_STATUS___';
const proxyArg = getProxyArg();
// Per workflow-12102025.md: Use curl-impersonate for TLS fingerprint matching
const curlBinary = getCurlBinaryForSession();
const cmd = `${curlBinary} -s --compressed ${proxyArg} -w '${separator}%{http_code}' --max-time ${timeoutSec} ${headerArgs} -d '${bodyEscaped}' '${url}'`;
const cmd = `curl -s --compressed ${proxyArg} -w '${separator}%{http_code}' --max-time ${timeoutSec} ${filteredHeaders} -d '${bodyJson}' '${url}'`;
try {
const output = execSync(cmd, {
@@ -388,29 +427,19 @@ export function curlPost(url: string, body: any, timeout = 30000): CurlResponse
}
/**
* Per workflow-12102025.md: Execute HTTP GET using curl/curl-impersonate
* - Uses browser-specific TLS fingerprint via curl-impersonate
* - Headers sent in browser-specific order
* - Dynamic Referer per dispensary
* Execute HTTP GET using curl (bypasses TLS fingerprinting)
* Returns HTML or JSON depending on response content-type
*/
export function curlGet(url: string, timeout = 30000): CurlResponse {
// Per workflow-12102025.md: Build ordered headers for GET request
const { headers, orderedHeaders } = buildHeaders(false);
// Per workflow-12102025.md: Build header args in browser-specific order
const headerArgs = orderedHeaders
.filter(h => h !== 'Host' && h !== 'Content-Length') // curl handles these
.map(h => `-H '${h}: ${headers[h]}'`)
export function curlGet(url: string, headers: Record<string, string>, timeout = 30000): CurlResponse {
const filteredHeaders = Object.entries(headers)
.filter(([k]) => k.toLowerCase() !== 'accept-encoding')
.map(([k, v]) => `-H '${k}: ${v}'`)
.join(' ');
const timeoutSec = Math.ceil(timeout / 1000);
const separator = '___HTTP_STATUS___';
const proxyArg = getProxyArg();
// Per workflow-12102025.md: Use curl-impersonate for TLS fingerprint matching
const curlBinary = getCurlBinaryForSession();
const cmd = `${curlBinary} -s --compressed ${proxyArg} -w '${separator}%{http_code}' --max-time ${timeoutSec} ${headerArgs} '${url}'`;
const cmd = `curl -s --compressed ${proxyArg} -w '${separator}%{http_code}' --max-time ${timeoutSec} ${filteredHeaders} '${url}'`;
try {
const output = execSync(cmd, {
@@ -430,6 +459,7 @@ export function curlGet(url: string, timeout = 30000): CurlResponse {
const responseBody = output.slice(0, separatorIndex);
const statusCode = parseInt(output.slice(separatorIndex + separator.length).trim(), 10);
// Try to parse as JSON, otherwise return as string (HTML)
try {
return { status: statusCode, data: JSON.parse(responseBody) };
} catch {
@@ -446,22 +476,16 @@ export function curlGet(url: string, timeout = 30000): CurlResponse {
// ============================================================
// GRAPHQL EXECUTION
// Per workflow-12102025.md:
// - On 403: immediately rotate IP + fingerprint (no delay first)
// - Then retry
// ============================================================
export interface ExecuteGraphQLOptions {
maxRetries?: number;
retryOn403?: boolean;
cName?: string;
cName?: string; // Optional - used for Referer header, defaults to 'cities'
}
/**
* Per workflow-12102025.md: Execute GraphQL query with curl/curl-impersonate
* - Uses browser-specific TLS fingerprint
* - Headers in browser-specific order
* - On 403: immediately rotate IP + fingerprint, then retry
* Execute GraphQL query with curl (bypasses TLS fingerprinting)
*/
export async function executeGraphQL(
operationName: string,
@@ -469,12 +493,7 @@ export async function executeGraphQL(
hash: string,
options: ExecuteGraphQLOptions
): Promise<any> {
const { maxRetries = 3, retryOn403 = true } = options;
// Per workflow-12102025.md: Session must be active for requests
if (!currentSession) {
throw new Error('[Dutchie Client] Cannot execute GraphQL without active session - call startSession() first');
}
const { maxRetries = 3, retryOn403 = true, cName = 'cities' } = options;
const body = {
operationName,
@@ -488,14 +507,14 @@ export async function executeGraphQL(
let attempt = 0;
while (attempt <= maxRetries) {
const fingerprint = getFingerprint();
const headers = buildHeaders(`/embedded-menu/${cName}`, fingerprint);
console.log(`[Dutchie Client] curl POST ${operationName} (attempt ${attempt + 1}/${maxRetries + 1})`);
const startTime = Date.now();
// Per workflow-12102025.md: curlPost now uses ordered headers and curl-impersonate
const response = curlPost(DUTCHIE_CONFIG.graphqlEndpoint, body, DUTCHIE_CONFIG.timeout);
const responseTime = Date.now() - startTime;
const response = curlPost(DUTCHIE_CONFIG.graphqlEndpoint, body, headers, DUTCHIE_CONFIG.timeout);
console.log(`[Dutchie Client] Response status: ${response.status} (${responseTime}ms)`);
console.log(`[Dutchie Client] Response status: ${response.status}`);
if (response.error) {
console.error(`[Dutchie Client] curl error: ${response.error}`);
@@ -508,9 +527,6 @@ export async function executeGraphQL(
}
if (response.status === 200) {
// Per workflow-12102025.md: success resets consecutive 403 count
await recordProxySuccess(responseTime);
if (response.data?.errors?.length > 0) {
console.warn(`[Dutchie Client] GraphQL errors: ${JSON.stringify(response.data.errors[0])}`);
}
@@ -518,20 +534,11 @@ export async function executeGraphQL(
}
if (response.status === 403 && retryOn403) {
// Per workflow-12102025.md: immediately rotate IP + fingerprint
console.warn(`[Dutchie Client] 403 blocked - immediately rotating proxy + fingerprint...`);
const hasMoreProxies = await handle403Block();
if (!hasMoreProxies) {
throw new Error('All proxies exhausted - no more IPs available');
}
// Per workflow-12102025.md: Update session referer after rotation
currentSession.referer = buildRefererFromMenuUrl(currentSession.menuUrl);
console.warn(`[Dutchie Client] 403 blocked - rotating proxy and fingerprint...`);
await rotateProxyOn403('403 Forbidden on GraphQL');
rotateFingerprint();
attempt++;
// Per workflow-12102025.md: small backoff after rotation
await sleep(500);
await sleep(1000 * attempt);
continue;
}
@@ -560,10 +567,8 @@ export interface FetchPageOptions {
}
/**
* Per workflow-12102025.md: Fetch HTML page from Dutchie
* - Uses browser-specific TLS fingerprint
* - Headers in browser-specific order
* - Same 403 handling as GraphQL
* Fetch HTML page from Dutchie (for city pages, dispensary pages, etc.)
* Returns raw HTML string
*/
export async function fetchPage(
path: string,
@@ -572,22 +577,32 @@ export async function fetchPage(
const { maxRetries = 3, retryOn403 = true } = options;
const url = `${DUTCHIE_CONFIG.baseUrl}${path}`;
// Per workflow-12102025.md: Session must be active for requests
if (!currentSession) {
throw new Error('[Dutchie Client] Cannot fetch page without active session - call startSession() first');
}
let attempt = 0;
while (attempt <= maxRetries) {
// Per workflow-12102025.md: curlGet now uses ordered headers and curl-impersonate
const fingerprint = getFingerprint();
const headers: Record<string, string> = {
'accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8',
'accept-language': fingerprint.acceptLanguage,
'user-agent': fingerprint.userAgent,
};
if (fingerprint.secChUa) {
headers['sec-ch-ua'] = fingerprint.secChUa;
headers['sec-ch-ua-mobile'] = fingerprint.secChUaMobile || '?0';
headers['sec-ch-ua-platform'] = fingerprint.secChUaPlatform || '"Windows"';
headers['sec-fetch-dest'] = 'document';
headers['sec-fetch-mode'] = 'navigate';
headers['sec-fetch-site'] = 'none';
headers['sec-fetch-user'] = '?1';
headers['upgrade-insecure-requests'] = '1';
}
console.log(`[Dutchie Client] curl GET ${path} (attempt ${attempt + 1}/${maxRetries + 1})`);
const startTime = Date.now();
const response = curlGet(url, DUTCHIE_CONFIG.timeout);
const responseTime = Date.now() - startTime;
const response = curlGet(url, headers, DUTCHIE_CONFIG.timeout);
console.log(`[Dutchie Client] Response status: ${response.status} (${responseTime}ms)`);
console.log(`[Dutchie Client] Response status: ${response.status}`);
if (response.error) {
console.error(`[Dutchie Client] curl error: ${response.error}`);
@@ -599,26 +614,15 @@ export async function fetchPage(
}
if (response.status === 200) {
// Per workflow-12102025.md: success resets consecutive 403 count
await recordProxySuccess(responseTime);
return { html: response.data, status: response.status };
}
if (response.status === 403 && retryOn403) {
// Per workflow-12102025.md: immediately rotate IP + fingerprint
console.warn(`[Dutchie Client] 403 blocked - immediately rotating proxy + fingerprint...`);
const hasMoreProxies = await handle403Block();
if (!hasMoreProxies) {
throw new Error('All proxies exhausted - no more IPs available');
}
// Per workflow-12102025.md: Update session after rotation
currentSession.referer = buildRefererFromMenuUrl(currentSession.menuUrl);
console.warn(`[Dutchie Client] 403 blocked - rotating proxy and fingerprint...`);
await rotateProxyOn403('403 Forbidden on page fetch');
rotateFingerprint();
attempt++;
// Per workflow-12102025.md: small backoff after rotation
await sleep(500);
await sleep(1000 * attempt);
continue;
}

View File

@@ -6,17 +6,22 @@
*/
export {
// HTTP Client (per workflow-12102025.md: uses curl-impersonate + ordered headers)
// HTTP Client
curlPost,
curlGet,
executeGraphQL,
fetchPage,
extractNextData,
// Headers (per workflow-12102025.md: browser-specific ordering)
// Headers & Fingerprints
buildHeaders,
getFingerprint,
rotateFingerprint,
resetFingerprint,
getRandomFingerprint,
getLocaleForTimezone,
// Session Management (per workflow-12102025.md: menuUrl for dynamic Referer)
// Session Management (per-store fingerprint rotation)
startSession,
endSession,
getCurrentSession,

View File

@@ -47,27 +47,4 @@ router.post('/refresh', authMiddleware, async (req: AuthRequest, res) => {
res.json({ token });
});
// Verify password for sensitive actions (requires current user to be authenticated)
router.post('/verify-password', authMiddleware, async (req: AuthRequest, res) => {
try {
const { password } = req.body;
if (!password) {
return res.status(400).json({ error: 'Password required' });
}
// Re-authenticate the current user with the provided password
const user = await authenticateUser(req.user!.email, password);
if (!user) {
return res.status(401).json({ error: 'Invalid password', verified: false });
}
res.json({ verified: true });
} catch (error) {
console.error('Password verification error:', error);
res.status(500).json({ error: 'Internal server error' });
}
});
export default router;

View File

@@ -14,25 +14,13 @@ router.use(authMiddleware);
/**
* GET /api/admin/intelligence/brands
* List all brands with state presence, store counts, and pricing
* Query params:
* - state: Filter by state (e.g., "AZ")
* - limit: Max results (default 500)
* - offset: Pagination offset
*/
router.get('/brands', async (req: Request, res: Response) => {
try {
const { limit = '500', offset = '0', state } = req.query;
const { limit = '500', offset = '0' } = req.query;
const limitNum = Math.min(parseInt(limit as string, 10), 1000);
const offsetNum = parseInt(offset as string, 10);
// Build WHERE clause based on state filter
let stateFilter = '';
const params: any[] = [limitNum, offsetNum];
if (state && state !== 'all') {
stateFilter = 'AND d.state = $3';
params.push(state);
}
const { rows } = await pool.query(`
SELECT
sp.brand_name_raw as brand_name,
@@ -44,26 +32,17 @@ router.get('/brands', async (req: Request, res: Response) => {
FROM store_products sp
JOIN dispensaries d ON sp.dispensary_id = d.id
WHERE sp.brand_name_raw IS NOT NULL AND sp.brand_name_raw != ''
${stateFilter}
GROUP BY sp.brand_name_raw
ORDER BY store_count DESC, sku_count DESC
LIMIT $1 OFFSET $2
`, params);
`, [limitNum, offsetNum]);
// Get total count with same state filter
const countParams: any[] = [];
let countStateFilter = '';
if (state && state !== 'all') {
countStateFilter = 'AND d.state = $1';
countParams.push(state);
}
// Get total count
const { rows: countRows } = await pool.query(`
SELECT COUNT(DISTINCT sp.brand_name_raw) as total
FROM store_products sp
JOIN dispensaries d ON sp.dispensary_id = d.id
WHERE sp.brand_name_raw IS NOT NULL AND sp.brand_name_raw != ''
${countStateFilter}
`, countParams);
SELECT COUNT(DISTINCT brand_name_raw) as total
FROM store_products
WHERE brand_name_raw IS NOT NULL AND brand_name_raw != ''
`);
res.json({
brands: rows.map((r: any) => ({
@@ -168,42 +147,10 @@ router.get('/brands/:brandName/penetration', async (req: Request, res: Response)
/**
* GET /api/admin/intelligence/pricing
* Get pricing analytics by category
* Query params:
* - state: Filter by state (e.g., "AZ")
*/
router.get('/pricing', async (req: Request, res: Response) => {
try {
const { state } = req.query;
// Build WHERE clause based on state filter
let stateFilter = '';
const categoryParams: any[] = [];
const stateQueryParams: any[] = [];
const overallParams: any[] = [];
if (state && state !== 'all') {
stateFilter = 'AND d.state = $1';
categoryParams.push(state);
overallParams.push(state);
}
// Category pricing with optional state filter
const categoryQuery = state && state !== 'all'
? `
SELECT
sp.category_raw as category,
ROUND(AVG(sp.price_rec)::numeric, 2) as avg_price,
MIN(sp.price_rec) as min_price,
MAX(sp.price_rec) as max_price,
ROUND(PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY sp.price_rec)::numeric, 2) as median_price,
COUNT(*) as product_count
FROM store_products sp
JOIN dispensaries d ON sp.dispensary_id = d.id
WHERE sp.category_raw IS NOT NULL AND sp.price_rec > 0 ${stateFilter}
GROUP BY sp.category_raw
ORDER BY product_count DESC
`
: `
const { rows: categoryRows } = await pool.query(`
SELECT
sp.category_raw as category,
ROUND(AVG(sp.price_rec)::numeric, 2) as avg_price,
@@ -215,11 +162,8 @@ router.get('/pricing', async (req: Request, res: Response) => {
WHERE sp.category_raw IS NOT NULL AND sp.price_rec > 0
GROUP BY sp.category_raw
ORDER BY product_count DESC
`;
`);
const { rows: categoryRows } = await pool.query(categoryQuery, categoryParams);
// State pricing
const { rows: stateRows } = await pool.query(`
SELECT
d.state,
@@ -234,31 +178,6 @@ router.get('/pricing', async (req: Request, res: Response) => {
ORDER BY avg_price DESC
`);
// Overall stats with optional state filter
const overallQuery = state && state !== 'all'
? `
SELECT
ROUND(AVG(sp.price_rec)::numeric, 2) as avg_price,
MIN(sp.price_rec) as min_price,
MAX(sp.price_rec) as max_price,
COUNT(*) as total_products
FROM store_products sp
JOIN dispensaries d ON sp.dispensary_id = d.id
WHERE sp.price_rec > 0 ${stateFilter}
`
: `
SELECT
ROUND(AVG(sp.price_rec)::numeric, 2) as avg_price,
MIN(sp.price_rec) as min_price,
MAX(sp.price_rec) as max_price,
COUNT(*) as total_products
FROM store_products sp
WHERE sp.price_rec > 0
`;
const { rows: overallRows } = await pool.query(overallQuery, overallParams);
const overall = overallRows[0];
res.json({
byCategory: categoryRows.map((r: any) => ({
category: r.category,
@@ -275,12 +194,6 @@ router.get('/pricing', async (req: Request, res: Response) => {
maxPrice: r.max_price ? parseFloat(r.max_price) : null,
productCount: parseInt(r.product_count, 10),
})),
overall: {
avgPrice: overall?.avg_price ? parseFloat(overall.avg_price) : null,
minPrice: overall?.min_price ? parseFloat(overall.min_price) : null,
maxPrice: overall?.max_price ? parseFloat(overall.max_price) : null,
totalProducts: parseInt(overall?.total_products || '0', 10),
},
});
} catch (error: any) {
console.error('[Intelligence] Error fetching pricing:', error.message);
@@ -291,23 +204,9 @@ router.get('/pricing', async (req: Request, res: Response) => {
/**
* GET /api/admin/intelligence/stores
* Get store intelligence summary
* Query params:
* - state: Filter by state (e.g., "AZ")
* - limit: Max results (default 200)
*/
router.get('/stores', async (req: Request, res: Response) => {
try {
const { state, limit = '200' } = req.query;
const limitNum = Math.min(parseInt(limit as string, 10), 500);
// Build WHERE clause based on state filter
let stateFilter = '';
const params: any[] = [limitNum];
if (state && state !== 'all') {
stateFilter = 'AND d.state = $2';
params.push(state);
}
const { rows: storeRows } = await pool.query(`
SELECT
d.id,
@@ -317,22 +216,17 @@ router.get('/stores', async (req: Request, res: Response) => {
d.state,
d.menu_type,
d.crawl_enabled,
c.name as chain_name,
COUNT(DISTINCT sp.id) as sku_count,
COUNT(DISTINCT sp.id) as product_count,
COUNT(DISTINCT sp.brand_name_raw) as brand_count,
ROUND(AVG(sp.price_rec)::numeric, 2) as avg_price,
MAX(sp.updated_at) as last_crawl,
(SELECT COUNT(*) FROM store_product_snapshots sps
WHERE sps.store_product_id IN (SELECT id FROM store_products WHERE dispensary_id = d.id)) as snapshot_count
MAX(sp.updated_at) as last_product_update
FROM dispensaries d
LEFT JOIN store_products sp ON sp.dispensary_id = d.id
LEFT JOIN chains c ON d.chain_id = c.id
WHERE d.state IS NOT NULL AND d.crawl_enabled = true
${stateFilter}
GROUP BY d.id, d.name, d.dba_name, d.city, d.state, d.menu_type, d.crawl_enabled, c.name
ORDER BY sku_count DESC
LIMIT $1
`, params);
WHERE d.state IS NOT NULL
GROUP BY d.id, d.name, d.dba_name, d.city, d.state, d.menu_type, d.crawl_enabled
ORDER BY product_count DESC
LIMIT 200
`);
res.json({
stores: storeRows.map((r: any) => ({
@@ -343,13 +237,10 @@ router.get('/stores', async (req: Request, res: Response) => {
state: r.state,
menuType: r.menu_type,
crawlEnabled: r.crawl_enabled,
chainName: r.chain_name || null,
skuCount: parseInt(r.sku_count || '0', 10),
snapshotCount: parseInt(r.snapshot_count || '0', 10),
productCount: parseInt(r.product_count || '0', 10),
brandCount: parseInt(r.brand_count || '0', 10),
avgPrice: r.avg_price ? parseFloat(r.avg_price) : null,
lastCrawl: r.last_crawl,
crawlFrequencyHours: 4, // Default crawl frequency
lastProductUpdate: r.last_product_update,
})),
total: storeRows.length,
});

View File

@@ -543,9 +543,6 @@ router.post('/bulk-priority', async (req: Request, res: Response) => {
/**
* POST /api/job-queue/enqueue - Add a new job to the queue
*
* 2024-12-10: Rewired to use worker_tasks via taskService.
* Legacy dispensary_crawl_jobs code commented out below.
*/
router.post('/enqueue', async (req: Request, res: Response) => {
try {
@@ -555,59 +552,6 @@ router.post('/enqueue', async (req: Request, res: Response) => {
return res.status(400).json({ success: false, error: 'dispensary_id is required' });
}
// 2024-12-10: Map legacy job_type to new task role
const roleMap: Record<string, string> = {
'dutchie_product_crawl': 'product_refresh',
'menu_detection': 'entry_point_discovery',
'menu_detection_single': 'entry_point_discovery',
'product_discovery': 'product_discovery',
'store_discovery': 'store_discovery',
};
const role = roleMap[job_type] || 'product_refresh';
// 2024-12-10: Use taskService to create task in worker_tasks table
const { taskService } = await import('../tasks/task-service');
// Check if task already pending for this dispensary
const existingTasks = await taskService.listTasks({
dispensary_id,
role: role as any,
status: ['pending', 'claimed', 'running'],
limit: 1,
});
if (existingTasks.length > 0) {
return res.json({
success: true,
task_id: existingTasks[0].id,
message: 'Task already queued'
});
}
const task = await taskService.createTask({
role: role as any,
dispensary_id,
priority,
});
res.json({ success: true, task_id: task.id, message: 'Task enqueued' });
} catch (error: any) {
console.error('[JobQueue] Error enqueuing task:', error);
res.status(500).json({ success: false, error: error.message });
}
});
/*
* LEGACY CODE - 2024-12-10: Commented out, was using orphaned dispensary_crawl_jobs table
*
router.post('/enqueue', async (req: Request, res: Response) => {
try {
const { dispensary_id, job_type = 'dutchie_product_crawl', priority = 0 } = req.body;
if (!dispensary_id) {
return res.status(400).json({ success: false, error: 'dispensary_id is required' });
}
// Check if job already pending for this dispensary
const existing = await pool.query(`
SELECT id FROM dispensary_crawl_jobs
@@ -641,7 +585,6 @@ router.post('/enqueue', async (req: Request, res: Response) => {
res.status(500).json({ success: false, error: error.message });
}
});
*/
/**
* POST /api/job-queue/pause - Pause queue processing
@@ -669,8 +612,6 @@ router.get('/paused', async (_req: Request, res: Response) => {
/**
* POST /api/job-queue/enqueue-batch - Queue multiple dispensaries at once
* Body: { dispensary_ids: number[], job_type?: string, priority?: number }
*
* 2024-12-10: Rewired to use worker_tasks via taskService.
*/
router.post('/enqueue-batch', async (req: Request, res: Response) => {
try {
@@ -684,30 +625,35 @@ router.post('/enqueue-batch', async (req: Request, res: Response) => {
return res.status(400).json({ success: false, error: 'Maximum 500 dispensaries per batch' });
}
// 2024-12-10: Map legacy job_type to new task role
const roleMap: Record<string, string> = {
'dutchie_product_crawl': 'product_refresh',
'menu_detection': 'entry_point_discovery',
'product_discovery': 'product_discovery',
};
const role = roleMap[job_type] || 'product_refresh';
// 2024-12-10: Use taskService to create tasks in worker_tasks table
const { taskService } = await import('../tasks/task-service');
const tasks = dispensary_ids.map(dispensary_id => ({
role: role as any,
dispensary_id,
priority,
}));
const createdCount = await taskService.createTasks(tasks);
// Insert jobs, skipping duplicates
const { rows } = await pool.query(`
INSERT INTO dispensary_crawl_jobs (dispensary_id, job_type, priority, trigger_type, status, created_at)
SELECT
d.id,
$2::text,
$3::integer,
'api_batch',
'pending',
NOW()
FROM dispensaries d
WHERE d.id = ANY($1::int[])
AND d.crawl_enabled = true
AND d.platform_dispensary_id IS NOT NULL
AND NOT EXISTS (
SELECT 1 FROM dispensary_crawl_jobs cj
WHERE cj.dispensary_id = d.id
AND cj.job_type = $2::text
AND cj.status IN ('pending', 'running')
)
RETURNING id, dispensary_id
`, [dispensary_ids, job_type, priority]);
res.json({
success: true,
queued: createdCount,
queued: rows.length,
requested: dispensary_ids.length,
message: `Queued ${createdCount} of ${dispensary_ids.length} dispensaries`
job_ids: rows.map(r => r.id),
message: `Queued ${rows.length} of ${dispensary_ids.length} dispensaries`
});
} catch (error: any) {
console.error('[JobQueue] Error batch enqueuing:', error);
@@ -718,8 +664,6 @@ router.post('/enqueue-batch', async (req: Request, res: Response) => {
/**
* POST /api/job-queue/enqueue-state - Queue all crawl-enabled dispensaries for a state
* Body: { state_code: string, job_type?: string, priority?: number, limit?: number }
*
* 2024-12-10: Rewired to use worker_tasks via taskService.
*/
router.post('/enqueue-state', async (req: Request, res: Response) => {
try {
@@ -729,55 +673,52 @@ router.post('/enqueue-state', async (req: Request, res: Response) => {
return res.status(400).json({ success: false, error: 'state_code is required (e.g., "AZ")' });
}
// 2024-12-10: Map legacy job_type to new task role
const roleMap: Record<string, string> = {
'dutchie_product_crawl': 'product_refresh',
'menu_detection': 'entry_point_discovery',
'product_discovery': 'product_discovery',
};
const role = roleMap[job_type] || 'product_refresh';
// Get dispensary IDs for the state
const dispensaryResult = await pool.query(`
SELECT d.id
FROM dispensaries d
JOIN states s ON s.id = d.state_id
WHERE s.code = $1
// Get state_id and queue jobs
const { rows } = await pool.query(`
WITH target_state AS (
SELECT id FROM states WHERE code = $1
)
INSERT INTO dispensary_crawl_jobs (dispensary_id, job_type, priority, trigger_type, status, created_at)
SELECT
d.id,
$2::text,
$3::integer,
'api_state',
'pending',
NOW()
FROM dispensaries d, target_state
WHERE d.state_id = target_state.id
AND d.crawl_enabled = true
AND d.platform_dispensary_id IS NOT NULL
LIMIT $2
`, [state_code.toUpperCase(), limit]);
const dispensary_ids = dispensaryResult.rows.map((r: any) => r.id);
// 2024-12-10: Use taskService to create tasks in worker_tasks table
const { taskService } = await import('../tasks/task-service');
const tasks = dispensary_ids.map((dispensary_id: number) => ({
role: role as any,
dispensary_id,
priority,
}));
const createdCount = await taskService.createTasks(tasks);
AND NOT EXISTS (
SELECT 1 FROM dispensary_crawl_jobs cj
WHERE cj.dispensary_id = d.id
AND cj.job_type = $2::text
AND cj.status IN ('pending', 'running')
)
LIMIT $4::integer
RETURNING id, dispensary_id
`, [state_code.toUpperCase(), job_type, priority, limit]);
// Get total available count
const countResult = await pool.query(`
WITH target_state AS (
SELECT id FROM states WHERE code = $1
)
SELECT COUNT(*) as total
FROM dispensaries d
JOIN states s ON s.id = d.state_id
WHERE s.code = $1
FROM dispensaries d, target_state
WHERE d.state_id = target_state.id
AND d.crawl_enabled = true
AND d.platform_dispensary_id IS NOT NULL
`, [state_code.toUpperCase()]);
res.json({
success: true,
queued: createdCount,
queued: rows.length,
total_available: parseInt(countResult.rows[0].total),
state: state_code.toUpperCase(),
role,
message: `Queued ${createdCount} dispensaries for ${state_code.toUpperCase()}`
job_type,
message: `Queued ${rows.length} dispensaries for ${state_code.toUpperCase()}`
});
} catch (error: any) {
console.error('[JobQueue] Error enqueuing state:', error);

View File

@@ -1,140 +0,0 @@
/**
* Kubernetes Control Routes
*
* Provides admin UI control over k8s resources like worker scaling.
* Uses in-cluster config when running in k8s, or kubeconfig locally.
*/
import { Router, Request, Response } from 'express';
import * as k8s from '@kubernetes/client-node';
const router = Router();
// K8s client setup - lazy initialization
let appsApi: k8s.AppsV1Api | null = null;
let k8sError: string | null = null;
function getK8sClient(): k8s.AppsV1Api | null {
if (appsApi) return appsApi;
if (k8sError) return null;
try {
const kc = new k8s.KubeConfig();
// Try in-cluster config first (when running in k8s)
try {
kc.loadFromCluster();
console.log('[K8s] Loaded in-cluster config');
} catch {
// Fall back to default kubeconfig (local dev)
try {
kc.loadFromDefault();
console.log('[K8s] Loaded default kubeconfig');
} catch (e) {
k8sError = 'No k8s config available';
console.log('[K8s] No config available - k8s routes disabled');
return null;
}
}
appsApi = kc.makeApiClient(k8s.AppsV1Api);
return appsApi;
} catch (e: any) {
k8sError = e.message;
console.error('[K8s] Failed to initialize client:', e.message);
return null;
}
}
const NAMESPACE = process.env.K8S_NAMESPACE || 'dispensary-scraper';
const WORKER_DEPLOYMENT = 'scraper-worker';
/**
* GET /api/k8s/workers
* Get current worker deployment status
*/
router.get('/workers', async (_req: Request, res: Response) => {
const client = getK8sClient();
if (!client) {
return res.json({
success: true,
available: false,
error: k8sError || 'K8s not available',
replicas: 0,
readyReplicas: 0,
});
}
try {
const deployment = await client.readNamespacedDeployment({
name: WORKER_DEPLOYMENT,
namespace: NAMESPACE,
});
res.json({
success: true,
available: true,
replicas: deployment.spec?.replicas || 0,
readyReplicas: deployment.status?.readyReplicas || 0,
availableReplicas: deployment.status?.availableReplicas || 0,
updatedReplicas: deployment.status?.updatedReplicas || 0,
});
} catch (e: any) {
console.error('[K8s] Error getting deployment:', e.message);
res.status(500).json({
success: false,
error: e.message,
});
}
});
/**
* POST /api/k8s/workers/scale
* Scale worker deployment
* Body: { replicas: number }
*/
router.post('/workers/scale', async (req: Request, res: Response) => {
const client = getK8sClient();
if (!client) {
return res.status(503).json({
success: false,
error: k8sError || 'K8s not available',
});
}
const { replicas } = req.body;
if (typeof replicas !== 'number' || replicas < 0 || replicas > 50) {
return res.status(400).json({
success: false,
error: 'replicas must be a number between 0 and 50',
});
}
try {
// Patch the deployment to set replicas
await client.patchNamespacedDeploymentScale({
name: WORKER_DEPLOYMENT,
namespace: NAMESPACE,
body: { spec: { replicas } },
});
console.log(`[K8s] Scaled ${WORKER_DEPLOYMENT} to ${replicas} replicas`);
res.json({
success: true,
replicas,
message: `Scaled to ${replicas} workers`,
});
} catch (e: any) {
console.error('[K8s] Error scaling deployment:', e.message);
res.status(500).json({
success: false,
error: e.message,
});
}
});
export default router;

View File

@@ -291,107 +291,6 @@ router.get('/stores/:id/summary', async (req: Request, res: Response) => {
}
});
/**
* GET /api/markets/stores/:id/crawl-history
* Get crawl history for a specific store
*/
router.get('/stores/:id/crawl-history', async (req: Request, res: Response) => {
try {
const { id } = req.params;
const { limit = '50' } = req.query;
const dispensaryId = parseInt(id, 10);
const limitNum = Math.min(parseInt(limit as string, 10), 100);
// Get crawl history from crawl_orchestration_traces
const { rows: historyRows } = await pool.query(`
SELECT
id,
run_id,
profile_key,
crawler_module,
state_at_start,
state_at_end,
total_steps,
duration_ms,
success,
error_message,
products_found,
started_at,
completed_at
FROM crawl_orchestration_traces
WHERE dispensary_id = $1
ORDER BY started_at DESC
LIMIT $2
`, [dispensaryId, limitNum]);
// Get next scheduled crawl if available
const { rows: scheduleRows } = await pool.query(`
SELECT
js.id as schedule_id,
js.job_name,
js.enabled,
js.base_interval_minutes,
js.jitter_minutes,
js.next_run_at,
js.last_run_at,
js.last_status
FROM job_schedules js
WHERE js.enabled = true
AND js.job_config->>'dispensaryId' = $1::text
ORDER BY js.next_run_at
LIMIT 1
`, [dispensaryId.toString()]);
// Get dispensary info for slug
const { rows: dispRows } = await pool.query(`
SELECT
id,
name,
dba_name,
slug,
state,
city,
menu_type,
platform_dispensary_id,
last_menu_scrape
FROM dispensaries
WHERE id = $1
`, [dispensaryId]);
res.json({
dispensary: dispRows[0] || null,
history: historyRows.map(row => ({
id: row.id,
runId: row.run_id,
profileKey: row.profile_key,
crawlerModule: row.crawler_module,
stateAtStart: row.state_at_start,
stateAtEnd: row.state_at_end,
totalSteps: row.total_steps,
durationMs: row.duration_ms,
success: row.success,
errorMessage: row.error_message,
productsFound: row.products_found,
startedAt: row.started_at?.toISOString() || null,
completedAt: row.completed_at?.toISOString() || null,
})),
nextSchedule: scheduleRows[0] ? {
scheduleId: scheduleRows[0].schedule_id,
jobName: scheduleRows[0].job_name,
enabled: scheduleRows[0].enabled,
baseIntervalMinutes: scheduleRows[0].base_interval_minutes,
jitterMinutes: scheduleRows[0].jitter_minutes,
nextRunAt: scheduleRows[0].next_run_at?.toISOString() || null,
lastRunAt: scheduleRows[0].last_run_at?.toISOString() || null,
lastStatus: scheduleRows[0].last_status,
} : null,
});
} catch (error: any) {
console.error('[Markets] Error fetching crawl history:', error.message);
res.status(500).json({ error: error.message });
}
});
/**
* GET /api/markets/stores/:id/products
* Get products for a store with filtering and pagination

View File

@@ -78,14 +78,14 @@ router.get('/metrics', async (_req: Request, res: Response) => {
/**
* GET /api/admin/orchestrator/states
* Returns array of states with at least one crawl-enabled dispensary
* Returns array of states with at least one known dispensary
*/
router.get('/states', async (_req: Request, res: Response) => {
try {
const { rows } = await pool.query(`
SELECT DISTINCT state, COUNT(*) as store_count
FROM dispensaries
WHERE state IS NOT NULL AND crawl_enabled = true
WHERE state IS NOT NULL
GROUP BY state
ORDER BY state
`);

View File

@@ -1,334 +0,0 @@
/**
* Payload Routes
*
* Per TASK_WORKFLOW_2024-12-10.md: API access to raw crawl payloads.
*
* Endpoints:
* - GET /api/payloads - List payload metadata (paginated)
* - GET /api/payloads/:id - Get payload metadata by ID
* - GET /api/payloads/:id/data - Get full payload JSON
* - GET /api/payloads/store/:dispensaryId - List payloads for a store
* - GET /api/payloads/store/:dispensaryId/latest - Get latest payload for a store
* - GET /api/payloads/store/:dispensaryId/diff - Diff two payloads
*/
import { Router, Request, Response } from 'express';
import { getPool } from '../db/pool';
import {
loadRawPayloadById,
getLatestPayload,
getRecentPayloads,
listPayloadMetadata,
} from '../utils/payload-storage';
import { Pool } from 'pg';
const router = Router();
// Get pool instance for queries
const getDbPool = (): Pool => getPool() as unknown as Pool;
/**
* GET /api/payloads
* List payload metadata (paginated)
*/
router.get('/', async (req: Request, res: Response) => {
try {
const pool = getDbPool();
const limit = Math.min(parseInt(req.query.limit as string) || 50, 100);
const offset = parseInt(req.query.offset as string) || 0;
const dispensaryId = req.query.dispensary_id ? parseInt(req.query.dispensary_id as string) : undefined;
const payloads = await listPayloadMetadata(pool, {
dispensaryId,
limit,
offset,
});
res.json({
success: true,
payloads,
pagination: { limit, offset },
});
} catch (error: any) {
console.error('[Payloads] List error:', error.message);
res.status(500).json({ success: false, error: error.message });
}
});
/**
* GET /api/payloads/:id
* Get payload metadata by ID
*/
router.get('/:id', async (req: Request, res: Response) => {
try {
const pool = getDbPool();
const id = parseInt(req.params.id);
const result = await pool.query(`
SELECT
p.id,
p.dispensary_id,
p.crawl_run_id,
p.storage_path,
p.product_count,
p.size_bytes,
p.size_bytes_raw,
p.fetched_at,
p.processed_at,
p.checksum_sha256,
d.name as dispensary_name
FROM raw_crawl_payloads p
LEFT JOIN dispensaries d ON d.id = p.dispensary_id
WHERE p.id = $1
`, [id]);
if (result.rows.length === 0) {
return res.status(404).json({ success: false, error: 'Payload not found' });
}
res.json({
success: true,
payload: result.rows[0],
});
} catch (error: any) {
console.error('[Payloads] Get error:', error.message);
res.status(500).json({ success: false, error: error.message });
}
});
/**
* GET /api/payloads/:id/data
* Get full payload JSON (decompressed from disk)
*/
router.get('/:id/data', async (req: Request, res: Response) => {
try {
const pool = getDbPool();
const id = parseInt(req.params.id);
const result = await loadRawPayloadById(pool, id);
if (!result) {
return res.status(404).json({ success: false, error: 'Payload not found' });
}
res.json({
success: true,
metadata: result.metadata,
data: result.payload,
});
} catch (error: any) {
console.error('[Payloads] Get data error:', error.message);
res.status(500).json({ success: false, error: error.message });
}
});
/**
* GET /api/payloads/store/:dispensaryId
* List payloads for a specific store
*/
router.get('/store/:dispensaryId', async (req: Request, res: Response) => {
try {
const pool = getDbPool();
const dispensaryId = parseInt(req.params.dispensaryId);
const limit = Math.min(parseInt(req.query.limit as string) || 20, 100);
const offset = parseInt(req.query.offset as string) || 0;
const payloads = await listPayloadMetadata(pool, {
dispensaryId,
limit,
offset,
});
res.json({
success: true,
dispensaryId,
payloads,
pagination: { limit, offset },
});
} catch (error: any) {
console.error('[Payloads] Store list error:', error.message);
res.status(500).json({ success: false, error: error.message });
}
});
/**
* GET /api/payloads/store/:dispensaryId/latest
* Get the latest payload for a store (with full data)
*/
router.get('/store/:dispensaryId/latest', async (req: Request, res: Response) => {
try {
const pool = getDbPool();
const dispensaryId = parseInt(req.params.dispensaryId);
const result = await getLatestPayload(pool, dispensaryId);
if (!result) {
return res.status(404).json({
success: false,
error: `No payloads found for dispensary ${dispensaryId}`,
});
}
res.json({
success: true,
metadata: result.metadata,
data: result.payload,
});
} catch (error: any) {
console.error('[Payloads] Latest error:', error.message);
res.status(500).json({ success: false, error: error.message });
}
});
/**
* GET /api/payloads/store/:dispensaryId/diff
* Compare two payloads for a store
*
* Query params:
* - from: payload ID (older)
* - to: payload ID (newer) - optional, defaults to latest
*/
router.get('/store/:dispensaryId/diff', async (req: Request, res: Response) => {
try {
const pool = getDbPool();
const dispensaryId = parseInt(req.params.dispensaryId);
const fromId = req.query.from ? parseInt(req.query.from as string) : undefined;
const toId = req.query.to ? parseInt(req.query.to as string) : undefined;
let fromPayload: any;
let toPayload: any;
if (fromId && toId) {
// Load specific payloads
const [from, to] = await Promise.all([
loadRawPayloadById(pool, fromId),
loadRawPayloadById(pool, toId),
]);
fromPayload = from;
toPayload = to;
} else {
// Load two most recent
const recent = await getRecentPayloads(pool, dispensaryId, 2);
if (recent.length < 2) {
return res.status(400).json({
success: false,
error: 'Need at least 2 payloads to diff. Only found ' + recent.length,
});
}
toPayload = recent[0]; // Most recent
fromPayload = recent[1]; // Previous
}
if (!fromPayload || !toPayload) {
return res.status(404).json({ success: false, error: 'One or both payloads not found' });
}
// Build product maps by ID
const fromProducts = new Map<string, any>();
const toProducts = new Map<string, any>();
for (const p of fromPayload.payload.products || []) {
const id = p._id || p.id;
if (id) fromProducts.set(id, p);
}
for (const p of toPayload.payload.products || []) {
const id = p._id || p.id;
if (id) toProducts.set(id, p);
}
// Find differences
const added: any[] = [];
const removed: any[] = [];
const priceChanges: any[] = [];
const stockChanges: any[] = [];
// Products in "to" but not in "from" = added
for (const [id, product] of toProducts) {
if (!fromProducts.has(id)) {
added.push({
id,
name: product.name,
brand: product.brand?.name,
price: product.Prices?.[0]?.price,
});
}
}
// Products in "from" but not in "to" = removed
for (const [id, product] of fromProducts) {
if (!toProducts.has(id)) {
removed.push({
id,
name: product.name,
brand: product.brand?.name,
price: product.Prices?.[0]?.price,
});
}
}
// Products in both - check for changes
for (const [id, toProduct] of toProducts) {
const fromProduct = fromProducts.get(id);
if (!fromProduct) continue;
const fromPrice = fromProduct.Prices?.[0]?.price;
const toPrice = toProduct.Prices?.[0]?.price;
if (fromPrice !== toPrice) {
priceChanges.push({
id,
name: toProduct.name,
brand: toProduct.brand?.name,
oldPrice: fromPrice,
newPrice: toPrice,
change: toPrice && fromPrice ? toPrice - fromPrice : null,
});
}
const fromStock = fromProduct.Status || fromProduct.status;
const toStock = toProduct.Status || toProduct.status;
if (fromStock !== toStock) {
stockChanges.push({
id,
name: toProduct.name,
brand: toProduct.brand?.name,
oldStatus: fromStock,
newStatus: toStock,
});
}
}
res.json({
success: true,
from: {
id: fromPayload.metadata.id,
fetchedAt: fromPayload.metadata.fetchedAt,
productCount: fromPayload.metadata.productCount,
},
to: {
id: toPayload.metadata.id,
fetchedAt: toPayload.metadata.fetchedAt,
productCount: toPayload.metadata.productCount,
},
diff: {
added: added.length,
removed: removed.length,
priceChanges: priceChanges.length,
stockChanges: stockChanges.length,
},
details: {
added,
removed,
priceChanges,
stockChanges,
},
});
} catch (error: any) {
console.error('[Payloads] Diff error:', error.message);
res.status(500).json({ success: false, error: error.message });
}
});
export default router;

View File

@@ -278,7 +278,7 @@ router.post('/update-locations', requireRole('superadmin', 'admin'), async (req,
// Run in background
updateAllProxyLocations().catch(err => {
console.error('Location update failed:', err);
console.error('Location update failed:', err);
});
res.json({ message: 'Location update job started' });

View File

@@ -13,12 +13,6 @@ import {
TaskFilter,
} from '../tasks/task-service';
import { pool } from '../db/pool';
import {
isTaskPoolPaused,
pauseTaskPool,
resumeTaskPool,
getTaskPoolStatus,
} from '../tasks/task-pool-state';
const router = Router();
@@ -598,42 +592,4 @@ router.post('/migration/full-migrate', async (req: Request, res: Response) => {
}
});
/**
* GET /api/tasks/pool/status
* Check if task pool is paused
*/
router.get('/pool/status', async (_req: Request, res: Response) => {
const status = getTaskPoolStatus();
res.json({
success: true,
...status,
});
});
/**
* POST /api/tasks/pool/pause
* Pause the task pool - workers won't pick up new tasks
*/
router.post('/pool/pause', async (_req: Request, res: Response) => {
pauseTaskPool();
res.json({
success: true,
paused: true,
message: 'Task pool paused - workers will not pick up new tasks',
});
});
/**
* POST /api/tasks/pool/resume
* Resume the task pool - workers will pick up tasks again
*/
router.post('/pool/resume', async (_req: Request, res: Response) => {
resumeTaskPool();
res.json({
success: true,
paused: false,
message: 'Task pool resumed - workers will pick up new tasks',
});
});
export default router;

View File

@@ -70,20 +70,21 @@ router.post('/register', async (req: Request, res: Response) => {
);
if (existing.rows.length > 0) {
// Re-activate existing worker - keep existing pod_name (fantasy name), don't overwrite with K8s name
// Re-activate existing worker
const { rows } = await pool.query(`
UPDATE worker_registry
SET status = 'active',
role = $1,
hostname = $2,
ip_address = $3,
pod_name = $2,
hostname = $3,
ip_address = $4,
last_heartbeat_at = NOW(),
started_at = NOW(),
metadata = $4,
metadata = $5,
updated_at = NOW()
WHERE worker_id = $5
RETURNING id, worker_id, friendly_name, pod_name, role
`, [role, finalHostname, clientIp, metadata, finalWorkerId]);
WHERE worker_id = $6
RETURNING id, worker_id, friendly_name, role
`, [role, pod_name, finalHostname, clientIp, metadata, finalWorkerId]);
const worker = rows[0];
const roleMsg = role ? `for ${role}` : 'as role-agnostic';
@@ -104,13 +105,13 @@ router.post('/register', async (req: Request, res: Response) => {
const nameResult = await pool.query('SELECT assign_worker_name($1) as name', [finalWorkerId]);
const friendlyName = nameResult.rows[0].name;
// Register the worker - use friendlyName as pod_name (not K8s name)
// Register the worker
const { rows } = await pool.query(`
INSERT INTO worker_registry (
worker_id, friendly_name, role, pod_name, hostname, ip_address, status, metadata
) VALUES ($1, $2, $3, $4, $5, $6, 'active', $7)
RETURNING id, worker_id, friendly_name, pod_name, role
`, [finalWorkerId, friendlyName, role, friendlyName, finalHostname, clientIp, metadata]);
RETURNING id, worker_id, friendly_name, role
`, [finalWorkerId, friendlyName, role, pod_name, finalHostname, clientIp, metadata]);
const worker = rows[0];
const roleMsg = role ? `for ${role}` : 'as role-agnostic';
@@ -137,36 +138,17 @@ router.post('/register', async (req: Request, res: Response) => {
*
* Body:
* - worker_id: string (required)
* - current_task_id: number (optional) - task currently being processed (primary task)
* - current_task_ids: number[] (optional) - all tasks currently being processed (concurrent)
* - active_task_count: number (optional) - number of tasks currently running
* - max_concurrent_tasks: number (optional) - max concurrent tasks this worker can handle
* - current_task_id: number (optional) - task currently being processed
* - status: string (optional) - 'active', 'idle'
* - resources: object (optional) - memory_mb, cpu_user_ms, cpu_system_ms, etc.
*/
router.post('/heartbeat', async (req: Request, res: Response) => {
try {
const {
worker_id,
current_task_id,
current_task_ids,
active_task_count,
max_concurrent_tasks,
status = 'active',
resources
} = req.body;
const { worker_id, current_task_id, status = 'active', resources } = req.body;
if (!worker_id) {
return res.status(400).json({ success: false, error: 'worker_id is required' });
}
// Build metadata object with all the new fields
const metadata: Record<string, unknown> = {};
if (resources) Object.assign(metadata, resources);
if (current_task_ids) metadata.current_task_ids = current_task_ids;
if (active_task_count !== undefined) metadata.active_task_count = active_task_count;
if (max_concurrent_tasks !== undefined) metadata.max_concurrent_tasks = max_concurrent_tasks;
// Store resources in metadata jsonb column
const { rows } = await pool.query(`
UPDATE worker_registry
@@ -177,7 +159,7 @@ router.post('/heartbeat', async (req: Request, res: Response) => {
updated_at = NOW()
WHERE worker_id = $3
RETURNING id, friendly_name, status
`, [current_task_id || null, status, worker_id, Object.keys(metadata).length > 0 ? JSON.stringify(metadata) : null]);
`, [current_task_id || null, status, worker_id, resources ? JSON.stringify(resources) : null]);
if (rows.length === 0) {
return res.status(404).json({ success: false, error: 'Worker not found - please register first' });
@@ -348,27 +330,12 @@ router.get('/workers', async (req: Request, res: Response) => {
tasks_completed,
tasks_failed,
current_task_id,
-- Concurrent task fields from metadata
(metadata->>'current_task_ids')::jsonb as current_task_ids,
(metadata->>'active_task_count')::int as active_task_count,
(metadata->>'max_concurrent_tasks')::int as max_concurrent_tasks,
-- Decommission fields
COALESCE(decommission_requested, false) as decommission_requested,
decommission_reason,
-- Preflight fields (dual-transport verification)
curl_ip,
http_ip,
preflight_status,
preflight_at,
fingerprint_data,
-- Full metadata for resources
metadata,
EXTRACT(EPOCH FROM (NOW() - last_heartbeat_at)) as seconds_since_heartbeat,
CASE
WHEN status = 'offline' OR status = 'terminated' THEN status
WHEN last_heartbeat_at < NOW() - INTERVAL '2 minutes' THEN 'stale'
WHEN current_task_id IS NOT NULL THEN 'busy'
WHEN (metadata->>'active_task_count')::int > 0 THEN 'busy'
ELSE 'ready'
END as health_status,
created_at
@@ -705,163 +672,4 @@ router.get('/capacity', async (_req: Request, res: Response) => {
}
});
// ============================================================
// WORKER LIFECYCLE MANAGEMENT
// ============================================================
/**
* POST /api/worker-registry/workers/:workerId/decommission
* Request graceful decommission of a worker (will stop after current task)
*/
router.post('/workers/:workerId/decommission', async (req: Request, res: Response) => {
try {
const { workerId } = req.params;
const { reason, issued_by } = req.body;
// Update worker_registry to flag for decommission
const result = await pool.query(
`UPDATE worker_registry
SET decommission_requested = true,
decommission_reason = $2,
decommission_requested_at = NOW()
WHERE worker_id = $1
RETURNING friendly_name, status, current_task_id`,
[workerId, reason || 'Manual decommission from admin']
);
if (result.rows.length === 0) {
return res.status(404).json({ success: false, error: 'Worker not found' });
}
const worker = result.rows[0];
// Also log to worker_commands for audit trail
await pool.query(
`INSERT INTO worker_commands (worker_id, command, reason, issued_by)
VALUES ($1, 'decommission', $2, $3)
ON CONFLICT DO NOTHING`,
[workerId, reason || 'Manual decommission', issued_by || 'admin']
).catch(() => {
// Table might not exist yet - ignore
});
res.json({
success: true,
message: worker.current_task_id
? `Worker ${worker.friendly_name} will stop after completing task #${worker.current_task_id}`
: `Worker ${worker.friendly_name} will stop on next poll`,
worker: {
friendly_name: worker.friendly_name,
status: worker.status,
current_task_id: worker.current_task_id,
decommission_requested: true
}
});
} catch (error: any) {
res.status(500).json({ success: false, error: error.message });
}
});
/**
* POST /api/worker-registry/workers/:workerId/cancel-decommission
* Cancel a pending decommission request
*/
router.post('/workers/:workerId/cancel-decommission', async (req: Request, res: Response) => {
try {
const { workerId } = req.params;
const result = await pool.query(
`UPDATE worker_registry
SET decommission_requested = false,
decommission_reason = NULL,
decommission_requested_at = NULL
WHERE worker_id = $1
RETURNING friendly_name`,
[workerId]
);
if (result.rows.length === 0) {
return res.status(404).json({ success: false, error: 'Worker not found' });
}
res.json({
success: true,
message: `Decommission cancelled for ${result.rows[0].friendly_name}`
});
} catch (error: any) {
res.status(500).json({ success: false, error: error.message });
}
});
/**
* POST /api/worker-registry/spawn
* Spawn a new worker in the current pod (only works in multi-worker-per-pod mode)
* For now, this is a placeholder - actual spawning requires the pod supervisor
*/
router.post('/spawn', async (req: Request, res: Response) => {
try {
const { pod_name, role } = req.body;
// For now, we can't actually spawn workers from the API
// This would require a supervisor process in each pod that listens for spawn commands
// Instead, return instructions for how to scale
res.json({
success: false,
error: 'Direct worker spawning not yet implemented',
instructions: 'To add workers, scale the K8s deployment: kubectl scale deployment/scraper-worker --replicas=N'
});
} catch (error: any) {
res.status(500).json({ success: false, error: error.message });
}
});
/**
* GET /api/worker-registry/pods
* Get workers grouped by pod
*/
router.get('/pods', async (_req: Request, res: Response) => {
try {
const { rows } = await pool.query(`
SELECT
COALESCE(pod_name, 'Unknown') as pod_name,
COUNT(*) as worker_count,
COUNT(*) FILTER (WHERE current_task_id IS NOT NULL) as busy_count,
COUNT(*) FILTER (WHERE current_task_id IS NULL) as idle_count,
SUM(tasks_completed) as total_completed,
SUM(tasks_failed) as total_failed,
SUM((metadata->>'memory_rss_mb')::int) as total_memory_mb,
array_agg(json_build_object(
'worker_id', worker_id,
'friendly_name', friendly_name,
'status', status,
'current_task_id', current_task_id,
'tasks_completed', tasks_completed,
'tasks_failed', tasks_failed,
'decommission_requested', COALESCE(decommission_requested, false),
'last_heartbeat_at', last_heartbeat_at
)) as workers
FROM worker_registry
WHERE status NOT IN ('offline', 'terminated')
GROUP BY pod_name
ORDER BY pod_name
`);
res.json({
success: true,
pods: rows.map(row => ({
pod_name: row.pod_name,
worker_count: parseInt(row.worker_count),
busy_count: parseInt(row.busy_count),
idle_count: parseInt(row.idle_count),
total_completed: parseInt(row.total_completed) || 0,
total_failed: parseInt(row.total_failed) || 0,
total_memory_mb: parseInt(row.total_memory_mb) || 0,
workers: row.workers
}))
});
} catch (error: any) {
res.status(500).json({ success: false, error: error.message });
}
});
export default router;

View File

@@ -17,234 +17,13 @@
* GET /api/monitor/jobs - Get recent job history
* GET /api/monitor/active-jobs - Get currently running jobs
* GET /api/monitor/summary - Get monitoring summary
*
* K8s Scaling (added 2024-12-10):
* GET /api/workers/k8s/replicas - Get current replica count
* POST /api/workers/k8s/scale - Scale worker replicas up/down
*/
import { Router, Request, Response } from 'express';
import { pool } from '../db/pool';
import * as k8s from '@kubernetes/client-node';
const router = Router();
// ============================================================
// K8S SCALING CONFIGURATION (added 2024-12-10)
// Per TASK_WORKFLOW_2024-12-10.md: Admin can scale workers from UI
// ============================================================
const K8S_NAMESPACE = process.env.K8S_NAMESPACE || 'dispensary-scraper';
const K8S_DEPLOYMENT_NAME = process.env.K8S_WORKER_DEPLOYMENT || 'scraper-worker';
// Initialize K8s client - uses in-cluster config when running in K8s,
// or kubeconfig when running locally
let k8sAppsApi: k8s.AppsV1Api | null = null;
function getK8sClient(): k8s.AppsV1Api | null {
if (k8sAppsApi) return k8sAppsApi;
try {
const kc = new k8s.KubeConfig();
// Try in-cluster config first (when running as a pod)
// Falls back to default kubeconfig (~/.kube/config) for local dev
try {
kc.loadFromCluster();
} catch {
kc.loadFromDefault();
}
k8sAppsApi = kc.makeApiClient(k8s.AppsV1Api);
return k8sAppsApi;
} catch (err: any) {
console.warn('[Workers] K8s client not available:', err.message);
return null;
}
}
// ============================================================
// K8S SCALING ROUTES (added 2024-12-10)
// Per TASK_WORKFLOW_2024-12-10.md: Admin can scale workers from UI
// ============================================================
/**
* GET /api/workers/k8s/replicas - Get current worker replica count
* Returns current and desired replica counts from the Deployment
*/
router.get('/k8s/replicas', async (_req: Request, res: Response) => {
const client = getK8sClient();
if (!client) {
return res.status(503).json({
success: false,
error: 'K8s client not available (not running in cluster or no kubeconfig)',
replicas: null,
});
}
try {
const response = await client.readNamespacedDeployment({
name: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
});
const deployment = response;
res.json({
success: true,
replicas: {
current: deployment.status?.readyReplicas || 0,
desired: deployment.spec?.replicas || 0,
available: deployment.status?.availableReplicas || 0,
updated: deployment.status?.updatedReplicas || 0,
},
deployment: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
});
} catch (err: any) {
console.error('[Workers] K8s replicas error:', err.body?.message || err.message);
res.status(500).json({
success: false,
error: err.body?.message || err.message,
});
}
});
/**
* POST /api/workers/k8s/scale - Scale worker replicas
* Body: { replicas: number } - desired replica count (0-20)
*/
router.post('/k8s/scale', async (req: Request, res: Response) => {
const client = getK8sClient();
if (!client) {
return res.status(503).json({
success: false,
error: 'K8s client not available (not running in cluster or no kubeconfig)',
});
}
const { replicas } = req.body;
// Validate replica count
if (typeof replicas !== 'number' || replicas < 0 || replicas > 20) {
return res.status(400).json({
success: false,
error: 'replicas must be a number between 0 and 20',
});
}
try {
// Get current state first
const currentResponse = await client.readNamespacedDeploymentScale({
name: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
});
const currentReplicas = currentResponse.spec?.replicas || 0;
// Update scale using replaceNamespacedDeploymentScale
await client.replaceNamespacedDeploymentScale({
name: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
body: {
apiVersion: 'autoscaling/v1',
kind: 'Scale',
metadata: {
name: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
},
spec: {
replicas: replicas,
},
},
});
console.log(`[Workers] Scaled ${K8S_DEPLOYMENT_NAME} from ${currentReplicas} to ${replicas} replicas`);
res.json({
success: true,
message: `Scaled from ${currentReplicas} to ${replicas} replicas`,
previous: currentReplicas,
desired: replicas,
deployment: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
});
} catch (err: any) {
console.error('[Workers] K8s scale error:', err.body?.message || err.message);
res.status(500).json({
success: false,
error: err.body?.message || err.message,
});
}
});
/**
* POST /api/workers/k8s/scale-up - Scale up worker replicas by 1
* Convenience endpoint for adding a single worker
*/
router.post('/k8s/scale-up', async (_req: Request, res: Response) => {
const client = getK8sClient();
if (!client) {
return res.status(503).json({
success: false,
error: 'K8s client not available (not running in cluster or no kubeconfig)',
});
}
try {
// Get current replica count
const currentResponse = await client.readNamespacedDeploymentScale({
name: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
});
const currentReplicas = currentResponse.spec?.replicas || 0;
const newReplicas = currentReplicas + 1;
// Cap at 20 replicas
if (newReplicas > 20) {
return res.status(400).json({
success: false,
error: 'Maximum replica count (20) reached',
});
}
// Scale up by 1
await client.replaceNamespacedDeploymentScale({
name: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
body: {
apiVersion: 'autoscaling/v1',
kind: 'Scale',
metadata: {
name: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
},
spec: {
replicas: newReplicas,
},
},
});
console.log(`[Workers] Scaled up ${K8S_DEPLOYMENT_NAME} from ${currentReplicas} to ${newReplicas} replicas`);
res.json({
success: true,
message: `Added worker (${currentReplicas}${newReplicas} replicas)`,
previous: currentReplicas,
desired: newReplicas,
deployment: K8S_DEPLOYMENT_NAME,
namespace: K8S_NAMESPACE,
});
} catch (err: any) {
console.error('[Workers] K8s scale-up error:', err.body?.message || err.message);
res.status(500).json({
success: false,
error: err.body?.message || err.message,
});
}
});
// ============================================================
// STATIC ROUTES (must come before parameterized routes)
// ============================================================

View File

@@ -16,11 +16,10 @@ import {
executeGraphQL,
startSession,
endSession,
setCrawlRotator,
getFingerprint,
GRAPHQL_HASHES,
DUTCHIE_CONFIG,
} from '../platforms/dutchie';
import { CrawlRotator } from '../services/crawl-rotator';
dotenv.config();
@@ -109,27 +108,19 @@ async function main() {
// ============================================================
// STEP 2: Start stealth session
// Per workflow-12102025.md: Initialize CrawlRotator and start session with menuUrl
// ============================================================
console.log('┌─────────────────────────────────────────────────────────────┐');
console.log('│ STEP 2: Start Stealth Session │');
console.log('└─────────────────────────────────────────────────────────────┘');
// Per workflow-12102025.md: Initialize CrawlRotator (required for sessions)
const rotator = new CrawlRotator();
setCrawlRotator(rotator);
// Use Arizona timezone for this store
const session = startSession(disp.state || 'AZ', 'America/Phoenix');
// Per workflow-12102025.md: startSession takes menuUrl for dynamic Referer
const session = startSession(disp.menu_url);
const fp = session.fingerprint;
const fp = getFingerprint();
console.log(` Session ID: ${session.sessionId}`);
console.log(` Browser: ${fp.browserName} (${fp.deviceCategory})`);
console.log(` User-Agent: ${fp.userAgent.slice(0, 60)}...`);
console.log(` Accept-Language: ${fp.acceptLanguage}`);
console.log(` Referer: ${session.referer}`);
console.log(` DNT: ${fp.httpFingerprint.hasDNT ? 'enabled' : 'disabled'}`);
console.log(` TLS: ${fp.httpFingerprint.curlImpersonateBinary}`);
console.log(` Sec-CH-UA: ${fp.secChUa || '(not set)'}`);
console.log('');
// ============================================================

View File

@@ -1,10 +1,10 @@
/**
* Test script for stealth session management
*
* Per workflow-12102025.md:
* - Tests HTTP fingerprinting (browser-specific headers + ordering)
* - Tests UA generation (device distribution, browser filtering)
* - Tests dynamic Referer per dispensary
* Tests:
* 1. Per-session fingerprint rotation
* 2. Geographic consistency (timezone → Accept-Language)
* 3. Proxy location loading from database
*
* Usage:
* npx tsx src/scripts/test-stealth-session.ts
@@ -14,142 +14,104 @@ import {
startSession,
endSession,
getCurrentSession,
getFingerprint,
getRandomFingerprint,
getLocaleForTimezone,
buildHeaders,
setCrawlRotator,
} from '../platforms/dutchie';
import { CrawlRotator } from '../services/crawl-rotator';
import {
generateHTTPFingerprint,
buildRefererFromMenuUrl,
BrowserType,
} from '../services/http-fingerprint';
console.log('='.repeat(60));
console.log('STEALTH SESSION TEST (per workflow-12102025.md)');
console.log('STEALTH SESSION TEST');
console.log('='.repeat(60));
// Initialize CrawlRotator (required for sessions)
console.log('\n[Setup] Initializing CrawlRotator...');
const rotator = new CrawlRotator();
setCrawlRotator(rotator);
console.log(' CrawlRotator initialized');
// Test 1: HTTP Fingerprint Generation
console.log('\n[Test 1] HTTP Fingerprint Generation:');
const browsers: BrowserType[] = ['Chrome', 'Firefox', 'Safari', 'Edge'];
for (const browser of browsers) {
const httpFp = generateHTTPFingerprint(browser);
console.log(` ${browser}:`);
console.log(` TLS binary: ${httpFp.curlImpersonateBinary}`);
console.log(` DNT: ${httpFp.hasDNT ? 'enabled' : 'disabled'}`);
console.log(` Header order: ${httpFp.headerOrder.slice(0, 5).join(', ')}...`);
}
// Test 2: Dynamic Referer from menu URLs
console.log('\n[Test 2] Dynamic Referer from Menu URLs:');
const testUrls = [
'https://dutchie.com/embedded-menu/harvest-of-tempe',
'https://dutchie.com/dispensary/zen-leaf-mesa',
'/embedded-menu/deeply-rooted',
'/dispensary/curaleaf-phoenix',
null,
// Test 1: Timezone to Locale mapping
console.log('\n[Test 1] Timezone to Locale Mapping:');
const testTimezones = [
'America/Phoenix',
'America/Los_Angeles',
'America/New_York',
'America/Chicago',
undefined,
'Invalid/Timezone',
];
for (const url of testUrls) {
const referer = buildRefererFromMenuUrl(url);
console.log(` ${url || '(null/undefined)'}`);
console.log(`${referer}`);
for (const tz of testTimezones) {
const locale = getLocaleForTimezone(tz);
console.log(` ${tz || '(undefined)'}${locale}`);
}
// Test 3: Session with Dynamic Referer
console.log('\n[Test 3] Session with Dynamic Referer:');
const testMenuUrl = 'https://dutchie.com/dispensary/harvest-of-tempe';
console.log(` Starting session with menuUrl: ${testMenuUrl}`);
// Test 2: Random fingerprint selection
console.log('\n[Test 2] Random Fingerprint Selection (5 samples):');
for (let i = 0; i < 5; i++) {
const fp = getRandomFingerprint();
console.log(` ${i + 1}. ${fp.userAgent.slice(0, 60)}...`);
}
const session1 = startSession(testMenuUrl);
// Test 3: Session Management
console.log('\n[Test 3] Session Management:');
// Before session - should use default fingerprint
console.log(' Before session:');
const beforeFp = getFingerprint();
console.log(` getFingerprint(): ${beforeFp.userAgent.slice(0, 50)}...`);
console.log(` getCurrentSession(): ${getCurrentSession()}`);
// Start session with Arizona timezone
console.log('\n Starting session (AZ, America/Phoenix):');
const session1 = startSession('AZ', 'America/Phoenix');
console.log(` Session ID: ${session1.sessionId}`);
console.log(` Browser: ${session1.fingerprint.browserName}`);
console.log(` Device: ${session1.fingerprint.deviceCategory}`);
console.log(` Referer: ${session1.referer}`);
console.log(` DNT: ${session1.fingerprint.httpFingerprint.hasDNT ? 'enabled' : 'disabled'}`);
console.log(` TLS: ${session1.fingerprint.httpFingerprint.curlImpersonateBinary}`);
console.log(` Fingerprint UA: ${session1.fingerprint.userAgent.slice(0, 50)}...`);
console.log(` Accept-Language: ${session1.fingerprint.acceptLanguage}`);
console.log(` Timezone: ${session1.timezone}`);
// Test 4: Build Headers (browser-specific order)
console.log('\n[Test 4] Build Headers (browser-specific order):');
const { headers, orderedHeaders } = buildHeaders(true, 1000);
console.log(` Headers built for ${session1.fingerprint.browserName}:`);
console.log(` Order: ${orderedHeaders.join(' → ')}`);
console.log(` Sample headers:`);
console.log(` User-Agent: ${headers['User-Agent']?.slice(0, 50)}...`);
console.log(` Accept: ${headers['Accept']}`);
console.log(` Accept-Language: ${headers['Accept-Language']}`);
console.log(` Referer: ${headers['Referer']}`);
if (headers['sec-ch-ua']) {
console.log(` sec-ch-ua: ${headers['sec-ch-ua']}`);
}
if (headers['DNT']) {
console.log(` DNT: ${headers['DNT']}`);
}
// During session - should use session fingerprint
console.log('\n During session:');
const duringFp = getFingerprint();
console.log(` getFingerprint(): ${duringFp.userAgent.slice(0, 50)}...`);
console.log(` Same as session? ${duringFp.userAgent === session1.fingerprint.userAgent}`);
// Test buildHeaders with session
console.log('\n buildHeaders() during session:');
const headers = buildHeaders('/embedded-menu/test-store');
console.log(` User-Agent: ${headers['user-agent'].slice(0, 50)}...`);
console.log(` Accept-Language: ${headers['accept-language']}`);
console.log(` Origin: ${headers['origin']}`);
console.log(` Referer: ${headers['referer']}`);
// End session
console.log('\n Ending session:');
endSession();
console.log(` getCurrentSession(): ${getCurrentSession()}`);
// Test 5: Multiple Sessions (UA variety)
console.log('\n[Test 5] Multiple Sessions (UA & fingerprint variety):');
const sessions: {
browser: string;
device: string;
hasDNT: boolean;
}[] = [];
// Test 4: Multiple sessions should have different fingerprints
console.log('\n[Test 4] Multiple Sessions (fingerprint variety):');
const fingerprints: string[] = [];
for (let i = 0; i < 10; i++) {
const session = startSession(`/dispensary/store-${i}`);
sessions.push({
browser: session.fingerprint.browserName,
device: session.fingerprint.deviceCategory,
hasDNT: session.fingerprint.httpFingerprint.hasDNT,
});
const session = startSession('CA', 'America/Los_Angeles');
fingerprints.push(session.fingerprint.userAgent);
endSession();
}
// Count distribution
const browserCounts: Record<string, number> = {};
const deviceCounts: Record<string, number> = {};
let dntCount = 0;
const uniqueCount = new Set(fingerprints).size;
console.log(` 10 sessions created, ${uniqueCount} unique fingerprints`);
console.log(` Variety: ${uniqueCount >= 3 ? '✅ Good' : '⚠️ Low - may need more fingerprint options'}`);
for (const s of sessions) {
browserCounts[s.browser] = (browserCounts[s.browser] || 0) + 1;
deviceCounts[s.device] = (deviceCounts[s.device] || 0) + 1;
if (s.hasDNT) dntCount++;
}
// Test 5: Geographic consistency check
console.log('\n[Test 5] Geographic Consistency:');
const geoTests = [
{ state: 'AZ', tz: 'America/Phoenix' },
{ state: 'CA', tz: 'America/Los_Angeles' },
{ state: 'NY', tz: 'America/New_York' },
{ state: 'IL', tz: 'America/Chicago' },
];
console.log(` 10 sessions created:`);
console.log(` Browsers: ${JSON.stringify(browserCounts)}`);
console.log(` Devices: ${JSON.stringify(deviceCounts)}`);
console.log(` DNT enabled: ${dntCount}/10 (expected ~30%)`);
// Test 6: Device distribution check (per workflow-12102025.md: 62/36/2)
console.log('\n[Test 6] Device Distribution (larger sample):');
const deviceSamples: string[] = [];
for (let i = 0; i < 100; i++) {
const session = startSession();
deviceSamples.push(session.fingerprint.deviceCategory);
for (const { state, tz } of geoTests) {
const session = startSession(state, tz);
const consistent = session.fingerprint.acceptLanguage.includes('en-US');
console.log(` ${state} (${tz}): Accept-Language=${session.fingerprint.acceptLanguage} ${consistent ? '✅' : '❌'}`);
endSession();
}
const mobileCount = deviceSamples.filter(d => d === 'mobile').length;
const desktopCount = deviceSamples.filter(d => d === 'desktop').length;
const tabletCount = deviceSamples.filter(d => d === 'tablet').length;
console.log(` 100 sessions (expected: 62% mobile, 36% desktop, 2% tablet):`);
console.log(` Mobile: ${mobileCount}%`);
console.log(` Desktop: ${desktopCount}%`);
console.log(` Tablet: ${tabletCount}%`);
console.log(` Distribution: ${Math.abs(mobileCount - 62) < 15 && Math.abs(desktopCount - 36) < 15 ? '✅ Reasonable' : '⚠️ Off target'}`);
console.log('\n' + '='.repeat(60));
console.log('TEST COMPLETE');
console.log('='.repeat(60));

View File

@@ -1,53 +1,49 @@
/**
* Crawl Rotator - Proxy & User Agent Rotation for Crawlers
*
* Updated: 2025-12-10 per workflow-12102025.md
*
* KEY BEHAVIORS (per workflow-12102025.md):
* 1. Task determines WHAT work to do, proxy determines SESSION IDENTITY
* 2. Proxy location (timezone) sets Accept-Language headers (always English)
* 3. On 403: immediately get new IP, new fingerprint, retry
* 4. After 3 consecutive 403s on same proxy with different fingerprints → disable proxy
*
* USER-AGENT GENERATION (per workflow-12102025.md):
* - Device distribution: Mobile 62%, Desktop 36%, Tablet 2%
* - Browser whitelist: Chrome, Safari, Edge, Firefox only
* - UA sticks until IP rotates
* - Failure = alert admin + stop crawl (no fallback)
*
* Uses intoli/user-agents for realistic UA generation with daily-updated data.
* Manages rotation of proxies and user agents to avoid blocks.
* Used by platform-specific crawlers (Dutchie, Jane, etc.)
*
* Canonical location: src/services/crawl-rotator.ts
*/
import { Pool } from 'pg';
import UserAgent from 'user-agents';
import {
HTTPFingerprint,
generateHTTPFingerprint,
BrowserType,
} from './http-fingerprint';
// ============================================================
// UA CONSTANTS (per workflow-12102025.md)
// USER AGENT CONFIGURATION
// ============================================================
/**
* Per workflow-12102025.md: Device category distribution (hardcoded)
* Mobile: 62%, Desktop: 36%, Tablet: 2%
* Modern browser user agents (Chrome, Firefox, Safari, Edge on various platforms)
* Updated: 2024
*/
const DEVICE_WEIGHTS = {
mobile: 62,
desktop: 36,
tablet: 2,
} as const;
export const USER_AGENTS = [
// Chrome on Windows
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36',
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36',
/**
* Per workflow-12102025.md: Browser whitelist
* Only Chrome (67%), Safari (20%), Edge (6%), Firefox (3%)
* Samsung Internet, Opera, and other niche browsers are filtered out
*/
const ALLOWED_BROWSERS = ['Chrome', 'Safari', 'Edge', 'Firefox'] as const;
// Chrome on macOS
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36',
// Firefox on Windows
'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:121.0) Gecko/20100101 Firefox/121.0',
'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:120.0) Gecko/20100101 Firefox/120.0',
// Firefox on macOS
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:121.0) Gecko/20100101 Firefox/121.0',
// Safari on macOS
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.2 Safari/605.1.15',
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.1 Safari/605.1.15',
// Edge on Windows
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36 Edg/120.0.0.0',
// Chrome on Linux
'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
];
// ============================================================
// PROXY TYPES
@@ -65,13 +61,8 @@ export interface Proxy {
failureCount: number;
successCount: number;
avgResponseTimeMs: number | null;
maxConnections: number;
/**
* Per workflow-12102025.md: Track consecutive 403s with different fingerprints.
* After 3 consecutive 403s → disable proxy (it's burned).
*/
consecutive403Count: number;
// Location info - determines session headers per workflow-12102025.md
maxConnections: number; // Number of concurrent connections allowed (for rotating proxies)
// Location info (if known)
city?: string;
state?: string;
country?: string;
@@ -86,40 +77,6 @@ export interface ProxyStats {
avgSuccessRate: number;
}
// ============================================================
// FINGERPRINT TYPE
// Per workflow-12102025.md: Full browser fingerprint from user-agents
// ============================================================
export interface BrowserFingerprint {
userAgent: string;
platform: string;
screenWidth: number;
screenHeight: number;
viewportWidth: number;
viewportHeight: number;
deviceCategory: string;
browserName: string; // Per workflow-12102025.md: for session logging
// Derived headers for anti-detect
acceptLanguage: string;
secChUa?: string;
secChUaPlatform?: string;
secChUaMobile?: string;
// Per workflow-12102025.md: HTTP Fingerprinting section
httpFingerprint: HTTPFingerprint;
}
/**
* Per workflow-12102025.md: Session log entry for debugging blocked sessions
*/
export interface UASessionLog {
deviceCategory: string;
browserName: string;
userAgent: string;
proxyIp: string | null;
sessionStartedAt: Date;
}
// ============================================================
// PROXY ROTATOR CLASS
// ============================================================
@@ -134,6 +91,9 @@ export class ProxyRotator {
this.pool = pool || null;
}
/**
* Initialize with database pool
*/
setPool(pool: Pool): void {
this.pool = pool;
}
@@ -162,7 +122,6 @@ export class ProxyRotator {
0 as "successCount",
response_time_ms as "avgResponseTimeMs",
COALESCE(max_connections, 1) as "maxConnections",
COALESCE(consecutive_403_count, 0) as "consecutive403Count",
city,
state,
country,
@@ -175,9 +134,11 @@ export class ProxyRotator {
this.proxies = result.rows;
// Calculate total concurrent capacity
const totalCapacity = this.proxies.reduce((sum, p) => sum + p.maxConnections, 0);
console.log(`[ProxyRotator] Loaded ${this.proxies.length} active proxies (${totalCapacity} max concurrent connections)`);
} catch (error) {
// Table might not exist - that's okay
console.warn(`[ProxyRotator] Could not load proxies: ${error}`);
this.proxies = [];
}
@@ -189,6 +150,7 @@ export class ProxyRotator {
getNext(): Proxy | null {
if (this.proxies.length === 0) return null;
// Round-robin rotation
this.currentIndex = (this.currentIndex + 1) % this.proxies.length;
this.lastRotation = new Date();
@@ -223,68 +185,23 @@ export class ProxyRotator {
}
/**
* Mark proxy as blocked (403 received)
* Per workflow-12102025.md:
* - Increment consecutive_403_count
* - After 3 consecutive 403s with different fingerprints → disable proxy
* - This is separate from general failures (timeouts, etc.)
*/
async markBlocked(proxyId: number): Promise<boolean> {
const proxy = this.proxies.find(p => p.id === proxyId);
let shouldDisable = false;
if (proxy) {
proxy.consecutive403Count++;
// Per workflow-12102025.md: 3 consecutive 403s → proxy is burned
if (proxy.consecutive403Count >= 3) {
proxy.isActive = false;
this.proxies = this.proxies.filter(p => p.id !== proxyId);
console.log(`[ProxyRotator] Proxy ${proxyId} DISABLED after ${proxy.consecutive403Count} consecutive 403s (burned)`);
shouldDisable = true;
} else {
console.log(`[ProxyRotator] Proxy ${proxyId} blocked (403 #${proxy.consecutive403Count}/3)`);
}
}
// Update database
if (this.pool) {
try {
await this.pool.query(`
UPDATE proxies
SET
consecutive_403_count = COALESCE(consecutive_403_count, 0) + 1,
last_failure_at = NOW(),
test_result = '403 Forbidden',
active = CASE WHEN COALESCE(consecutive_403_count, 0) >= 2 THEN false ELSE active END,
updated_at = NOW()
WHERE id = $1
`, [proxyId]);
} catch (err) {
console.error(`[ProxyRotator] Failed to update proxy ${proxyId}:`, err);
}
}
return shouldDisable;
}
/**
* Mark proxy as failed (general error - timeout, connection error, etc.)
* Separate from 403 blocking per workflow-12102025.md
* Mark proxy as failed (temporarily remove from rotation)
*/
async markFailed(proxyId: number, error?: string): Promise<void> {
// Update in-memory
const proxy = this.proxies.find(p => p.id === proxyId);
if (proxy) {
proxy.failureCount++;
// Deactivate if too many general failures
// Deactivate if too many failures
if (proxy.failureCount >= 5) {
proxy.isActive = false;
this.proxies = this.proxies.filter(p => p.id !== proxyId);
console.log(`[ProxyRotator] Proxy ${proxyId} deactivated after ${proxy.failureCount} general failures`);
console.log(`[ProxyRotator] Proxy ${proxyId} deactivated after ${proxy.failureCount} failures`);
}
}
// Update database
if (this.pool) {
try {
await this.pool.query(`
@@ -303,22 +220,23 @@ export class ProxyRotator {
}
/**
* Mark proxy as successful - resets consecutive 403 count
* Per workflow-12102025.md: successful request clears the 403 counter
* Mark proxy as successful
*/
async markSuccess(proxyId: number, responseTimeMs?: number): Promise<void> {
// Update in-memory
const proxy = this.proxies.find(p => p.id === proxyId);
if (proxy) {
proxy.successCount++;
proxy.consecutive403Count = 0; // Reset on success per workflow-12102025.md
proxy.lastUsedAt = new Date();
if (responseTimeMs !== undefined) {
// Rolling average
proxy.avgResponseTimeMs = proxy.avgResponseTimeMs
? (proxy.avgResponseTimeMs * 0.8) + (responseTimeMs * 0.2)
: responseTimeMs;
}
}
// Update database
if (this.pool) {
try {
await this.pool.query(`
@@ -326,7 +244,6 @@ export class ProxyRotator {
SET
last_tested_at = NOW(),
test_result = 'success',
consecutive_403_count = 0,
response_time_ms = CASE
WHEN response_time_ms IS NULL THEN $2
ELSE (response_time_ms * 0.8 + $2 * 0.2)::integer
@@ -355,8 +272,8 @@ export class ProxyRotator {
*/
getStats(): ProxyStats {
const totalProxies = this.proxies.length;
const activeProxies = this.proxies.reduce((sum, p) => sum + p.maxConnections, 0);
const blockedProxies = this.proxies.filter(p => p.failureCount >= 5 || p.consecutive403Count >= 3).length;
const activeProxies = this.proxies.reduce((sum, p) => sum + p.maxConnections, 0); // Total concurrent capacity
const blockedProxies = this.proxies.filter(p => p.failureCount >= 5).length;
const successRates = this.proxies
.filter(p => p.successCount + p.failureCount > 0)
@@ -368,12 +285,15 @@ export class ProxyRotator {
return {
totalProxies,
activeProxies,
activeProxies, // Total concurrent capacity across all proxies
blockedProxies,
avgSuccessRate,
};
}
/**
* Check if proxy pool has available proxies
*/
hasAvailableProxies(): boolean {
return this.proxies.length > 0;
}
@@ -381,194 +301,53 @@ export class ProxyRotator {
// ============================================================
// USER AGENT ROTATOR CLASS
// Per workflow-12102025.md: Uses intoli/user-agents for realistic fingerprints
// ============================================================
export class UserAgentRotator {
private currentFingerprint: BrowserFingerprint | null = null;
private sessionLog: UASessionLog | null = null;
private userAgents: string[];
private currentIndex: number = 0;
private lastRotation: Date = new Date();
constructor() {
// Per workflow-12102025.md: Initialize with first fingerprint
this.rotate();
constructor(userAgents: string[] = USER_AGENTS) {
this.userAgents = userAgents;
// Start at random index to avoid patterns
this.currentIndex = Math.floor(Math.random() * userAgents.length);
}
/**
* Per workflow-12102025.md: Roll device category based on distribution
* Mobile: 62%, Desktop: 36%, Tablet: 2%
* Get next user agent in rotation
*/
private rollDeviceCategory(): 'mobile' | 'desktop' | 'tablet' {
const roll = Math.random() * 100;
if (roll < DEVICE_WEIGHTS.mobile) {
return 'mobile';
} else if (roll < DEVICE_WEIGHTS.mobile + DEVICE_WEIGHTS.desktop) {
return 'desktop';
} else {
return 'tablet';
}
getNext(): string {
this.currentIndex = (this.currentIndex + 1) % this.userAgents.length;
this.lastRotation = new Date();
return this.userAgents[this.currentIndex];
}
/**
* Per workflow-12102025.md: Extract browser name from UA string
* Get current user agent without rotating
*/
private extractBrowserName(userAgent: string): string {
if (userAgent.includes('Edg/')) return 'Edge';
if (userAgent.includes('Firefox/')) return 'Firefox';
if (userAgent.includes('Safari/') && !userAgent.includes('Chrome/')) return 'Safari';
if (userAgent.includes('Chrome/')) return 'Chrome';
return 'Unknown';
getCurrent(): string {
return this.userAgents[this.currentIndex];
}
/**
* Per workflow-12102025.md: Check if browser is in whitelist
* Get a random user agent
*/
private isAllowedBrowser(userAgent: string): boolean {
const browserName = this.extractBrowserName(userAgent);
return ALLOWED_BROWSERS.includes(browserName as typeof ALLOWED_BROWSERS[number]);
getRandom(): string {
const index = Math.floor(Math.random() * this.userAgents.length);
return this.userAgents[index];
}
/**
* Generate a new random fingerprint
* Per workflow-12102025.md:
* - Roll device category (62/36/2)
* - Filter to top 4 browsers only
* - Failure = alert admin + stop (no fallback)
* Get total available user agents
*/
rotate(proxyIp?: string): BrowserFingerprint {
// Per workflow-12102025.md: Roll device category
const deviceCategory = this.rollDeviceCategory();
// Per workflow-12102025.md: Generate UA filtered to device category
const generator = new UserAgent({ deviceCategory });
// Per workflow-12102025.md: Try to get an allowed browser (max 50 attempts)
let ua: ReturnType<typeof generator>;
let attempts = 0;
const maxAttempts = 50;
do {
ua = generator();
attempts++;
} while (!this.isAllowedBrowser(ua.data.userAgent) && attempts < maxAttempts);
// Per workflow-12102025.md: If we can't get allowed browser, this is a failure
if (!this.isAllowedBrowser(ua.data.userAgent)) {
const errorMsg = `[UserAgentRotator] CRITICAL: Failed to generate allowed browser after ${maxAttempts} attempts. Device: ${deviceCategory}. Last UA: ${ua.data.userAgent}`;
console.error(errorMsg);
// Per workflow-12102025.md: Alert admin + stop crawl
// TODO: Post alert to admin dashboard
throw new Error(errorMsg);
}
const data = ua.data;
const browserName = this.extractBrowserName(data.userAgent);
// Build sec-ch-ua headers from user agent string
const secChUa = this.buildSecChUa(data.userAgent, deviceCategory);
// Per workflow-12102025.md: HTTP Fingerprinting - generate full HTTP fingerprint
const httpFingerprint = generateHTTPFingerprint(browserName as BrowserType);
this.currentFingerprint = {
userAgent: data.userAgent,
platform: data.platform,
screenWidth: data.screenWidth,
screenHeight: data.screenHeight,
viewportWidth: data.viewportWidth,
viewportHeight: data.viewportHeight,
deviceCategory: data.deviceCategory,
browserName, // Per workflow-12102025.md: for session logging
// Per workflow-12102025.md: always English
acceptLanguage: 'en-US,en;q=0.9',
...secChUa,
// Per workflow-12102025.md: HTTP Fingerprinting section
httpFingerprint,
};
// Per workflow-12102025.md: Log session data
this.sessionLog = {
deviceCategory,
browserName,
userAgent: data.userAgent,
proxyIp: proxyIp || null,
sessionStartedAt: new Date(),
};
console.log(`[UserAgentRotator] New fingerprint: device=${deviceCategory}, browser=${browserName}, UA=${data.userAgent.slice(0, 50)}...`);
return this.currentFingerprint;
}
/**
* Get current fingerprint without rotating
*/
getCurrent(): BrowserFingerprint {
if (!this.currentFingerprint) {
return this.rotate();
}
return this.currentFingerprint;
}
/**
* Get a random fingerprint (rotates and returns)
*/
getRandom(proxyIp?: string): BrowserFingerprint {
return this.rotate(proxyIp);
}
/**
* Per workflow-12102025.md: Get session log for debugging
*/
getSessionLog(): UASessionLog | null {
return this.sessionLog;
}
/**
* Build sec-ch-ua headers from user agent string
* Per workflow-12102025.md: Include mobile indicator based on device category
*/
private buildSecChUa(userAgent: string, deviceCategory: string): { secChUa?: string; secChUaPlatform?: string; secChUaMobile?: string } {
const isMobile = deviceCategory === 'mobile' || deviceCategory === 'tablet';
// Extract Chrome version if present
const chromeMatch = userAgent.match(/Chrome\/(\d+)/);
const edgeMatch = userAgent.match(/Edg\/(\d+)/);
if (edgeMatch) {
const version = edgeMatch[1];
return {
secChUa: `"Microsoft Edge";v="${version}", "Chromium";v="${version}", "Not_A Brand";v="24"`,
secChUaPlatform: userAgent.includes('Windows') ? '"Windows"' : userAgent.includes('Android') ? '"Android"' : '"macOS"',
secChUaMobile: isMobile ? '?1' : '?0',
};
}
if (chromeMatch) {
const version = chromeMatch[1];
let platform = '"Linux"';
if (userAgent.includes('Windows')) platform = '"Windows"';
else if (userAgent.includes('Mac')) platform = '"macOS"';
else if (userAgent.includes('Android')) platform = '"Android"';
else if (userAgent.includes('iPhone') || userAgent.includes('iPad')) platform = '"iOS"';
return {
secChUa: `"Google Chrome";v="${version}", "Chromium";v="${version}", "Not_A Brand";v="24"`,
secChUaPlatform: platform,
secChUaMobile: isMobile ? '?1' : '?0',
};
}
// Firefox/Safari don't send sec-ch-ua
return {};
}
getCount(): number {
return 1; // user-agents generates dynamically
return this.userAgents.length;
}
}
// ============================================================
// COMBINED ROTATOR
// Per workflow-12102025.md: Coordinates proxy + fingerprint rotation
// COMBINED ROTATOR (for convenience)
// ============================================================
export class CrawlRotator {
@@ -580,51 +359,49 @@ export class CrawlRotator {
this.userAgent = new UserAgentRotator();
}
/**
* Initialize rotator (load proxies from DB)
*/
async initialize(): Promise<void> {
await this.proxy.loadProxies();
}
/**
* Rotate proxy only (get new IP)
* Rotate proxy only
*/
rotateProxy(): Proxy | null {
return this.proxy.getNext();
}
/**
* Rotate fingerprint only (new UA, screen size, etc.)
* Rotate user agent only
*/
rotateFingerprint(): BrowserFingerprint {
return this.userAgent.rotate();
rotateUserAgent(): string {
return this.userAgent.getNext();
}
/**
* Rotate both proxy and fingerprint
* Per workflow-12102025.md: called on 403 for fresh identity
* Passes proxy IP to UA rotation for session logging
* Rotate both proxy and user agent
*/
rotateBoth(): { proxy: Proxy | null; fingerprint: BrowserFingerprint } {
const proxy = this.proxy.getNext();
const proxyIp = proxy ? proxy.host : undefined;
rotateBoth(): { proxy: Proxy | null; userAgent: string } {
return {
proxy,
fingerprint: this.userAgent.rotate(proxyIp),
proxy: this.proxy.getNext(),
userAgent: this.userAgent.getNext(),
};
}
/**
* Get current proxy and fingerprint without rotating
* Get current proxy and user agent without rotating
*/
getCurrent(): { proxy: Proxy | null; fingerprint: BrowserFingerprint } {
getCurrent(): { proxy: Proxy | null; userAgent: string } {
return {
proxy: this.proxy.getCurrent(),
fingerprint: this.userAgent.getCurrent(),
userAgent: this.userAgent.getCurrent(),
};
}
/**
* Record success for current proxy
* Per workflow-12102025.md: resets consecutive 403 count
*/
async recordSuccess(responseTimeMs?: number): Promise<void> {
const current = this.proxy.getCurrent();
@@ -634,20 +411,7 @@ export class CrawlRotator {
}
/**
* Record 403 block for current proxy
* Per workflow-12102025.md: increments consecutive_403_count, disables after 3
* Returns true if proxy was disabled
*/
async recordBlock(): Promise<boolean> {
const current = this.proxy.getCurrent();
if (current) {
return await this.proxy.markBlocked(current.id);
}
return false;
}
/**
* Record general failure (not 403)
* Record failure for current proxy
*/
async recordFailure(error?: string): Promise<void> {
const current = this.proxy.getCurrent();
@@ -657,13 +421,14 @@ export class CrawlRotator {
}
/**
* Get current proxy location info
* Per workflow-12102025.md: proxy location determines session headers
* Get current proxy location info (for reporting)
* Note: For rotating proxies (like IPRoyal), the actual exit location varies per request
*/
getProxyLocation(): { city?: string; state?: string; country?: string; timezone?: string; isRotating: boolean } | null {
const current = this.proxy.getCurrent();
if (!current) return null;
// Check if this is a rotating proxy (max_connections > 1 usually indicates rotating)
const isRotating = current.maxConnections > 1;
return {
@@ -674,127 +439,6 @@ export class CrawlRotator {
isRotating
};
}
/**
* Get timezone from current proxy
* Per workflow-12102025.md: used for Accept-Language header
*/
getProxyTimezone(): string | undefined {
const current = this.proxy.getCurrent();
return current?.timezone;
}
/**
* Preflight check - verifies proxy and anti-detect are working
* MUST be called before any task execution to ensure anonymity.
*
* Tests:
* 1. Proxy available - a proxy must be loaded and active
* 2. Proxy connectivity - makes HTTP request through proxy to verify connection
* 3. Anti-detect headers - verifies fingerprint is set with required headers
*
* @returns Promise<PreflightResult> with pass/fail status and details
*/
async preflight(): Promise<PreflightResult> {
const result: PreflightResult = {
passed: false,
proxyAvailable: false,
proxyConnected: false,
antidetectReady: false,
proxyIp: null,
fingerprint: null,
error: null,
responseTimeMs: null,
};
// Step 1: Check proxy is available
const currentProxy = this.proxy.getCurrent();
if (!currentProxy) {
result.error = 'No proxy available';
console.log('[Preflight] FAILED - No proxy available');
return result;
}
result.proxyAvailable = true;
result.proxyIp = currentProxy.host;
// Step 2: Check fingerprint/anti-detect is ready
const fingerprint = this.userAgent.getCurrent();
if (!fingerprint || !fingerprint.userAgent) {
result.error = 'Anti-detect fingerprint not initialized';
console.log('[Preflight] FAILED - No fingerprint');
return result;
}
result.antidetectReady = true;
result.fingerprint = {
userAgent: fingerprint.userAgent,
browserName: fingerprint.browserName,
deviceCategory: fingerprint.deviceCategory,
};
// Step 3: Test proxy connectivity with an actual HTTP request
// Use httpbin.org/ip to verify request goes through proxy
const proxyUrl = this.proxy.getProxyUrl(currentProxy);
const testUrl = 'https://httpbin.org/ip';
try {
const { default: axios } = await import('axios');
const { HttpsProxyAgent } = await import('https-proxy-agent');
const agent = new HttpsProxyAgent(proxyUrl);
const startTime = Date.now();
const response = await axios.get(testUrl, {
httpsAgent: agent,
timeout: 15000, // 15 second timeout
headers: {
'User-Agent': fingerprint.userAgent,
'Accept-Language': fingerprint.acceptLanguage,
...(fingerprint.secChUa && { 'sec-ch-ua': fingerprint.secChUa }),
...(fingerprint.secChUaPlatform && { 'sec-ch-ua-platform': fingerprint.secChUaPlatform }),
...(fingerprint.secChUaMobile && { 'sec-ch-ua-mobile': fingerprint.secChUaMobile }),
},
});
result.responseTimeMs = Date.now() - startTime;
result.proxyConnected = true;
result.passed = true;
// Mark success on proxy stats
await this.proxy.markSuccess(currentProxy.id, result.responseTimeMs);
console.log(`[Preflight] PASSED - Proxy ${currentProxy.host} connected (${result.responseTimeMs}ms), UA: ${fingerprint.browserName}/${fingerprint.deviceCategory}`);
} catch (err: any) {
result.error = `Proxy connection failed: ${err.message || 'Unknown error'}`;
console.log(`[Preflight] FAILED - Proxy connection error: ${err.message}`);
// Mark failure on proxy stats
await this.proxy.markFailed(currentProxy.id, err.message);
}
return result;
}
}
/**
* Result from preflight check
*/
export interface PreflightResult {
/** Overall pass/fail */
passed: boolean;
/** Step 1: Is a proxy loaded? */
proxyAvailable: boolean;
/** Step 2: Did HTTP request through proxy succeed? */
proxyConnected: boolean;
/** Step 3: Is fingerprint/anti-detect ready? */
antidetectReady: boolean;
/** Current proxy IP */
proxyIp: string | null;
/** Fingerprint summary */
fingerprint: { userAgent: string; browserName: string; deviceCategory: string } | null;
/** Error message if failed */
error: string | null;
/** Proxy response time in ms */
responseTimeMs: number | null;
}
// ============================================================

View File

@@ -1,100 +0,0 @@
/**
* Curl Preflight - Verify curl/axios transport works through proxy
*
* Tests:
* 1. Proxy is available and active
* 2. HTTP request through proxy succeeds
* 3. Anti-detect headers are properly set
*
* Use case: Fast, simple API requests that don't need browser fingerprint
*/
import axios from 'axios';
import { HttpsProxyAgent } from 'https-proxy-agent';
import { CrawlRotator, PreflightResult } from './crawl-rotator';
export interface CurlPreflightResult extends PreflightResult {
method: 'curl';
}
/**
* Run curl preflight check
* Tests proxy connectivity using axios/curl through the proxy
*/
export async function runCurlPreflight(
crawlRotator: CrawlRotator
): Promise<CurlPreflightResult> {
const result: CurlPreflightResult = {
method: 'curl',
passed: false,
proxyAvailable: false,
proxyConnected: false,
antidetectReady: false,
proxyIp: null,
fingerprint: null,
error: null,
responseTimeMs: null,
};
// Step 1: Check proxy is available
const currentProxy = crawlRotator.proxy.getCurrent();
if (!currentProxy) {
result.error = 'No proxy available';
console.log('[CurlPreflight] FAILED - No proxy available');
return result;
}
result.proxyAvailable = true;
result.proxyIp = currentProxy.host;
// Step 2: Check fingerprint/anti-detect is ready
const fingerprint = crawlRotator.userAgent.getCurrent();
if (!fingerprint || !fingerprint.userAgent) {
result.error = 'Anti-detect fingerprint not initialized';
console.log('[CurlPreflight] FAILED - No fingerprint');
return result;
}
result.antidetectReady = true;
result.fingerprint = {
userAgent: fingerprint.userAgent,
browserName: fingerprint.browserName,
deviceCategory: fingerprint.deviceCategory,
};
// Step 3: Test proxy connectivity with an actual HTTP request
const proxyUrl = crawlRotator.proxy.getProxyUrl(currentProxy);
const testUrl = 'https://httpbin.org/ip';
try {
const agent = new HttpsProxyAgent(proxyUrl);
const startTime = Date.now();
const response = await axios.get(testUrl, {
httpsAgent: agent,
timeout: 15000, // 15 second timeout
headers: {
'User-Agent': fingerprint.userAgent,
'Accept-Language': fingerprint.acceptLanguage,
...(fingerprint.secChUa && { 'sec-ch-ua': fingerprint.secChUa }),
...(fingerprint.secChUaPlatform && { 'sec-ch-ua-platform': fingerprint.secChUaPlatform }),
...(fingerprint.secChUaMobile && { 'sec-ch-ua-mobile': fingerprint.secChUaMobile }),
},
});
result.responseTimeMs = Date.now() - startTime;
result.proxyConnected = true;
result.passed = true;
// Mark success on proxy stats
await crawlRotator.proxy.markSuccess(currentProxy.id, result.responseTimeMs);
console.log(`[CurlPreflight] PASSED - Proxy ${currentProxy.host} connected (${result.responseTimeMs}ms), UA: ${fingerprint.browserName}/${fingerprint.deviceCategory}`);
} catch (err: any) {
result.error = `Proxy connection failed: ${err.message || 'Unknown error'}`;
console.log(`[CurlPreflight] FAILED - Proxy connection error: ${err.message}`);
// Mark failure on proxy stats
await crawlRotator.proxy.markFailed(currentProxy.id, err.message);
}
return result;
}

View File

@@ -1,315 +0,0 @@
/**
* HTTP Fingerprinting Service
*
* Per workflow-12102025.md - HTTP Fingerprinting section:
* - Full header set per browser type
* - Browser-specific header ordering
* - Natural randomization (DNT, Accept quality)
* - Dynamic Referer per dispensary
*
* Canonical location: src/services/http-fingerprint.ts
*/
// ============================================================
// TYPES
// ============================================================
export type BrowserType = 'Chrome' | 'Firefox' | 'Safari' | 'Edge';
/**
* Per workflow-12102025.md: Full HTTP fingerprint for a session
*/
export interface HTTPFingerprint {
browserType: BrowserType;
headers: Record<string, string>;
headerOrder: string[];
curlImpersonateBinary: string;
hasDNT: boolean;
}
/**
* Per workflow-12102025.md: Context for building headers
*/
export interface HeaderContext {
userAgent: string;
secChUa?: string;
secChUaPlatform?: string;
secChUaMobile?: string;
referer: string;
isPost: boolean;
contentLength?: number;
}
// ============================================================
// CONSTANTS (per workflow-12102025.md)
// ============================================================
/**
* Per workflow-12102025.md: DNT header distribution (~30% of users)
*/
const DNT_PROBABILITY = 0.30;
/**
* Per workflow-12102025.md: Accept header variations for natural traffic
*/
const ACCEPT_VARIATIONS = [
'application/json, text/plain, */*',
'application/json,text/plain,*/*',
'*/*',
];
/**
* Per workflow-12102025.md: Accept-Language variations
*/
const ACCEPT_LANGUAGE_VARIATIONS = [
'en-US,en;q=0.9',
'en-US,en;q=0.8',
'en-US;q=0.9,en;q=0.8',
];
/**
* Per workflow-12102025.md: curl-impersonate binaries per browser
*/
const CURL_IMPERSONATE_BINARIES: Record<BrowserType, string> = {
Chrome: 'curl_chrome131',
Edge: 'curl_chrome131', // Edge uses Chromium
Firefox: 'curl_ff133',
Safari: 'curl_safari17',
};
// ============================================================
// HEADER ORDERING (per workflow-12102025.md)
// ============================================================
/**
* Per workflow-12102025.md: Chrome header order for GraphQL requests
*/
const CHROME_HEADER_ORDER = [
'Host',
'Connection',
'Content-Length',
'sec-ch-ua',
'DNT',
'sec-ch-ua-mobile',
'User-Agent',
'sec-ch-ua-platform',
'Content-Type',
'Accept',
'Origin',
'sec-fetch-site',
'sec-fetch-mode',
'sec-fetch-dest',
'Referer',
'Accept-Encoding',
'Accept-Language',
];
/**
* Per workflow-12102025.md: Firefox header order for GraphQL requests
*/
const FIREFOX_HEADER_ORDER = [
'Host',
'User-Agent',
'Accept',
'Accept-Language',
'Accept-Encoding',
'Content-Type',
'Content-Length',
'Origin',
'DNT',
'Connection',
'Referer',
'sec-fetch-dest',
'sec-fetch-mode',
'sec-fetch-site',
];
/**
* Per workflow-12102025.md: Safari header order for GraphQL requests
*/
const SAFARI_HEADER_ORDER = [
'Host',
'Connection',
'Content-Length',
'Accept',
'User-Agent',
'Content-Type',
'Origin',
'Referer',
'Accept-Encoding',
'Accept-Language',
];
/**
* Per workflow-12102025.md: Edge uses Chrome order (Chromium-based)
*/
const HEADER_ORDERS: Record<BrowserType, string[]> = {
Chrome: CHROME_HEADER_ORDER,
Edge: CHROME_HEADER_ORDER,
Firefox: FIREFOX_HEADER_ORDER,
Safari: SAFARI_HEADER_ORDER,
};
// ============================================================
// FINGERPRINT GENERATION
// ============================================================
/**
* Per workflow-12102025.md: Generate HTTP fingerprint for a session
* Randomization is done once per session for consistency
*/
export function generateHTTPFingerprint(browserType: BrowserType): HTTPFingerprint {
// Per workflow-12102025.md: DNT randomized per session (~30%)
const hasDNT = Math.random() < DNT_PROBABILITY;
return {
browserType,
headers: {}, // Built dynamically per request
headerOrder: HEADER_ORDERS[browserType],
curlImpersonateBinary: CURL_IMPERSONATE_BINARIES[browserType],
hasDNT,
};
}
/**
* Per workflow-12102025.md: Build complete headers for a request
* Returns headers in browser-specific order
*/
export function buildOrderedHeaders(
fingerprint: HTTPFingerprint,
context: HeaderContext
): { headers: Record<string, string>; orderedHeaders: string[] } {
const { browserType, hasDNT, headerOrder } = fingerprint;
const { userAgent, secChUa, secChUaPlatform, secChUaMobile, referer, isPost, contentLength } = context;
// Per workflow-12102025.md: Natural randomization for Accept
const accept = ACCEPT_VARIATIONS[Math.floor(Math.random() * ACCEPT_VARIATIONS.length)];
const acceptLanguage = ACCEPT_LANGUAGE_VARIATIONS[Math.floor(Math.random() * ACCEPT_LANGUAGE_VARIATIONS.length)];
// Build all possible headers
const allHeaders: Record<string, string> = {
'Connection': 'keep-alive',
'User-Agent': userAgent,
'Accept': accept,
'Accept-Language': acceptLanguage,
'Accept-Encoding': 'gzip, deflate, br',
};
// Per workflow-12102025.md: POST-only headers
if (isPost) {
allHeaders['Content-Type'] = 'application/json';
allHeaders['Origin'] = 'https://dutchie.com';
if (contentLength !== undefined) {
allHeaders['Content-Length'] = String(contentLength);
}
}
// Per workflow-12102025.md: Dynamic Referer per dispensary
allHeaders['Referer'] = referer;
// Per workflow-12102025.md: DNT randomized per session
if (hasDNT) {
allHeaders['DNT'] = '1';
}
// Per workflow-12102025.md: Chromium-only headers (Chrome, Edge)
if (browserType === 'Chrome' || browserType === 'Edge') {
if (secChUa) allHeaders['sec-ch-ua'] = secChUa;
if (secChUaMobile) allHeaders['sec-ch-ua-mobile'] = secChUaMobile;
if (secChUaPlatform) allHeaders['sec-ch-ua-platform'] = secChUaPlatform;
allHeaders['sec-fetch-site'] = 'same-origin';
allHeaders['sec-fetch-mode'] = 'cors';
allHeaders['sec-fetch-dest'] = 'empty';
}
// Per workflow-12102025.md: Firefox has sec-fetch but no sec-ch
if (browserType === 'Firefox') {
allHeaders['sec-fetch-site'] = 'same-origin';
allHeaders['sec-fetch-mode'] = 'cors';
allHeaders['sec-fetch-dest'] = 'empty';
}
// Per workflow-12102025.md: Safari has no sec-* headers
// Filter to only headers that exist and order them
const orderedHeaders: string[] = [];
const headers: Record<string, string> = {};
for (const headerName of headerOrder) {
if (allHeaders[headerName]) {
orderedHeaders.push(headerName);
headers[headerName] = allHeaders[headerName];
}
}
return { headers, orderedHeaders };
}
/**
* Per workflow-12102025.md: Build curl command arguments for headers
* Headers are added in browser-specific order
*/
export function buildCurlHeaderArgs(
fingerprint: HTTPFingerprint,
context: HeaderContext
): string[] {
const { headers, orderedHeaders } = buildOrderedHeaders(fingerprint, context);
const args: string[] = [];
for (const headerName of orderedHeaders) {
// Skip Host and Content-Length - curl handles these
if (headerName === 'Host' || headerName === 'Content-Length') continue;
args.push('-H', `${headerName}: ${headers[headerName]}`);
}
return args;
}
/**
* Per workflow-12102025.md: Extract Referer from dispensary menu_url
*/
export function buildRefererFromMenuUrl(menuUrl: string | null | undefined): string {
if (!menuUrl) {
return 'https://dutchie.com/';
}
// Extract slug from menu_url
// Formats: /embedded-menu/<slug> or /dispensary/<slug> or full URL
let slug: string | null = null;
const embeddedMatch = menuUrl.match(/\/embedded-menu\/([^/?]+)/);
const dispensaryMatch = menuUrl.match(/\/dispensary\/([^/?]+)/);
if (embeddedMatch) {
slug = embeddedMatch[1];
} else if (dispensaryMatch) {
slug = dispensaryMatch[1];
}
if (slug) {
return `https://dutchie.com/dispensary/${slug}`;
}
return 'https://dutchie.com/';
}
/**
* Per workflow-12102025.md: Get curl-impersonate binary for browser
*/
export function getCurlBinary(browserType: BrowserType): string {
return CURL_IMPERSONATE_BINARIES[browserType];
}
/**
* Per workflow-12102025.md: Check if curl-impersonate is available
*/
export function isCurlImpersonateAvailable(browserType: BrowserType): boolean {
const binary = CURL_IMPERSONATE_BINARIES[browserType];
try {
const { execSync } = require('child_process');
execSync(`which ${binary}`, { stdio: 'ignore' });
return true;
} catch {
return false;
}
}

View File

@@ -1,399 +0,0 @@
/**
* Puppeteer Preflight - Verify browser-based transport works with anti-detect
*
* Uses Puppeteer + StealthPlugin to:
* 1. Launch headless browser with stealth mode + PROXY
* 2. Visit fingerprint.com demo to verify anti-detect and confirm proxy IP
* 3. Establish session by visiting Dutchie embedded menu
* 4. Make GraphQL request from browser context
* 5. Verify we get a valid response (not blocked)
*
* Use case: Anti-detect scraping that needs real browser fingerprint through proxy
*
* Based on test-intercept.js which successfully captures 1000+ products
*/
import { PreflightResult, CrawlRotator } from './crawl-rotator';
// GraphQL hash for FilteredProducts query - MUST match CLAUDE.md
const FILTERED_PRODUCTS_HASH = 'ee29c060826dc41c527e470e9ae502c9b2c169720faa0a9f5d25e1b9a530a4a0';
// Test dispensary - AZ-Deeply-Rooted (known working)
const TEST_CNAME = 'AZ-Deeply-Rooted';
const TEST_PLATFORM_ID = '6405ef617056e8014d79101b';
// Anti-detect verification sites (primary + fallback)
const FINGERPRINT_DEMO_URL = 'https://demo.fingerprint.com/';
const AMIUNIQUE_URL = 'https://amiunique.org/fingerprint';
export interface PuppeteerPreflightResult extends PreflightResult {
method: 'http';
/** Number of products returned (proves API access) */
productsReturned?: number;
/** Browser user agent used */
browserUserAgent?: string;
/** Bot detection result from fingerprint.com */
botDetection?: {
detected: boolean;
probability?: number;
type?: string;
};
/** Expected proxy IP (from pool) */
expectedProxyIp?: string;
/** Whether IP verification passed (detected IP matches proxy) */
ipVerified?: boolean;
}
/**
* Run Puppeteer preflight check with proxy
* Tests browser-based access with anti-detect verification via fingerprint.com
*
* @param crawlRotator - CrawlRotator instance to get proxy from pool
*/
export async function runPuppeteerPreflight(
crawlRotator?: CrawlRotator
): Promise<PuppeteerPreflightResult> {
const result: PuppeteerPreflightResult = {
method: 'http',
passed: false,
proxyAvailable: false,
proxyConnected: false,
antidetectReady: false,
proxyIp: null,
fingerprint: null,
error: null,
responseTimeMs: null,
productsReturned: 0,
ipVerified: false,
};
let browser: any = null;
try {
// Step 0: Get a proxy from the pool
let proxyUrl: string | null = null;
let expectedProxyHost: string | null = null;
if (crawlRotator) {
const currentProxy = crawlRotator.proxy.getCurrent();
if (currentProxy) {
result.proxyAvailable = true;
proxyUrl = crawlRotator.proxy.getProxyUrl(currentProxy);
expectedProxyHost = currentProxy.host;
result.expectedProxyIp = expectedProxyHost;
console.log(`[PuppeteerPreflight] Using proxy: ${currentProxy.host}:${currentProxy.port}`);
} else {
result.error = 'No proxy available from pool';
console.log(`[PuppeteerPreflight] FAILED - No proxy available`);
return result;
}
} else {
console.log(`[PuppeteerPreflight] WARNING: No CrawlRotator provided - using direct connection`);
result.proxyAvailable = true; // No proxy needed for direct
}
// Dynamic imports to avoid loading Puppeteer unless needed
const puppeteer = require('puppeteer-extra');
const StealthPlugin = require('puppeteer-extra-plugin-stealth');
puppeteer.use(StealthPlugin());
const startTime = Date.now();
// Build browser args
const browserArgs = ['--no-sandbox', '--disable-setuid-sandbox'];
if (proxyUrl) {
// Extract host:port for Puppeteer (it handles auth separately)
const proxyUrlParsed = new URL(proxyUrl);
browserArgs.push(`--proxy-server=${proxyUrlParsed.host}`);
}
// Launch browser with stealth + proxy
browser = await puppeteer.launch({
headless: 'new',
args: browserArgs,
});
const page = await browser.newPage();
// If proxy has auth, set it up
if (proxyUrl) {
const proxyUrlParsed = new URL(proxyUrl);
if (proxyUrlParsed.username && proxyUrlParsed.password) {
await page.authenticate({
username: decodeURIComponent(proxyUrlParsed.username),
password: decodeURIComponent(proxyUrlParsed.password),
});
}
}
// Get browser user agent
const userAgent = await page.evaluate(() => navigator.userAgent);
result.browserUserAgent = userAgent;
result.fingerprint = {
userAgent,
browserName: 'Chrome (Puppeteer)',
deviceCategory: 'desktop',
};
// =========================================================================
// STEP 1: Visit fingerprint.com demo to verify anti-detect and get IP
// =========================================================================
console.log(`[PuppeteerPreflight] Testing anti-detect at ${FINGERPRINT_DEMO_URL}...`);
try {
await page.goto(FINGERPRINT_DEMO_URL, {
waitUntil: 'networkidle2',
timeout: 30000,
});
result.proxyConnected = true; // If we got here, proxy is working
// Wait for fingerprint results to load
await page.waitForSelector('[data-test="visitor-id"]', { timeout: 10000 }).catch(() => {});
// Extract fingerprint data from the page
const fingerprintData = await page.evaluate(() => {
// Try to find the IP address displayed on the page
const ipElement = document.querySelector('[data-test="ip-address"]');
const ip = ipElement?.textContent?.trim() || null;
// Try to find bot detection info
const botElement = document.querySelector('[data-test="bot-detected"]');
const botDetected = botElement?.textContent?.toLowerCase().includes('true') || false;
// Try to find visitor ID (proves fingerprinting worked)
const visitorIdElement = document.querySelector('[data-test="visitor-id"]');
const visitorId = visitorIdElement?.textContent?.trim() || null;
// Alternative: look for common UI patterns if data-test attrs not present
let detectedIp = ip;
if (!detectedIp) {
// Look for IP in any element containing IP-like pattern
const allText = document.body.innerText;
const ipMatch = allText.match(/\b(\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})\b/);
detectedIp = ipMatch ? ipMatch[1] : null;
}
return {
ip: detectedIp,
botDetected,
visitorId,
pageLoaded: !!document.body,
};
});
if (fingerprintData.ip) {
result.proxyIp = fingerprintData.ip;
console.log(`[PuppeteerPreflight] Detected IP: ${fingerprintData.ip}`);
// Verify IP matches expected proxy
if (expectedProxyHost) {
// Check if detected IP contains the proxy host (or is close match)
if (fingerprintData.ip === expectedProxyHost ||
expectedProxyHost.includes(fingerprintData.ip) ||
fingerprintData.ip.includes(expectedProxyHost.split('.').slice(0, 3).join('.'))) {
result.ipVerified = true;
console.log(`[PuppeteerPreflight] IP VERIFIED - matches proxy`);
} else {
console.log(`[PuppeteerPreflight] IP mismatch: expected ${expectedProxyHost}, got ${fingerprintData.ip}`);
// Don't fail - residential proxies often show different egress IPs
}
}
}
if (fingerprintData.visitorId) {
console.log(`[PuppeteerPreflight] Fingerprint visitor ID: ${fingerprintData.visitorId}`);
}
result.botDetection = {
detected: fingerprintData.botDetected,
};
if (fingerprintData.botDetected) {
console.log(`[PuppeteerPreflight] WARNING: Bot detection triggered!`);
} else {
console.log(`[PuppeteerPreflight] Anti-detect check: NOT detected as bot`);
result.antidetectReady = true;
}
} catch (fpErr: any) {
// Could mean proxy connection failed
console.log(`[PuppeteerPreflight] Fingerprint.com check failed: ${fpErr.message}`);
if (fpErr.message.includes('net::ERR_PROXY') || fpErr.message.includes('ECONNREFUSED')) {
result.error = `Proxy connection failed: ${fpErr.message}`;
return result;
}
// Try fallback: amiunique.org
console.log(`[PuppeteerPreflight] Trying fallback: ${AMIUNIQUE_URL}...`);
try {
await page.goto(AMIUNIQUE_URL, {
waitUntil: 'networkidle2',
timeout: 30000,
});
result.proxyConnected = true;
// Extract IP from amiunique.org page
const amiData = await page.evaluate(() => {
const allText = document.body.innerText;
const ipMatch = allText.match(/\b(\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})\b/);
return {
ip: ipMatch ? ipMatch[1] : null,
pageLoaded: !!document.body,
};
});
if (amiData.ip) {
result.proxyIp = amiData.ip;
console.log(`[PuppeteerPreflight] Detected IP via amiunique.org: ${amiData.ip}`);
}
result.antidetectReady = true;
console.log(`[PuppeteerPreflight] amiunique.org fallback succeeded`);
} catch (amiErr: any) {
console.log(`[PuppeteerPreflight] amiunique.org fallback also failed: ${amiErr.message}`);
// Continue with Dutchie test anyway
result.proxyConnected = true;
result.antidetectReady = true;
}
}
// =========================================================================
// STEP 2: Test Dutchie API access (the real test)
// =========================================================================
const embedUrl = `https://dutchie.com/embedded-menu/${TEST_CNAME}?menuType=rec`;
console.log(`[PuppeteerPreflight] Establishing session at ${embedUrl}...`);
await page.goto(embedUrl, {
waitUntil: 'networkidle2',
timeout: 30000,
});
// Make GraphQL request from browser context
const graphqlResult = await page.evaluate(
async (platformId: string, hash: string) => {
try {
const variables = {
includeEnterpriseSpecials: false,
productsFilter: {
dispensaryId: platformId,
pricingType: 'rec',
Status: 'Active', // CRITICAL: Must be 'Active' per CLAUDE.md
types: [],
useCache: true,
isDefaultSort: true,
sortBy: 'popularSortIdx',
sortDirection: 1,
bypassOnlineThresholds: true,
isKioskMenu: false,
removeProductsBelowOptionThresholds: false,
},
page: 0,
perPage: 10, // Just need a few to prove it works
};
const extensions = {
persistedQuery: {
version: 1,
sha256Hash: hash,
},
};
const qs = new URLSearchParams({
operationName: 'FilteredProducts',
variables: JSON.stringify(variables),
extensions: JSON.stringify(extensions),
});
const url = `https://dutchie.com/api-3/graphql?${qs.toString()}`;
const sessionId = 'preflight-' + Date.now();
const response = await fetch(url, {
method: 'GET',
headers: {
Accept: 'application/json',
'content-type': 'application/json',
'x-dutchie-session': sessionId,
'apollographql-client-name': 'Marketplace (production)',
},
credentials: 'include',
});
if (!response.ok) {
return { error: `HTTP ${response.status}`, products: 0 };
}
const json = await response.json();
if (json.errors) {
return { error: JSON.stringify(json.errors).slice(0, 200), products: 0 };
}
const products = json?.data?.filteredProducts?.products || [];
return { error: null, products: products.length };
} catch (err: any) {
return { error: err.message || 'Unknown error', products: 0 };
}
},
TEST_PLATFORM_ID,
FILTERED_PRODUCTS_HASH
);
result.responseTimeMs = Date.now() - startTime;
if (graphqlResult.error) {
result.error = `GraphQL error: ${graphqlResult.error}`;
console.log(`[PuppeteerPreflight] FAILED - ${result.error}`);
} else if (graphqlResult.products === 0) {
result.error = 'GraphQL returned 0 products';
console.log(`[PuppeteerPreflight] FAILED - No products returned`);
} else {
result.passed = true;
result.productsReturned = graphqlResult.products;
console.log(
`[PuppeteerPreflight] PASSED - Got ${graphqlResult.products} products in ${result.responseTimeMs}ms`
);
if (result.proxyIp) {
console.log(`[PuppeteerPreflight] Browser IP via proxy: ${result.proxyIp}`);
}
}
} catch (err: any) {
result.error = `Browser error: ${err.message || 'Unknown error'}`;
console.log(`[PuppeteerPreflight] FAILED - ${result.error}`);
} finally {
if (browser) {
await browser.close().catch(() => {});
}
}
return result;
}
/**
* Run Puppeteer preflight with retry
* Retries once on failure to handle transient issues
*
* @param crawlRotator - CrawlRotator instance to get proxy from pool
* @param maxRetries - Number of retry attempts (default 1)
*/
export async function runPuppeteerPreflightWithRetry(
crawlRotator?: CrawlRotator,
maxRetries: number = 1
): Promise<PuppeteerPreflightResult> {
let lastResult: PuppeteerPreflightResult | null = null;
for (let attempt = 0; attempt <= maxRetries; attempt++) {
if (attempt > 0) {
console.log(`[PuppeteerPreflight] Retry attempt ${attempt}/${maxRetries}...`);
await new Promise((r) => setTimeout(r, 5000)); // Wait 5s between retries
}
lastResult = await runPuppeteerPreflight(crawlRotator);
if (lastResult.passed) {
return lastResult;
}
}
return lastResult!;
}

View File

@@ -1,38 +1,116 @@
/**
* LEGACY SCHEDULER - DEPRECATED 2024-12-10
*
* DO NOT USE THIS FILE.
*
* Per TASK_WORKFLOW_2024-12-10.md:
* This node-cron scheduler has been replaced by the database-driven
* task scheduler in src/services/task-scheduler.ts
*
* The new scheduler:
* - Stores schedules in PostgreSQL (survives restarts)
* - Uses SELECT FOR UPDATE SKIP LOCKED (multi-replica safe)
* - Creates tasks in worker_tasks table (processed by task-worker.ts)
*
* This file is kept for reference only. All exports are no-ops.
* Legacy code has been removed - see git history for original implementation.
*/
import cron from 'node-cron';
import { pool } from '../db/pool';
import { scrapeStore, scrapeCategory } from '../scraper-v2';
let scheduledJobs: cron.ScheduledTask[] = [];
async function getSettings(): Promise<{
scrapeIntervalHours: number;
scrapeSpecialsTime: string;
}> {
const result = await pool.query(`
SELECT key, value FROM settings
WHERE key IN ('scrape_interval_hours', 'scrape_specials_time')
`);
const settings: Record<string, string> = {};
result.rows.forEach((row: { key: string; value: string }) => {
settings[row.key] = row.value;
});
return {
scrapeIntervalHours: parseInt(settings.scrape_interval_hours || '4'),
scrapeSpecialsTime: settings.scrape_specials_time || '00:01'
};
}
async function scrapeAllStores(): Promise<void> {
console.log('🔄 Starting scheduled scrape for all stores...');
const result = await pool.query(`
SELECT id, name FROM stores WHERE active = true AND scrape_enabled = true
`);
for (const store of result.rows) {
try {
console.log(`Scraping store: ${store.name}`);
await scrapeStore(store.id);
} catch (error) {
console.error(`Failed to scrape store ${store.name}:`, error);
}
}
console.log('✅ Scheduled scrape completed');
}
async function scrapeSpecials(): Promise<void> {
console.log('🌟 Starting scheduled specials scrape...');
const result = await pool.query(`
SELECT s.id, s.name, c.id as category_id
FROM stores s
JOIN categories c ON c.store_id = s.id
WHERE s.active = true AND s.scrape_enabled = true
AND c.slug = 'specials' AND c.scrape_enabled = true
`);
for (const row of result.rows) {
try {
console.log(`Scraping specials for: ${row.name}`);
await scrapeCategory(row.id, row.category_id);
} catch (error) {
console.error(`Failed to scrape specials for ${row.name}:`, error);
}
}
console.log('✅ Specials scrape completed');
}
// 2024-12-10: All functions are now no-ops
export async function startScheduler(): Promise<void> {
console.warn('[DEPRECATED] startScheduler() called - use taskScheduler from task-scheduler.ts instead');
// Stop any existing jobs
stopScheduler();
const settings = await getSettings();
// Schedule regular store scrapes (every N hours)
const scrapeIntervalCron = `0 */${settings.scrapeIntervalHours} * * *`;
const storeJob = cron.schedule(scrapeIntervalCron, scrapeAllStores);
scheduledJobs.push(storeJob);
console.log(`📅 Scheduled store scraping: every ${settings.scrapeIntervalHours} hours`);
// Schedule specials scraping (daily at specified time)
const [hours, minutes] = settings.scrapeSpecialsTime.split(':');
const specialsCron = `${minutes} ${hours} * * *`;
const specialsJob = cron.schedule(specialsCron, scrapeSpecials);
scheduledJobs.push(specialsJob);
console.log(`📅 Scheduled specials scraping: daily at ${settings.scrapeSpecialsTime}`);
// Initial scrape on startup (after 10 seconds)
setTimeout(() => {
console.log('🚀 Running initial scrape...');
scrapeAllStores().catch(console.error);
}, 10000);
}
export function stopScheduler(): void {
console.warn('[DEPRECATED] stopScheduler() called - use taskScheduler from task-scheduler.ts instead');
scheduledJobs.forEach(job => job.stop());
scheduledJobs = [];
console.log('🛑 Scheduler stopped');
}
export async function restartScheduler(): Promise<void> {
console.warn('[DEPRECATED] restartScheduler() called - use taskScheduler from task-scheduler.ts instead');
console.log('🔄 Restarting scheduler...');
stopScheduler();
await startScheduler();
}
export async function triggerStoreScrape(_storeId: number): Promise<void> {
console.warn('[DEPRECATED] triggerStoreScrape() called - use taskService.createTask() instead');
// Manual trigger functions for admin
export async function triggerStoreScrape(storeId: number): Promise<void> {
console.log(`🔧 Manual scrape triggered for store ID: ${storeId}`);
await scrapeStore(storeId);
}
export async function triggerAllStoresScrape(): Promise<void> {
console.warn('[DEPRECATED] triggerAllStoresScrape() called - use taskScheduler.triggerSchedule() instead');
console.log('🔧 Manual scrape triggered for all stores');
await scrapeAllStores();
}

View File

@@ -1,375 +0,0 @@
/**
* Database-Driven Task Scheduler
*
* Per TASK_WORKFLOW_2024-12-10.md:
* - Schedules stored in DB (survives restarts)
* - Uses SELECT FOR UPDATE to prevent duplicate execution across replicas
* - Polls every 60s to check if schedules are due
* - Generates tasks into worker_tasks table for task-worker.ts to process
*
* 2024-12-10: Created to replace legacy node-cron scheduler
*/
import { pool } from '../db/pool';
import { taskService, TaskRole } from '../tasks/task-service';
// Per TASK_WORKFLOW_2024-12-10.md: Poll interval for checking schedules
const POLL_INTERVAL_MS = 60_000; // 60 seconds
interface TaskSchedule {
id: number;
name: string;
role: TaskRole;
enabled: boolean;
interval_hours: number;
last_run_at: Date | null;
next_run_at: Date | null;
state_code: string | null;
priority: number;
}
class TaskScheduler {
private pollTimer: NodeJS.Timeout | null = null;
private isRunning = false;
/**
* Start the scheduler
* Per TASK_WORKFLOW_2024-12-10.md: Called on API server startup
*/
async start(): Promise<void> {
if (this.isRunning) {
console.log('[TaskScheduler] Already running');
return;
}
console.log('[TaskScheduler] Starting database-driven scheduler...');
this.isRunning = true;
// Per TASK_WORKFLOW_2024-12-10.md: On startup, recover stale tasks
try {
const recovered = await taskService.recoverStaleTasks(10);
if (recovered > 0) {
console.log(`[TaskScheduler] Recovered ${recovered} stale tasks from dead workers`);
}
} catch (err: any) {
console.error('[TaskScheduler] Failed to recover stale tasks:', err.message);
}
// Per TASK_WORKFLOW_2024-12-10.md: Ensure default schedules exist
await this.ensureDefaultSchedules();
// Per TASK_WORKFLOW_2024-12-10.md: Check immediately on startup
await this.checkAndRunDueSchedules();
// Per TASK_WORKFLOW_2024-12-10.md: Then poll every 60 seconds
this.pollTimer = setInterval(async () => {
await this.checkAndRunDueSchedules();
}, POLL_INTERVAL_MS);
console.log('[TaskScheduler] Started - polling every 60s');
}
/**
* Stop the scheduler
*/
stop(): void {
if (this.pollTimer) {
clearInterval(this.pollTimer);
this.pollTimer = null;
}
this.isRunning = false;
console.log('[TaskScheduler] Stopped');
}
/**
* Ensure default schedules exist in the database
* Per TASK_WORKFLOW_2024-12-10.md: Creates schedules if they don't exist
*/
private async ensureDefaultSchedules(): Promise<void> {
// Per TASK_WORKFLOW_2024-12-10.md: Default schedules for task generation
// NOTE: payload_fetch replaces direct product_refresh - it chains to product_refresh
const defaults = [
{
name: 'payload_fetch_all',
role: 'payload_fetch' as TaskRole,
interval_hours: 4,
priority: 0,
description: 'Fetch payloads from Dutchie API for all crawl-enabled stores every 4 hours. Chains to product_refresh.',
},
{
name: 'store_discovery_dutchie',
role: 'store_discovery' as TaskRole,
interval_hours: 24,
priority: 5,
description: 'Discover new Dutchie stores daily',
},
{
name: 'analytics_refresh',
role: 'analytics_refresh' as TaskRole,
interval_hours: 6,
priority: 0,
description: 'Refresh analytics materialized views every 6 hours',
},
];
for (const sched of defaults) {
try {
await pool.query(`
INSERT INTO task_schedules (name, role, interval_hours, priority, description, enabled, next_run_at)
VALUES ($1, $2, $3, $4, $5, true, NOW())
ON CONFLICT (name) DO NOTHING
`, [sched.name, sched.role, sched.interval_hours, sched.priority, sched.description]);
} catch (err: any) {
// Table may not exist yet - will be created by migration
if (!err.message.includes('does not exist')) {
console.error(`[TaskScheduler] Failed to create default schedule ${sched.name}:`, err.message);
}
}
}
}
/**
* Check for and run any due schedules
* Per TASK_WORKFLOW_2024-12-10.md: Uses SELECT FOR UPDATE SKIP LOCKED to prevent duplicates
*/
private async checkAndRunDueSchedules(): Promise<void> {
const client = await pool.connect();
try {
await client.query('BEGIN');
// Per TASK_WORKFLOW_2024-12-10.md: Atomic claim of due schedules
const result = await client.query<TaskSchedule>(`
SELECT *
FROM task_schedules
WHERE enabled = true
AND (next_run_at IS NULL OR next_run_at <= NOW())
FOR UPDATE SKIP LOCKED
`);
for (const schedule of result.rows) {
console.log(`[TaskScheduler] Running schedule: ${schedule.name} (${schedule.role})`);
try {
const tasksCreated = await this.executeSchedule(schedule);
console.log(`[TaskScheduler] Schedule ${schedule.name} created ${tasksCreated} tasks`);
// Per TASK_WORKFLOW_2024-12-10.md: Update last_run_at and calculate next_run_at
await client.query(`
UPDATE task_schedules
SET
last_run_at = NOW(),
next_run_at = NOW() + ($1 || ' hours')::interval,
last_task_count = $2,
updated_at = NOW()
WHERE id = $3
`, [schedule.interval_hours, tasksCreated, schedule.id]);
} catch (err: any) {
console.error(`[TaskScheduler] Schedule ${schedule.name} failed:`, err.message);
// Still update next_run_at to prevent infinite retry loop
await client.query(`
UPDATE task_schedules
SET
next_run_at = NOW() + ($1 || ' hours')::interval,
last_error = $2,
updated_at = NOW()
WHERE id = $3
`, [schedule.interval_hours, err.message, schedule.id]);
}
}
await client.query('COMMIT');
} catch (err: any) {
await client.query('ROLLBACK');
console.error('[TaskScheduler] Failed to check schedules:', err.message);
} finally {
client.release();
}
}
/**
* Execute a schedule and create tasks
* Per TASK_WORKFLOW_2024-12-10.md: Different logic per role
*/
private async executeSchedule(schedule: TaskSchedule): Promise<number> {
switch (schedule.role) {
case 'payload_fetch':
// Per TASK_WORKFLOW_2024-12-10.md: payload_fetch replaces direct product_refresh
return this.generatePayloadFetchTasks(schedule);
case 'product_refresh':
// Legacy - kept for manual triggers, but scheduled crawls use payload_fetch
return this.generatePayloadFetchTasks(schedule);
case 'store_discovery':
return this.generateStoreDiscoveryTasks(schedule);
case 'analytics_refresh':
return this.generateAnalyticsRefreshTasks(schedule);
default:
console.warn(`[TaskScheduler] Unknown role: ${schedule.role}`);
return 0;
}
}
/**
* Generate payload_fetch tasks for stores that need crawling
* Per TASK_WORKFLOW_2024-12-10.md: payload_fetch hits API, saves to disk, chains to product_refresh
*/
private async generatePayloadFetchTasks(schedule: TaskSchedule): Promise<number> {
// Per TASK_WORKFLOW_2024-12-10.md: Find stores needing refresh
const result = await pool.query(`
SELECT d.id
FROM dispensaries d
WHERE d.crawl_enabled = true
AND d.platform_dispensary_id IS NOT NULL
-- No pending/running payload_fetch or product_refresh task already
AND NOT EXISTS (
SELECT 1 FROM worker_tasks t
WHERE t.dispensary_id = d.id
AND t.role IN ('payload_fetch', 'product_refresh')
AND t.status IN ('pending', 'claimed', 'running')
)
-- Never fetched OR last fetch > interval ago
AND (
d.last_fetch_at IS NULL
OR d.last_fetch_at < NOW() - ($1 || ' hours')::interval
)
${schedule.state_code ? 'AND d.state_id = (SELECT id FROM states WHERE code = $2)' : ''}
`, schedule.state_code ? [schedule.interval_hours, schedule.state_code] : [schedule.interval_hours]);
const dispensaryIds = result.rows.map((r: { id: number }) => r.id);
if (dispensaryIds.length === 0) {
return 0;
}
// Per TASK_WORKFLOW_2024-12-10.md: Create payload_fetch tasks (they chain to product_refresh)
const tasks = dispensaryIds.map((id: number) => ({
role: 'payload_fetch' as TaskRole,
dispensary_id: id,
priority: schedule.priority,
}));
return taskService.createTasks(tasks);
}
/**
* Generate store_discovery tasks
* Per TASK_WORKFLOW_2024-12-10.md: One task per platform
*/
private async generateStoreDiscoveryTasks(schedule: TaskSchedule): Promise<number> {
// Check if discovery task already pending
const existing = await taskService.listTasks({
role: 'store_discovery',
status: ['pending', 'claimed', 'running'],
limit: 1,
});
if (existing.length > 0) {
console.log('[TaskScheduler] Store discovery task already pending, skipping');
return 0;
}
await taskService.createTask({
role: 'store_discovery',
platform: 'dutchie',
priority: schedule.priority,
});
return 1;
}
/**
* Generate analytics_refresh tasks
* Per TASK_WORKFLOW_2024-12-10.md: Single task to refresh all MVs
*/
private async generateAnalyticsRefreshTasks(schedule: TaskSchedule): Promise<number> {
// Check if analytics task already pending
const existing = await taskService.listTasks({
role: 'analytics_refresh',
status: ['pending', 'claimed', 'running'],
limit: 1,
});
if (existing.length > 0) {
console.log('[TaskScheduler] Analytics refresh task already pending, skipping');
return 0;
}
await taskService.createTask({
role: 'analytics_refresh',
priority: schedule.priority,
});
return 1;
}
/**
* Get all schedules for dashboard display
*/
async getSchedules(): Promise<TaskSchedule[]> {
try {
const result = await pool.query(`
SELECT * FROM task_schedules ORDER BY name
`);
return result.rows as TaskSchedule[];
} catch {
return [];
}
}
/**
* Update a schedule
*/
async updateSchedule(id: number, updates: Partial<TaskSchedule>): Promise<void> {
const setClauses: string[] = [];
const values: any[] = [];
let paramIndex = 1;
if (updates.enabled !== undefined) {
setClauses.push(`enabled = $${paramIndex++}`);
values.push(updates.enabled);
}
if (updates.interval_hours !== undefined) {
setClauses.push(`interval_hours = $${paramIndex++}`);
values.push(updates.interval_hours);
}
if (updates.priority !== undefined) {
setClauses.push(`priority = $${paramIndex++}`);
values.push(updates.priority);
}
if (setClauses.length === 0) return;
setClauses.push('updated_at = NOW()');
values.push(id);
await pool.query(`
UPDATE task_schedules
SET ${setClauses.join(', ')}
WHERE id = $${paramIndex}
`, values);
}
/**
* Trigger a schedule to run immediately
*/
async triggerSchedule(id: number): Promise<number> {
const result = await pool.query(`
SELECT * FROM task_schedules WHERE id = $1
`, [id]);
if (result.rows.length === 0) {
throw new Error(`Schedule ${id} not found`);
}
return this.executeSchedule(result.rows[0] as TaskSchedule);
}
}
// Per TASK_WORKFLOW_2024-12-10.md: Singleton instance
export const taskScheduler = new TaskScheduler();

View File

@@ -1,30 +1,566 @@
/**
* System API Routes (Stub)
* System API Routes
*
* The full system routes depend on SyncOrchestrator which was moved to _deprecated.
* This stub provides empty routers to maintain backward compatibility.
* Provides REST API endpoints for system monitoring and control:
* - /api/system/sync/* - Sync orchestrator
* - /api/system/dlq/* - Dead-letter queue
* - /api/system/integrity/* - Integrity checks
* - /api/system/fix/* - Auto-fix routines
* - /api/system/alerts/* - System alerts
* - /metrics - Prometheus metrics
*
* Full implementation available at: src/_deprecated/system/routes/index.ts
* Phase 5: Full Production Sync + Monitoring
*/
import { Router, Request, Response } from 'express';
import { Pool } from 'pg';
import { MetricsService } from '../services';
import {
SyncOrchestrator,
MetricsService,
DLQService,
AlertService,
IntegrityService,
AutoFixService,
} from '../services';
export function createSystemRouter(_pool: Pool): Router {
export function createSystemRouter(pool: Pool): Router {
const router = Router();
// Stub - full sync/dlq/integrity/fix/alerts routes moved to _deprecated
router.get('/status', (_req: Request, res: Response) => {
res.json({
message: 'System routes temporarily disabled - see _deprecated/system/routes',
status: 'stub',
// Initialize services
const metrics = new MetricsService(pool);
const dlq = new DLQService(pool);
const alerts = new AlertService(pool);
const integrity = new IntegrityService(pool, alerts);
const autoFix = new AutoFixService(pool, alerts);
const orchestrator = new SyncOrchestrator(pool, metrics, dlq, alerts);
// ============================================================
// SYNC ORCHESTRATOR ENDPOINTS
// ============================================================
/**
* GET /api/system/sync/status
* Get current sync status
*/
router.get('/sync/status', async (_req: Request, res: Response) => {
try {
const status = await orchestrator.getStatus();
res.json(status);
} catch (error) {
console.error('[System] Sync status error:', error);
res.status(500).json({ error: 'Failed to get sync status' });
}
});
/**
* POST /api/system/sync/run
* Trigger a sync run
*/
router.post('/sync/run', async (req: Request, res: Response) => {
try {
const triggeredBy = req.body.triggeredBy || 'api';
const result = await orchestrator.runSync();
res.json({
success: true,
triggeredBy,
metrics: result,
});
} catch (error) {
console.error('[System] Sync run error:', error);
res.status(500).json({
success: false,
error: error instanceof Error ? error.message : 'Sync run failed',
});
}
});
/**
* GET /api/system/sync/queue-depth
* Get queue depth information
*/
router.get('/sync/queue-depth', async (_req: Request, res: Response) => {
try {
const depth = await orchestrator.getQueueDepth();
res.json(depth);
} catch (error) {
console.error('[System] Queue depth error:', error);
res.status(500).json({ error: 'Failed to get queue depth' });
}
});
/**
* GET /api/system/sync/health
* Get sync health status
*/
router.get('/sync/health', async (_req: Request, res: Response) => {
try {
const health = await orchestrator.getHealth();
res.status(health.healthy ? 200 : 503).json(health);
} catch (error) {
console.error('[System] Health check error:', error);
res.status(500).json({ healthy: false, error: 'Health check failed' });
}
});
/**
* POST /api/system/sync/pause
* Pause the orchestrator
*/
router.post('/sync/pause', async (req: Request, res: Response) => {
try {
const reason = req.body.reason || 'Manual pause';
await orchestrator.pause(reason);
res.json({ success: true, message: 'Orchestrator paused' });
} catch (error) {
console.error('[System] Pause error:', error);
res.status(500).json({ error: 'Failed to pause orchestrator' });
}
});
/**
* POST /api/system/sync/resume
* Resume the orchestrator
*/
router.post('/sync/resume', async (_req: Request, res: Response) => {
try {
await orchestrator.resume();
res.json({ success: true, message: 'Orchestrator resumed' });
} catch (error) {
console.error('[System] Resume error:', error);
res.status(500).json({ error: 'Failed to resume orchestrator' });
}
});
// ============================================================
// DLQ ENDPOINTS
// ============================================================
/**
* GET /api/system/dlq
* List DLQ payloads
*/
router.get('/dlq', async (req: Request, res: Response) => {
try {
const options = {
status: req.query.status as string,
errorType: req.query.errorType as string,
dispensaryId: req.query.dispensaryId ? parseInt(req.query.dispensaryId as string) : undefined,
limit: req.query.limit ? parseInt(req.query.limit as string) : 50,
offset: req.query.offset ? parseInt(req.query.offset as string) : 0,
};
const result = await dlq.listPayloads(options);
res.json(result);
} catch (error) {
console.error('[System] DLQ list error:', error);
res.status(500).json({ error: 'Failed to list DLQ payloads' });
}
});
/**
* GET /api/system/dlq/stats
* Get DLQ statistics
*/
router.get('/dlq/stats', async (_req: Request, res: Response) => {
try {
const stats = await dlq.getStats();
res.json(stats);
} catch (error) {
console.error('[System] DLQ stats error:', error);
res.status(500).json({ error: 'Failed to get DLQ stats' });
}
});
/**
* GET /api/system/dlq/summary
* Get DLQ summary by error type
*/
router.get('/dlq/summary', async (_req: Request, res: Response) => {
try {
const summary = await dlq.getSummary();
res.json(summary);
} catch (error) {
console.error('[System] DLQ summary error:', error);
res.status(500).json({ error: 'Failed to get DLQ summary' });
}
});
/**
* GET /api/system/dlq/:id
* Get a specific DLQ payload
*/
router.get('/dlq/:id', async (req: Request, res: Response) => {
try {
const payload = await dlq.getPayload(req.params.id);
if (!payload) {
return res.status(404).json({ error: 'Payload not found' });
}
res.json(payload);
} catch (error) {
console.error('[System] DLQ get error:', error);
res.status(500).json({ error: 'Failed to get DLQ payload' });
}
});
/**
* POST /api/system/dlq/:id/retry
* Retry a DLQ payload
*/
router.post('/dlq/:id/retry', async (req: Request, res: Response) => {
try {
const result = await dlq.retryPayload(req.params.id);
if (result.success) {
res.json(result);
} else {
res.status(400).json(result);
}
} catch (error) {
console.error('[System] DLQ retry error:', error);
res.status(500).json({ error: 'Failed to retry payload' });
}
});
/**
* POST /api/system/dlq/:id/abandon
* Abandon a DLQ payload
*/
router.post('/dlq/:id/abandon', async (req: Request, res: Response) => {
try {
const reason = req.body.reason || 'Manually abandoned';
const abandonedBy = req.body.abandonedBy || 'api';
const success = await dlq.abandonPayload(req.params.id, reason, abandonedBy);
res.json({ success });
} catch (error) {
console.error('[System] DLQ abandon error:', error);
res.status(500).json({ error: 'Failed to abandon payload' });
}
});
/**
* POST /api/system/dlq/bulk-retry
* Bulk retry payloads by error type
*/
router.post('/dlq/bulk-retry', async (req: Request, res: Response) => {
try {
const { errorType } = req.body;
if (!errorType) {
return res.status(400).json({ error: 'errorType is required' });
}
const result = await dlq.bulkRetryByErrorType(errorType);
res.json(result);
} catch (error) {
console.error('[System] DLQ bulk retry error:', error);
res.status(500).json({ error: 'Failed to bulk retry' });
}
});
// ============================================================
// INTEGRITY CHECK ENDPOINTS
// ============================================================
/**
* POST /api/system/integrity/run
* Run all integrity checks
*/
router.post('/integrity/run', async (req: Request, res: Response) => {
try {
const triggeredBy = req.body.triggeredBy || 'api';
const result = await integrity.runAllChecks(triggeredBy);
res.json(result);
} catch (error) {
console.error('[System] Integrity run error:', error);
res.status(500).json({ error: 'Failed to run integrity checks' });
}
});
/**
* GET /api/system/integrity/runs
* Get recent integrity check runs
*/
router.get('/integrity/runs', async (req: Request, res: Response) => {
try {
const limit = req.query.limit ? parseInt(req.query.limit as string) : 10;
const runs = await integrity.getRecentRuns(limit);
res.json(runs);
} catch (error) {
console.error('[System] Integrity runs error:', error);
res.status(500).json({ error: 'Failed to get integrity runs' });
}
});
/**
* GET /api/system/integrity/runs/:runId
* Get results for a specific integrity run
*/
router.get('/integrity/runs/:runId', async (req: Request, res: Response) => {
try {
const results = await integrity.getRunResults(req.params.runId);
res.json(results);
} catch (error) {
console.error('[System] Integrity run results error:', error);
res.status(500).json({ error: 'Failed to get run results' });
}
});
// ============================================================
// AUTO-FIX ENDPOINTS
// ============================================================
/**
* GET /api/system/fix/routines
* Get available fix routines
*/
router.get('/fix/routines', (_req: Request, res: Response) => {
try {
const routines = autoFix.getAvailableRoutines();
res.json(routines);
} catch (error) {
console.error('[System] Get routines error:', error);
res.status(500).json({ error: 'Failed to get routines' });
}
});
/**
* POST /api/system/fix/:routine
* Run a fix routine
*/
router.post('/fix/:routine', async (req: Request, res: Response) => {
try {
const routineName = req.params.routine;
const dryRun = req.body.dryRun === true;
const triggeredBy = req.body.triggeredBy || 'api';
const result = await autoFix.runRoutine(routineName as any, triggeredBy, { dryRun });
res.json(result);
} catch (error) {
console.error('[System] Fix routine error:', error);
res.status(500).json({ error: 'Failed to run fix routine' });
}
});
/**
* GET /api/system/fix/runs
* Get recent fix runs
*/
router.get('/fix/runs', async (req: Request, res: Response) => {
try {
const limit = req.query.limit ? parseInt(req.query.limit as string) : 20;
const runs = await autoFix.getRecentRuns(limit);
res.json(runs);
} catch (error) {
console.error('[System] Fix runs error:', error);
res.status(500).json({ error: 'Failed to get fix runs' });
}
});
// ============================================================
// ALERTS ENDPOINTS
// ============================================================
/**
* GET /api/system/alerts
* List alerts
*/
router.get('/alerts', async (req: Request, res: Response) => {
try {
const options = {
status: req.query.status as any,
severity: req.query.severity as any,
type: req.query.type as string,
limit: req.query.limit ? parseInt(req.query.limit as string) : 50,
offset: req.query.offset ? parseInt(req.query.offset as string) : 0,
};
const result = await alerts.listAlerts(options);
res.json(result);
} catch (error) {
console.error('[System] Alerts list error:', error);
res.status(500).json({ error: 'Failed to list alerts' });
}
});
/**
* GET /api/system/alerts/active
* Get active alerts
*/
router.get('/alerts/active', async (_req: Request, res: Response) => {
try {
const activeAlerts = await alerts.getActiveAlerts();
res.json(activeAlerts);
} catch (error) {
console.error('[System] Active alerts error:', error);
res.status(500).json({ error: 'Failed to get active alerts' });
}
});
/**
* GET /api/system/alerts/summary
* Get alert summary
*/
router.get('/alerts/summary', async (_req: Request, res: Response) => {
try {
const summary = await alerts.getSummary();
res.json(summary);
} catch (error) {
console.error('[System] Alerts summary error:', error);
res.status(500).json({ error: 'Failed to get alerts summary' });
}
});
/**
* POST /api/system/alerts/:id/acknowledge
* Acknowledge an alert
*/
router.post('/alerts/:id/acknowledge', async (req: Request, res: Response) => {
try {
const alertId = parseInt(req.params.id);
const acknowledgedBy = req.body.acknowledgedBy || 'api';
const success = await alerts.acknowledgeAlert(alertId, acknowledgedBy);
res.json({ success });
} catch (error) {
console.error('[System] Acknowledge alert error:', error);
res.status(500).json({ error: 'Failed to acknowledge alert' });
}
});
/**
* POST /api/system/alerts/:id/resolve
* Resolve an alert
*/
router.post('/alerts/:id/resolve', async (req: Request, res: Response) => {
try {
const alertId = parseInt(req.params.id);
const resolvedBy = req.body.resolvedBy || 'api';
const success = await alerts.resolveAlert(alertId, resolvedBy);
res.json({ success });
} catch (error) {
console.error('[System] Resolve alert error:', error);
res.status(500).json({ error: 'Failed to resolve alert' });
}
});
/**
* POST /api/system/alerts/bulk-acknowledge
* Bulk acknowledge alerts
*/
router.post('/alerts/bulk-acknowledge', async (req: Request, res: Response) => {
try {
const { ids, acknowledgedBy } = req.body;
if (!ids || !Array.isArray(ids)) {
return res.status(400).json({ error: 'ids array is required' });
}
const count = await alerts.bulkAcknowledge(ids, acknowledgedBy || 'api');
res.json({ acknowledged: count });
} catch (error) {
console.error('[System] Bulk acknowledge error:', error);
res.status(500).json({ error: 'Failed to bulk acknowledge' });
}
});
// ============================================================
// METRICS ENDPOINTS
// ============================================================
/**
* GET /api/system/metrics
* Get all current metrics
*/
router.get('/metrics', async (_req: Request, res: Response) => {
try {
const allMetrics = await metrics.getAllMetrics();
res.json(allMetrics);
} catch (error) {
console.error('[System] Metrics error:', error);
res.status(500).json({ error: 'Failed to get metrics' });
}
});
/**
* GET /api/system/metrics/:name
* Get a specific metric
*/
router.get('/metrics/:name', async (req: Request, res: Response) => {
try {
const metric = await metrics.getMetric(req.params.name);
if (!metric) {
return res.status(404).json({ error: 'Metric not found' });
}
res.json(metric);
} catch (error) {
console.error('[System] Metric error:', error);
res.status(500).json({ error: 'Failed to get metric' });
}
});
/**
* GET /api/system/metrics/:name/history
* Get metric time series
*/
router.get('/metrics/:name/history', async (req: Request, res: Response) => {
try {
const hours = req.query.hours ? parseInt(req.query.hours as string) : 24;
const history = await metrics.getMetricHistory(req.params.name, hours);
res.json(history);
} catch (error) {
console.error('[System] Metric history error:', error);
res.status(500).json({ error: 'Failed to get metric history' });
}
});
/**
* GET /api/system/errors
* Get error summary
*/
router.get('/errors', async (_req: Request, res: Response) => {
try {
const summary = await metrics.getErrorSummary();
res.json(summary);
} catch (error) {
console.error('[System] Error summary error:', error);
res.status(500).json({ error: 'Failed to get error summary' });
}
});
/**
* GET /api/system/errors/recent
* Get recent errors
*/
router.get('/errors/recent', async (req: Request, res: Response) => {
try {
const limit = req.query.limit ? parseInt(req.query.limit as string) : 50;
const errorType = req.query.type as string;
const errors = await metrics.getRecentErrors(limit, errorType);
res.json(errors);
} catch (error) {
console.error('[System] Recent errors error:', error);
res.status(500).json({ error: 'Failed to get recent errors' });
}
});
/**
* POST /api/system/errors/acknowledge
* Acknowledge errors
*/
router.post('/errors/acknowledge', async (req: Request, res: Response) => {
try {
const { ids, acknowledgedBy } = req.body;
if (!ids || !Array.isArray(ids)) {
return res.status(400).json({ error: 'ids array is required' });
}
const count = await metrics.acknowledgeErrors(ids, acknowledgedBy || 'api');
res.json({ acknowledged: count });
} catch (error) {
console.error('[System] Acknowledge errors error:', error);
res.status(500).json({ error: 'Failed to acknowledge errors' });
}
});
return router;
}
/**
* Create Prometheus metrics endpoint (standalone)
*/
export function createPrometheusRouter(pool: Pool): Router {
const router = Router();
const metrics = new MetricsService(pool);

View File

@@ -4,7 +4,7 @@
* Phase 5: Full Production Sync + Monitoring
*/
// SyncOrchestrator moved to _deprecated (depends on hydration module)
export { SyncOrchestrator, type SyncStatus, type QueueDepth, type SyncRunMetrics, type OrchestratorStatus } from './sync-orchestrator';
export { MetricsService, ERROR_TYPES, type Metric, type MetricTimeSeries, type ErrorBucket, type ErrorType } from './metrics';
export { DLQService, type DLQPayload, type DLQStats } from './dlq';
export { AlertService, type SystemAlert, type AlertSummary, type AlertSeverity, type AlertStatus } from './alerts';

View File

@@ -94,8 +94,7 @@ export async function handleEntryPointDiscovery(ctx: TaskContext): Promise<TaskR
// ============================================================
// STEP 3: Start stealth session
// ============================================================
// Per workflow-12102025.md: session identity comes from proxy location, not task params
const session = startSession();
const session = startSession(dispensary.state || 'AZ', 'America/Phoenix');
console.log(`[EntryPointDiscovery] Session started: ${session.sessionId}`);
try {

View File

@@ -4,9 +4,8 @@
* Exports all task handlers for the task worker.
*/
export { handleProductDiscovery } from './product-discovery';
export { handleProductRefresh } from './product-refresh';
export { handleProductDiscovery } from './product-discovery';
export { handleStoreDiscovery } from './store-discovery';
export { handleEntryPointDiscovery } from './entry-point-discovery';
export { handleAnalyticsRefresh } from './analytics-refresh';
export { handleWhoami } from './whoami';

View File

@@ -1,221 +0,0 @@
/**
* Payload Fetch Handler
*
* Per TASK_WORKFLOW_2024-12-10.md: Separates API fetch from data processing.
*
* This handler ONLY:
* 1. Hits Dutchie GraphQL API
* 2. Saves raw payload to filesystem (gzipped)
* 3. Records metadata in raw_crawl_payloads table
* 4. Queues a product_refresh task to process the payload
*
* Benefits of separation:
* - Retry-friendly: If normalize fails, re-run refresh without re-crawling
* - Faster refreshes: Local file read vs network call
* - Replay-able: Run refresh against any historical payload
* - Less API pressure: Only this role hits Dutchie
*/
import { TaskContext, TaskResult } from '../task-worker';
import {
executeGraphQL,
startSession,
endSession,
GRAPHQL_HASHES,
DUTCHIE_CONFIG,
} from '../../platforms/dutchie';
import { saveRawPayload } from '../../utils/payload-storage';
import { taskService } from '../task-service';
export async function handlePayloadFetch(ctx: TaskContext): Promise<TaskResult> {
const { pool, task } = ctx;
const dispensaryId = task.dispensary_id;
if (!dispensaryId) {
return { success: false, error: 'No dispensary_id specified for payload_fetch task' };
}
try {
// ============================================================
// STEP 1: Load dispensary info
// ============================================================
const dispResult = await pool.query(`
SELECT
id, name, platform_dispensary_id, menu_url, menu_type, city, state
FROM dispensaries
WHERE id = $1 AND crawl_enabled = true
`, [dispensaryId]);
if (dispResult.rows.length === 0) {
return { success: false, error: `Dispensary ${dispensaryId} not found or not crawl_enabled` };
}
const dispensary = dispResult.rows[0];
const platformId = dispensary.platform_dispensary_id;
if (!platformId) {
return { success: false, error: `Dispensary ${dispensaryId} has no platform_dispensary_id` };
}
// Extract cName from menu_url
const cNameMatch = dispensary.menu_url?.match(/\/(?:embedded-menu|dispensary)\/([^/?]+)/);
const cName = cNameMatch ? cNameMatch[1] : 'dispensary';
console.log(`[PayloadFetch] Starting fetch for ${dispensary.name} (ID: ${dispensaryId})`);
console.log(`[PayloadFetch] Platform ID: ${platformId}, cName: ${cName}`);
// ============================================================
// STEP 2: Start stealth session
// ============================================================
const session = startSession();
console.log(`[PayloadFetch] Session started: ${session.sessionId}`);
await ctx.heartbeat();
// ============================================================
// STEP 3: Fetch products via GraphQL (Status: 'All')
// ============================================================
const allProducts: any[] = [];
let page = 0;
let totalCount = 0;
const perPage = DUTCHIE_CONFIG.perPage;
const maxPages = DUTCHIE_CONFIG.maxPages;
try {
while (page < maxPages) {
const variables = {
includeEnterpriseSpecials: false,
productsFilter: {
dispensaryId: platformId,
pricingType: 'rec',
Status: 'All',
types: [],
useCache: false,
isDefaultSort: true,
sortBy: 'popularSortIdx',
sortDirection: 1,
bypassOnlineThresholds: true,
isKioskMenu: false,
removeProductsBelowOptionThresholds: false,
},
page,
perPage,
};
console.log(`[PayloadFetch] Fetching page ${page + 1}...`);
const result = await executeGraphQL(
'FilteredProducts',
variables,
GRAPHQL_HASHES.FilteredProducts,
{ cName, maxRetries: 3 }
);
const data = result?.data?.filteredProducts;
if (!data || !data.products) {
if (page === 0) {
throw new Error('No product data returned from GraphQL');
}
break;
}
const products = data.products;
allProducts.push(...products);
if (page === 0) {
totalCount = data.queryInfo?.totalCount || products.length;
console.log(`[PayloadFetch] Total products reported: ${totalCount}`);
}
if (allProducts.length >= totalCount || products.length < perPage) {
break;
}
page++;
if (page < maxPages) {
await new Promise(r => setTimeout(r, DUTCHIE_CONFIG.pageDelayMs));
}
if (page % 5 === 0) {
await ctx.heartbeat();
}
}
console.log(`[PayloadFetch] Fetched ${allProducts.length} products in ${page + 1} pages`);
} finally {
endSession();
}
if (allProducts.length === 0) {
return {
success: false,
error: 'No products returned from GraphQL',
productsProcessed: 0,
};
}
await ctx.heartbeat();
// ============================================================
// STEP 4: Save raw payload to filesystem
// Per TASK_WORKFLOW_2024-12-10.md: Metadata/Payload separation
// ============================================================
const rawPayload = {
dispensaryId,
platformId,
cName,
fetchedAt: new Date().toISOString(),
productCount: allProducts.length,
products: allProducts,
};
const payloadResult = await saveRawPayload(
pool,
dispensaryId,
rawPayload,
null, // crawl_run_id - not using crawl_runs in new system
allProducts.length
);
console.log(`[PayloadFetch] Saved payload #${payloadResult.id} (${(payloadResult.sizeBytes / 1024).toFixed(1)}KB)`);
// ============================================================
// STEP 5: Update dispensary last_fetch_at
// ============================================================
await pool.query(`
UPDATE dispensaries
SET last_fetch_at = NOW()
WHERE id = $1
`, [dispensaryId]);
// ============================================================
// STEP 6: Queue product_refresh task to process the payload
// Per TASK_WORKFLOW_2024-12-10.md: Task chaining
// ============================================================
await taskService.createTask({
role: 'product_refresh',
dispensary_id: dispensaryId,
priority: task.priority || 0,
payload: { payload_id: payloadResult.id },
});
console.log(`[PayloadFetch] Queued product_refresh task for payload #${payloadResult.id}`);
return {
success: true,
payloadId: payloadResult.id,
productCount: allProducts.length,
sizeBytes: payloadResult.sizeBytes,
};
} catch (error: unknown) {
const errorMessage = error instanceof Error ? error.message : 'Unknown error';
console.error(`[PayloadFetch] Error for dispensary ${dispensaryId}:`, errorMessage);
return {
success: false,
error: errorMessage,
};
}
}

View File

@@ -1,31 +1,16 @@
/**
* Product Discovery Handler
*
* Per TASK_WORKFLOW_2024-12-10.md: Initial product fetch for newly discovered stores.
*
* Flow:
* 1. Triggered after store_discovery promotes a new dispensary
* 2. Chains to payload_fetch to get initial product data
* 3. payload_fetch chains to product_refresh for DB upsert
*
* Chaining:
* store_discovery → (newStoreIds) → product_discovery → payload_fetch → product_refresh
* Initial product fetch for stores that have 0 products.
* Same logic as product_resync, but for initial discovery.
*/
import { TaskContext, TaskResult } from '../task-worker';
import { handlePayloadFetch } from './payload-fetch';
import { handleProductRefresh } from './product-refresh';
export async function handleProductDiscovery(ctx: TaskContext): Promise<TaskResult> {
const { task } = ctx;
const dispensaryId = task.dispensary_id;
if (!dispensaryId) {
return { success: false, error: 'No dispensary_id provided' };
}
console.log(`[ProductDiscovery] Starting initial product discovery for dispensary ${dispensaryId}`);
// Per TASK_WORKFLOW_2024-12-10.md: Chain to payload_fetch for API → disk
// payload_fetch will then chain to product_refresh for disk → DB
return handlePayloadFetch(ctx);
// Product discovery is essentially the same as refresh for the first time
// The main difference is in when this task is triggered (new store vs scheduled)
console.log(`[ProductDiscovery] Starting initial product fetch for dispensary ${ctx.task.dispensary_id}`);
return handleProductRefresh(ctx);
}

View File

@@ -1,32 +1,33 @@
/**
* Product Refresh Handler
*
* Per TASK_WORKFLOW_2024-12-10.md: Processes a locally-stored payload.
*
* This handler reads from the filesystem (NOT the Dutchie API).
* The payload_fetch handler is responsible for API calls.
* Re-crawls a store to capture price/stock changes using the GraphQL pipeline.
*
* Flow:
* 1. Load payload from filesystem (by payload_id or latest for dispensary)
* 2. Normalize data via DutchieNormalizer
* 3. Upsert to store_products and store_product_snapshots
* 4. Track missing products (increment consecutive_misses, mark OOS at 3)
* 5. Download new product images
*
* Benefits of separation:
* - Retry-friendly: If this fails, re-run without re-crawling
* - Replay-able: Run against any historical payload
* - Faster: Local file read vs network call
* 1. Load dispensary info from database
* 2. Start stealth session (fingerprint + optional proxy)
* 3. Fetch products via GraphQL (Status: 'All')
* 4. Normalize data via DutchieNormalizer
* 5. Upsert to store_products and store_product_snapshots
* 6. Track missing products (increment consecutive_misses, mark OOS at 3)
* 7. Download new product images
* 8. End session
*/
import { TaskContext, TaskResult } from '../task-worker';
import {
executeGraphQL,
startSession,
endSession,
GRAPHQL_HASHES,
DUTCHIE_CONFIG,
} from '../../platforms/dutchie';
import { DutchieNormalizer } from '../../hydration/normalizers/dutchie';
import {
upsertStoreProducts,
createStoreProductSnapshots,
downloadProductImages,
} from '../../hydration/canonical-upsert';
import { loadRawPayloadById, getLatestPayload } from '../../utils/payload-storage';
const normalizer = new DutchieNormalizer();
@@ -46,76 +47,129 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
SELECT
id, name, platform_dispensary_id, menu_url, menu_type, city, state
FROM dispensaries
WHERE id = $1
WHERE id = $1 AND crawl_enabled = true
`, [dispensaryId]);
if (dispResult.rows.length === 0) {
return { success: false, error: `Dispensary ${dispensaryId} not found` };
return { success: false, error: `Dispensary ${dispensaryId} not found or not crawl_enabled` };
}
const dispensary = dispResult.rows[0];
const platformId = dispensary.platform_dispensary_id;
// Extract cName from menu_url for image storage context
if (!platformId) {
return { success: false, error: `Dispensary ${dispensaryId} has no platform_dispensary_id` };
}
// Extract cName from menu_url
const cNameMatch = dispensary.menu_url?.match(/\/(?:embedded-menu|dispensary)\/([^/?]+)/);
const cName = cNameMatch ? cNameMatch[1] : 'dispensary';
console.log(`[ProductRefresh] Starting refresh for ${dispensary.name} (ID: ${dispensaryId})`);
console.log(`[ProductResync] Starting crawl for ${dispensary.name} (ID: ${dispensaryId})`);
console.log(`[ProductResync] Platform ID: ${platformId}, cName: ${cName}`);
// ============================================================
// STEP 2: Start stealth session
// ============================================================
const session = startSession(dispensary.state || 'AZ', 'America/Phoenix');
console.log(`[ProductResync] Session started: ${session.sessionId}`);
await ctx.heartbeat();
// ============================================================
// STEP 2: Load payload from filesystem
// Per TASK_WORKFLOW_2024-12-10.md: Read local payload, not API
// STEP 3: Fetch products via GraphQL (Status: 'All')
// ============================================================
let payloadData: any;
let payloadId: number;
const allProducts: any[] = [];
let page = 0;
let totalCount = 0;
const perPage = DUTCHIE_CONFIG.perPage;
const maxPages = DUTCHIE_CONFIG.maxPages;
// Check if specific payload_id was provided (from task chaining)
const taskPayload = task.payload as { payload_id?: number } | null;
try {
while (page < maxPages) {
const variables = {
includeEnterpriseSpecials: false,
productsFilter: {
dispensaryId: platformId,
pricingType: 'rec',
Status: 'All',
types: [],
useCache: false,
isDefaultSort: true,
sortBy: 'popularSortIdx',
sortDirection: 1,
bypassOnlineThresholds: true,
isKioskMenu: false,
removeProductsBelowOptionThresholds: false,
},
page,
perPage,
};
if (taskPayload?.payload_id) {
// Load specific payload (from payload_fetch chaining)
const result = await loadRawPayloadById(pool, taskPayload.payload_id);
if (!result) {
return { success: false, error: `Payload ${taskPayload.payload_id} not found` };
console.log(`[ProductResync] Fetching page ${page + 1}...`);
const result = await executeGraphQL(
'FilteredProducts',
variables,
GRAPHQL_HASHES.FilteredProducts,
{ cName, maxRetries: 3 }
);
const data = result?.data?.filteredProducts;
if (!data || !data.products) {
if (page === 0) {
throw new Error('No product data returned from GraphQL');
}
payloadData = result.payload;
payloadId = result.metadata.id;
console.log(`[ProductRefresh] Loaded specific payload #${payloadId}`);
} else {
// Load latest payload for this dispensary
const result = await getLatestPayload(pool, dispensaryId);
if (!result) {
return { success: false, error: `No payload found for dispensary ${dispensaryId}` };
}
payloadData = result.payload;
payloadId = result.metadata.id;
console.log(`[ProductRefresh] Loaded latest payload #${payloadId} (${result.metadata.fetchedAt})`);
break;
}
const allProducts = payloadData.products || [];
const products = data.products;
allProducts.push(...products);
if (page === 0) {
totalCount = data.queryInfo?.totalCount || products.length;
console.log(`[ProductResync] Total products reported: ${totalCount}`);
}
if (allProducts.length >= totalCount || products.length < perPage) {
break;
}
page++;
if (page < maxPages) {
await new Promise(r => setTimeout(r, DUTCHIE_CONFIG.pageDelayMs));
}
if (page % 5 === 0) {
await ctx.heartbeat();
}
}
console.log(`[ProductResync] Fetched ${allProducts.length} products in ${page + 1} pages`);
} finally {
endSession();
}
if (allProducts.length === 0) {
return {
success: false,
error: 'Payload contains no products',
payloadId,
error: 'No products returned from GraphQL',
productsProcessed: 0,
};
}
console.log(`[ProductRefresh] Processing ${allProducts.length} products from payload #${payloadId}`);
await ctx.heartbeat();
// ============================================================
// STEP 3: Normalize data
// STEP 4: Normalize data
// ============================================================
console.log(`[ProductRefresh] Normalizing ${allProducts.length} products...`);
console.log(`[ProductResync] Normalizing ${allProducts.length} products...`);
// Build RawPayload for the normalizer
const rawPayload = {
id: `refresh-${dispensaryId}-${Date.now()}`,
id: `resync-${dispensaryId}-${Date.now()}`,
dispensary_id: dispensaryId,
crawl_run_id: null,
platform: 'dutchie',
@@ -135,26 +189,25 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
const normalizationResult = normalizer.normalize(rawPayload);
if (normalizationResult.errors.length > 0) {
console.warn(`[ProductRefresh] Normalization warnings: ${normalizationResult.errors.map(e => e.message).join(', ')}`);
console.warn(`[ProductResync] Normalization warnings: ${normalizationResult.errors.map(e => e.message).join(', ')}`);
}
if (normalizationResult.products.length === 0) {
return {
success: false,
error: 'Normalization produced no products',
payloadId,
productsProcessed: 0,
};
}
console.log(`[ProductRefresh] Normalized ${normalizationResult.products.length} products`);
console.log(`[ProductResync] Normalized ${normalizationResult.products.length} products`);
await ctx.heartbeat();
// ============================================================
// STEP 4: Upsert to canonical tables
// STEP 5: Upsert to canonical tables
// ============================================================
console.log(`[ProductRefresh] Upserting to store_products...`);
console.log(`[ProductResync] Upserting to store_products...`);
const upsertResult = await upsertStoreProducts(
pool,
@@ -163,12 +216,12 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
normalizationResult.availability
);
console.log(`[ProductRefresh] Upserted: ${upsertResult.upserted} (${upsertResult.new} new, ${upsertResult.updated} updated)`);
console.log(`[ProductResync] Upserted: ${upsertResult.upserted} (${upsertResult.new} new, ${upsertResult.updated} updated)`);
await ctx.heartbeat();
// Create snapshots
console.log(`[ProductRefresh] Creating snapshots...`);
console.log(`[ProductResync] Creating snapshots...`);
const snapshotsResult = await createStoreProductSnapshots(
pool,
@@ -179,12 +232,12 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
null // No crawl_run_id in new system
);
console.log(`[ProductRefresh] Created ${snapshotsResult.created} snapshots`);
console.log(`[ProductResync] Created ${snapshotsResult.created} snapshots`);
await ctx.heartbeat();
// ============================================================
// STEP 5: Track missing products (consecutive_misses logic)
// STEP 6: Track missing products (consecutive_misses logic)
// - Products in feed: reset consecutive_misses to 0
// - Products not in feed: increment consecutive_misses
// - At 3 consecutive misses: mark as OOS
@@ -217,7 +270,7 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
const incrementedCount = incrementResult.rowCount || 0;
if (incrementedCount > 0) {
console.log(`[ProductRefresh] Incremented consecutive_misses for ${incrementedCount} products`);
console.log(`[ProductResync] Incremented consecutive_misses for ${incrementedCount} products`);
}
// Mark as OOS any products that hit 3 consecutive misses
@@ -233,16 +286,16 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
const markedOosCount = oosResult.rowCount || 0;
if (markedOosCount > 0) {
console.log(`[ProductRefresh] Marked ${markedOosCount} products as OOS (3+ consecutive misses)`);
console.log(`[ProductResync] Marked ${markedOosCount} products as OOS (3+ consecutive misses)`);
}
await ctx.heartbeat();
// ============================================================
// STEP 6: Download images for new products
// STEP 7: Download images for new products
// ============================================================
if (upsertResult.productsNeedingImages.length > 0) {
console.log(`[ProductRefresh] Downloading images for ${upsertResult.productsNeedingImages.length} products...`);
console.log(`[ProductResync] Downloading images for ${upsertResult.productsNeedingImages.length} products...`);
try {
const dispensaryContext = {
@@ -256,12 +309,12 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
);
} catch (imgError: any) {
// Image download errors shouldn't fail the whole task
console.warn(`[ProductRefresh] Image download error (non-fatal): ${imgError.message}`);
console.warn(`[ProductResync] Image download error (non-fatal): ${imgError.message}`);
}
}
// ============================================================
// STEP 7: Update dispensary last_crawl_at
// STEP 8: Update dispensary last_crawl_at
// ============================================================
await pool.query(`
UPDATE dispensaries
@@ -269,20 +322,10 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
WHERE id = $1
`, [dispensaryId]);
// ============================================================
// STEP 8: Mark payload as processed
// ============================================================
await pool.query(`
UPDATE raw_crawl_payloads
SET processed_at = NOW()
WHERE id = $1
`, [payloadId]);
console.log(`[ProductRefresh] Completed ${dispensary.name}`);
console.log(`[ProductResync] Completed ${dispensary.name}`);
return {
success: true,
payloadId,
productsProcessed: normalizationResult.products.length,
snapshotsCreated: snapshotsResult.created,
newProducts: upsertResult.new,
@@ -292,7 +335,7 @@ export async function handleProductRefresh(ctx: TaskContext): Promise<TaskResult
} catch (error: unknown) {
const errorMessage = error instanceof Error ? error.message : 'Unknown error';
console.error(`[ProductRefresh] Error for dispensary ${dispensaryId}:`, errorMessage);
console.error(`[ProductResync] Error for dispensary ${dispensaryId}:`, errorMessage);
return {
success: false,
error: errorMessage,

View File

@@ -1,16 +1,8 @@
/**
* Store Discovery Handler
*
* Per TASK_WORKFLOW_2024-12-10.md: Discovers new stores and returns their IDs for task chaining.
*
* Flow:
* 1. For each active state, run Dutchie discovery
* 2. Discover locations via GraphQL
* 3. Auto-promote valid locations to dispensaries table
* 4. Return newStoreIds[] for chaining to payload_fetch
*
* Chaining:
* store_discovery → (returns newStoreIds) → payload_fetch → product_refresh
* Discovers new stores by crawling location APIs and adding them
* to discovery_locations table.
*/
import { TaskContext, TaskResult } from '../task-worker';
@@ -18,25 +10,23 @@ import { discoverState } from '../../discovery';
export async function handleStoreDiscovery(ctx: TaskContext): Promise<TaskResult> {
const { pool, task } = ctx;
const platform = task.platform || 'dutchie';
const platform = task.platform || 'default';
console.log(`[StoreDiscovery] Starting discovery for platform: ${platform}`);
try {
// Get states to discover
const statesResult = await pool.query(`
SELECT code FROM states WHERE is_active = true ORDER BY code
SELECT code FROM states WHERE active = true ORDER BY code
`);
const stateCodes = statesResult.rows.map(r => r.code);
if (stateCodes.length === 0) {
return { success: true, storesDiscovered: 0, newStoreIds: [], message: 'No active states to discover' };
return { success: true, storesDiscovered: 0, message: 'No active states to discover' };
}
let totalDiscovered = 0;
let totalPromoted = 0;
// Per TASK_WORKFLOW_2024-12-10.md: Collect all new store IDs for task chaining
const allNewStoreIds: number[] = [];
// Run discovery for each state
for (const stateCode of stateCodes) {
@@ -49,13 +39,6 @@ export async function handleStoreDiscovery(ctx: TaskContext): Promise<TaskResult
const result = await discoverState(pool, stateCode);
totalDiscovered += result.totalLocationsFound || 0;
totalPromoted += result.totalLocationsUpserted || 0;
// Per TASK_WORKFLOW_2024-12-10.md: Collect new IDs for chaining
if (result.newDispensaryIds && result.newDispensaryIds.length > 0) {
allNewStoreIds.push(...result.newDispensaryIds);
console.log(`[StoreDiscovery] ${stateCode}: ${result.newDispensaryIds.length} new stores`);
}
console.log(`[StoreDiscovery] ${stateCode}: found ${result.totalLocationsFound}, upserted ${result.totalLocationsUpserted}`);
} catch (error: unknown) {
const errorMessage = error instanceof Error ? error.message : 'Unknown error';
@@ -64,15 +47,13 @@ export async function handleStoreDiscovery(ctx: TaskContext): Promise<TaskResult
}
}
console.log(`[StoreDiscovery] Complete: ${totalDiscovered} discovered, ${totalPromoted} promoted, ${allNewStoreIds.length} new stores`);
console.log(`[StoreDiscovery] Complete: ${totalDiscovered} discovered, ${totalPromoted} promoted`);
return {
success: true,
storesDiscovered: totalDiscovered,
storesPromoted: totalPromoted,
statesProcessed: stateCodes.length,
// Per TASK_WORKFLOW_2024-12-10.md: Return new IDs for task chaining
newStoreIds: allNewStoreIds,
};
} catch (error: unknown) {
const errorMessage = error instanceof Error ? error.message : 'Unknown error';
@@ -80,7 +61,6 @@ export async function handleStoreDiscovery(ctx: TaskContext): Promise<TaskResult
return {
success: false,
error: errorMessage,
newStoreIds: [],
};
}
}

View File

@@ -1,80 +0,0 @@
/**
* WhoAmI Handler
* Tests proxy connectivity and anti-detect by fetching public IP
* Reports: proxy IP, fingerprint info, and connection status
*/
import { TaskContext, TaskResult } from '../task-worker';
import { execSync } from 'child_process';
export async function handleWhoami(ctx: TaskContext): Promise<TaskResult> {
const { pool, crawlRotator } = ctx;
console.log('[WhoAmI] Testing proxy and anti-detect...');
try {
// Use the preflight check which tests proxy + anti-detect
if (crawlRotator) {
const preflight = await crawlRotator.preflight();
if (!preflight.passed) {
return {
success: false,
error: preflight.error || 'Preflight check failed',
proxyAvailable: preflight.proxyAvailable,
proxyConnected: preflight.proxyConnected,
antidetectReady: preflight.antidetectReady,
};
}
console.log(`[WhoAmI] Proxy IP: ${preflight.proxyIp}, Response: ${preflight.responseTimeMs}ms`);
console.log(`[WhoAmI] Fingerprint: ${preflight.fingerprint?.browserName}/${preflight.fingerprint?.deviceCategory}`);
return {
success: true,
proxyIp: preflight.proxyIp,
responseTimeMs: preflight.responseTimeMs,
fingerprint: preflight.fingerprint,
proxyAvailable: preflight.proxyAvailable,
proxyConnected: preflight.proxyConnected,
antidetectReady: preflight.antidetectReady,
};
}
// Fallback: Direct proxy test without CrawlRotator
const proxyResult = await pool.query(`
SELECT host, port, username, password
FROM proxies
WHERE is_active = true
LIMIT 1
`);
if (proxyResult.rows.length === 0) {
return { success: false, error: 'No active proxy configured' };
}
const p = proxyResult.rows[0];
const proxyUrl = p.username
? `http://${p.username}:${p.password}@${p.host}:${p.port}`
: `http://${p.host}:${p.port}`;
console.log(`[WhoAmI] Using proxy: ${p.host}:${p.port}`);
// Fetch IP via proxy
const cmd = `curl -s --proxy '${proxyUrl}' 'https://api.ipify.org?format=json'`;
const output = execSync(cmd, { timeout: 30000 }).toString().trim();
const data = JSON.parse(output);
console.log(`[WhoAmI] Proxy IP: ${data.ip}`);
return {
success: true,
proxyIp: data.ip,
proxyHost: p.host,
proxyPort: p.port,
};
} catch (error: any) {
console.error('[WhoAmI] Error:', error.message);
return { success: false, error: error.message };
}
}

View File

@@ -17,8 +17,8 @@ export {
export { TaskWorker, TaskContext, TaskResult } from './task-worker';
export {
handleProductDiscovery,
handleProductRefresh,
handleProductDiscovery,
handleStoreDiscovery,
handleEntryPointDiscovery,
handleAnalyticsRefresh,

View File

@@ -1,37 +0,0 @@
/**
* Task Pool State
*
* Shared state for task pool pause/resume functionality.
* This is kept separate to avoid circular dependencies between
* task-service.ts and routes/tasks.ts.
*
* State is in-memory and resets on server restart.
* By default, the pool is PAUSED (closed) - admin must explicitly start it.
* This prevents workers from immediately grabbing tasks on deploy before
* the system is ready.
*/
let taskPoolPaused = true;
export function isTaskPoolPaused(): boolean {
return taskPoolPaused;
}
export function pauseTaskPool(): void {
taskPoolPaused = true;
console.log('[TaskPool] Task pool PAUSED - workers will not pick up new tasks');
}
export function resumeTaskPool(): void {
taskPoolPaused = false;
console.log('[TaskPool] Task pool RESUMED - workers can pick up tasks');
}
export function getTaskPoolStatus(): { paused: boolean; message: string } {
return {
paused: taskPoolPaused,
message: taskPoolPaused
? 'Task pool is paused - workers will not pick up new tasks'
: 'Task pool is open - workers are picking up tasks',
};
}

View File

@@ -9,7 +9,6 @@
*/
import { pool } from '../db/pool';
import { isTaskPoolPaused } from './task-pool-state';
// Helper to check if a table exists
async function tableExists(tableName: string): Promise<boolean> {
@@ -22,18 +21,12 @@ async function tableExists(tableName: string): Promise<boolean> {
return result.rows[0].exists;
}
// Per TASK_WORKFLOW_2024-12-10.md: Task roles
// payload_fetch: Hits Dutchie API, saves raw payload to filesystem
// product_discovery: Main product crawl handler
// product_refresh: Legacy role (deprecated but kept for compatibility)
export type TaskRole =
| 'store_discovery'
| 'entry_point_discovery'
| 'product_discovery'
| 'payload_fetch' // Fetches from API, saves to disk
| 'product_refresh' // DEPRECATED: Use product_discovery instead
| 'analytics_refresh'
| 'whoami'; // Tests proxy + anti-detect connectivity
| 'product_refresh'
| 'analytics_refresh';
export type TaskStatus =
| 'pending'
@@ -52,7 +45,6 @@ export interface WorkerTask {
platform: string | null;
status: TaskStatus;
priority: number;
method: 'curl' | 'http' | null; // Transport method: curl=axios/proxy, http=Puppeteer/browser
scheduled_for: Date | null;
worker_id: string | null;
claimed_at: Date | null;
@@ -63,7 +55,6 @@ export interface WorkerTask {
error_message: string | null;
retry_count: number;
max_retries: number;
payload: Record<string, unknown> | null; // Per TASK_WORKFLOW_2024-12-10.md: Task chaining data
created_at: Date;
updated_at: Date;
}
@@ -74,7 +65,6 @@ export interface CreateTaskParams {
platform?: string;
priority?: number;
scheduled_for?: Date;
payload?: Record<string, unknown>; // Per TASK_WORKFLOW_2024-12-10.md: For task chaining data
}
export interface CapacityMetrics {
@@ -106,8 +96,8 @@ class TaskService {
*/
async createTask(params: CreateTaskParams): Promise<WorkerTask> {
const result = await pool.query(
`INSERT INTO worker_tasks (role, dispensary_id, platform, priority, scheduled_for, payload)
VALUES ($1, $2, $3, $4, $5, $6)
`INSERT INTO worker_tasks (role, dispensary_id, platform, priority, scheduled_for)
VALUES ($1, $2, $3, $4, $5)
RETURNING *`,
[
params.role,
@@ -115,7 +105,6 @@ class TaskService {
params.platform ?? null,
params.priority ?? 0,
params.scheduled_for ?? null,
params.payload ? JSON.stringify(params.payload) : null,
]
);
return result.rows[0] as WorkerTask;
@@ -153,34 +142,18 @@ class TaskService {
/**
* Claim a task atomically for a worker
* If role is null, claims ANY available task (role-agnostic worker)
* Returns null if task pool is paused.
*
* @param role - Task role to claim, or null for any task
* @param workerId - Worker ID claiming the task
* @param curlPassed - Whether worker passed curl preflight (default true for backward compat)
* @param httpPassed - Whether worker passed http/Puppeteer preflight (default false)
*/
async claimTask(
role: TaskRole | null,
workerId: string,
curlPassed: boolean = true,
httpPassed: boolean = false
): Promise<WorkerTask | null> {
// Check if task pool is paused - don't claim any tasks
if (isTaskPoolPaused()) {
return null;
}
async claimTask(role: TaskRole | null, workerId: string): Promise<WorkerTask | null> {
if (role) {
// Role-specific claiming - use the SQL function with preflight capabilities
// Role-specific claiming - use the SQL function
const result = await pool.query(
`SELECT * FROM claim_task($1, $2, $3, $4)`,
[role, workerId, curlPassed, httpPassed]
`SELECT * FROM claim_task($1, $2)`,
[role, workerId]
);
return (result.rows[0] as WorkerTask) || null;
}
// Role-agnostic claiming - claim ANY pending task matching worker capabilities
// Role-agnostic claiming - claim ANY pending task
const result = await pool.query(`
UPDATE worker_tasks
SET
@@ -191,12 +164,6 @@ class TaskService {
SELECT id FROM worker_tasks
WHERE status = 'pending'
AND (scheduled_for IS NULL OR scheduled_for <= NOW())
-- Method compatibility: worker must have passed the required preflight
AND (
method IS NULL -- No preference, any worker can claim
OR (method = 'curl' AND $2 = TRUE)
OR (method = 'http' AND $3 = TRUE)
)
-- Exclude stores that already have an active task
AND (dispensary_id IS NULL OR dispensary_id NOT IN (
SELECT dispensary_id FROM worker_tasks
@@ -208,7 +175,7 @@ class TaskService {
FOR UPDATE SKIP LOCKED
)
RETURNING *
`, [workerId, curlPassed, httpPassed]);
`, [workerId]);
return (result.rows[0] as WorkerTask) || null;
}
@@ -249,24 +216,6 @@ class TaskService {
);
}
/**
* Release a claimed task back to pending (e.g., when preflight fails)
* This allows another worker to pick it up.
*/
async releaseTask(taskId: number): Promise<void> {
await pool.query(
`UPDATE worker_tasks
SET status = 'pending',
worker_id = NULL,
claimed_at = NULL,
started_at = NULL,
updated_at = NOW()
WHERE id = $1 AND status IN ('claimed', 'running')`,
[taskId]
);
console.log(`[TaskService] Task ${taskId} released back to pending`);
}
/**
* Mark a task as failed, with auto-retry if under max_retries
* Returns true if task was re-queued for retry, false if permanently failed
@@ -452,17 +401,6 @@ class TaskService {
/**
* Chain next task after completion
* Called automatically when a task completes successfully
*
* Per TASK_WORKFLOW_2024-12-10.md: Task chaining flow:
*
* Discovery flow (new stores):
* store_discovery → product_discovery → payload_fetch → product_refresh
*
* Scheduled flow (existing stores):
* payload_fetch → product_refresh
*
* Note: entry_point_discovery is deprecated since platform_dispensary_id
* is now resolved during store promotion.
*/
async chainNextTask(completedTask: WorkerTask): Promise<WorkerTask | null> {
if (completedTask.status !== 'completed') {
@@ -471,14 +409,12 @@ class TaskService {
switch (completedTask.role) {
case 'store_discovery': {
// Per TASK_WORKFLOW_2024-12-10.md: New stores discovered -> create product_discovery tasks
// Skip entry_point_discovery since platform_dispensary_id is set during promotion
// New stores discovered -> create entry_point_discovery tasks
const newStoreIds = (completedTask.result as { newStoreIds?: number[] })?.newStoreIds;
if (newStoreIds && newStoreIds.length > 0) {
console.log(`[TaskService] Chaining ${newStoreIds.length} product_discovery tasks for new stores`);
for (const storeId of newStoreIds) {
await this.createTask({
role: 'product_discovery',
role: 'entry_point_discovery',
dispensary_id: storeId,
platform: completedTask.platform ?? undefined,
priority: 10, // High priority for new stores
@@ -489,8 +425,7 @@ class TaskService {
}
case 'entry_point_discovery': {
// DEPRECATED: Entry point resolution now happens during store promotion
// Kept for backward compatibility with any in-flight tasks
// Entry point resolved -> create product_discovery task
const success = (completedTask.result as { success?: boolean })?.success;
if (success && completedTask.dispensary_id) {
return this.createTask({
@@ -504,15 +439,8 @@ class TaskService {
}
case 'product_discovery': {
// Per TASK_WORKFLOW_2024-12-10.md: Product discovery chains internally to payload_fetch
// No external chaining needed - handleProductDiscovery calls handlePayloadFetch directly
break;
}
case 'payload_fetch': {
// Per TASK_WORKFLOW_2024-12-10.md: payload_fetch chains to product_refresh
// This is handled internally by the payload_fetch handler via taskService.createTask
// No external chaining needed here
// Product discovery done -> store is now ready for regular resync
// No immediate chaining needed; will be picked up by daily batch generation
break;
}
}

Some files were not shown because too many files have changed in this diff Show More