fix:fetch-page-image-urls-for-existing-records

fix:populate-page-image-urls-in-fetch-latest-endpoint

feat: scrape all cities every 2h add /api/scrape-all cron endpoint enable Sulphur Springs

fix: add asyncio timeouts to prevent Vercel 502 on slow Laserfiche fetches

fix: 502/503 errors, add Google Vision OCR, improve geo-locate robustness

fix: OCR async, connector fetch_page_images, lambda closures, parsers init

fix: force re-summarize with force parameter, remove broken _execute import

feat: add force parameter to fetch-latest to clear cached summaries

fix: reuse event loop across requests in ASGI bridge, make OCR.space sync to avoid loop issues

fix: combine ASGI-to-WSGI bridge with direct lifespan init for connectors/DB/summarizer

fix: use Mangum lifespan=auto so DB, connectors, and summarizer initialize on first request

fix: replace Mangum with custom ASGI-to-WSGI bridge for Vercel experimentalServices

fix: set Mangum handler to Handler.HTTP explicitly for Vercel experimentalServices

fix: absolute minimal WSGI entry point - imports only inside handler

fix: replace psycopg2-binary with pg8000 (pure Python) to avoid Vercel Lambda 250MB native dep crash

fix: make Vercel entry point lazy-load FastAPI on first request to prevent cold-start crash

fix: wrap lifespan startup in try/except so health endpoint works even if DB init fails

feat: Vercel-ready OpenCouncil with 3 cities connected (Paris, Sulphur Springs, Frisco)

Implement Three P's framework: Life-Centric categories, Big Picture TL;DR, What You Can Do

Add /api/minutes/seed endpoint for pushing pre-computed OCR data to HF Space PostgreSQL

fix: make OCR work always and consistently

fix: add direct page image download fallback when PDF generation fails on HF Spaces

fix: extract page_image_urls from raw_text to avoid second HTTP request

fix: OCR pipeline not receiving page images for summarization

Fix DeepSeek hallucination: draconian prompt, temperature 0.0, save OCR text to raw_text

Fix: reset DB endpoint, filter future dates, prevent DeepSeek hallucination

Replace GPT-4o Vision with free Tesseract OCR pipeline

Add GPT-4o Vision support for scanned document images; remove all agenda code; focus on minutes

Debug: use model_dump(mode='json') to bypass Pydantic serialization, add traceback

Fix 500 error: wrap DB lookups in try/except, add NULL-safe json.loads and datetime parsing in _row_to_agenda and _row_to_summary

Fix agenda detail 500 error: deterministic IDs via SHA-256 hash, fix timezone-naive datetime fallback, add better error logging

Add minutes support: fetch, store, summarize city council minutes from Laserfiche

fix: report actual storage backend in health endpoint (was hardcoded to sqlite)

fix: update comments from Railway to HF Spaces, fix DB init print message

Switch from experimental services to builds/routes approach, fix API_BASE to empty string

Fix Vercel backend 500 errors

Add Vercel experimental services monorepo config (frontend + backend)

Initial commit: Civic City Hub - DeepSeek API, SQLite storage, encrypted API keys, professional news site frontend