Commit History

  1. fix: Handle N-to-M word replacements in analyze post-filter - Process each original word individually against corrected segment - Accept 1-to-1 small edits and 1-to-N word splits within multi-word blocks - Previously all multi-word blocks were silently rejected

    d1f18ad
  2. fix: Relax analyze post-filters to show AraSpell corrections in editor - Allow 3 char edits (was 2), 50% ratio (was 40%) - Lower word-split threshold from 12 to 5 chars

    6d70738
  3. feat: Wire /api/analyze to use AraSpell service for real-time editor spelling

    bf32528
  4. fix: Remove duplicate /api/spelling route causing AssertionError

    738185d
  5. feat: Enable ContextualCorrector (BERT MLM) for spelling - Pre-download AutoModelForMaskedLM in Dockerfile

    bcf3607
  6. feat: Add AraSpell spelling correction endpoint (POST /api/spelling) with lazy model loading - Extract AraSpell pipeline into src/nlp/spelling/ (araspell_rules.py + araspell_service.py) - Pre-download AraBERT encoder-decoder + checkpoint in Dockerfile - Lazy load on first request, contextual=False to save RAM - Update health check to report spelling status

    53f433b
  7. fix: Always load models on import (RENDER/GUNICORN_LOADED env vars not set in HF Spaces)

    ccf3eef
  8. debug: Add startup error tracking + memory info to debug endpoint

    756045d
  9. fix: Pre-download summarization model during Docker build (runtime has no DNS)

    8060095
  10. fix: Load summarization model locally with float16 (HF free tier has no outbound DNS)⁠

    d229a9c
  11. fix: DNS endpoint discovery - test multiple HF hostnames to find reachable one

    922cc55
  12. fix: Construct RequestParameters with all 6 required args (url, task, model, json, data, headers)

    625b5d7
  13. fix: Construct RequestParameters properly using inspect.signature

    d3fe06e
  14. fix: Try multiple _inner_post calling conventions + inspect signature

    06ff15b
  15. fix: Use _inner_post transport from InferenceClient v1.19.0

    82a77dd
  16. debug: Deep diagnostics to find InferenceClient internal transport

    dc7f39b
  17. fix: Back to InferenceClient with correct method signatures + diagnostics

    8b888b0
  18. fix: Use requests.post() directly instead of InferenceClient (version compat issue)

    0405b6b
  19. fix: Use client.post() for all models, fix summarization params, better error capture

    e18c6c8
  20. fix: Use huggingface_hub InferenceClient instead of raw urllib (fixes DNS inside HF Spaces)

    7cd5afa
  21. fix: Add wait_for_model, debug endpoint, and better HF API error handling

    2df4f2d
  22. feat: Switch all NLP models to HuggingFace Inference API⁠

    239c0bb

  1. Add FINAL_PRODUCTION_AUDIT.md — 10/10 tests passed

    31023ed
  2. Update DEPLOYMENT_REPORT.md with final status and next steps

    f06d922
  3. Fix: Google link failure now falls back to full sign-in instead of error toast

    d487585
  4. Fix: Guest login navigates to home/landing page instead of editor

    349ddaa
  5. Add HuggingFace Spaces metadata to README.md (fixes CONFIG_ERROR)

    9c52ab4
  6. Use bayan10 HuggingFace org for Spaces deployment

    0959fbc
  7. Switch backend from Render to HuggingFace Spaces (free, no card required)

    510a90e
  8. authDone Ready for deployment

    bb51619
  9. Merge pull request #2 from mohamedatef24/UI/UX_atef⁠

    Mohamed Atef
    d3ef934
  10. added light theme and ISLAMIC RAG and fixed the UI/UX issues and added the auth

    012d0ac

  1. Merge pull request #1 from mohamedatef24/UI_ATEF⁠

    Mohamed Atef
    97fb09f
  2. initial UI

    0c086e7

  1. Initial commit: repository skeleton and folder READMEs

    b366dff