unsloth/studio/backend/core/rag
danielhanchen cbb9d2be7e Merge origin/main into feature/deep-research
Two files conflicted, both resolved by hand.

studio/backend/core/inference/tools.py
- Module constants: kept both sides (_RAG_SEARCH_SLOT from this branch,
  _DISABLE_DNS_PINNING_ENV from main).
- _fetch_url_raw redirect loop: took main's version, which is a superset.
  It keeps this branch's IPv6 bracketing and validated Host header and adds
  the UNSLOTH_STUDIO_DISABLE_DNS_PINNING escape hatch for proxies that need
  the hostname in CONNECT. Default path still pins to the validated IP.

studio/frontend/src/components/assistant-ui/thread.tsx
- Composer: nested this branch's Deep Research controls inside main's
  dictation wrapper so they hide while dictating.
- Kept both this branch's research stop handler and main's useAui hook.

studio/backend/tests/test_web_fetch_extraction.py
- main's new DNS-pinning opt-out test used a URL with embedded credentials.
  check_url_access on this branch rejects those outright rather than
  stripping them, so the fixture now uses a plain URL and a separate test
  asserts the credential rejection.

Verified: test_web_fetch_extraction.py 67 passed, test_sandbox_tools.py 145
passed, tests/studio/test_deep_research_frontend_contract.py 9 passed,
tsc --noEmit clean.
2026-07-26 11:39:53 +00:00
..
__init__.py Studio: clean-room compact RAG (knowledge bases, hybrid search, fast indexing) (#5910) 2026-06-09 21:17:04 -07:00
captioner.py Replace standalone Studio wording with Unsloth (#7221) 2026-07-19 00:47:04 -07:00
chunking.py Studio: clean-room compact RAG (knowledge bases, hybrid search, fast indexing) (#5910) 2026-06-09 21:17:04 -07:00
config.py Studio: hide the RAG embedder and llama.cpp probe from the hub cached inventory (#7018) 2026-07-19 03:20:56 -07:00
embed_llama_server.py Studio: add configurable model download location (#7274) 2026-07-23 01:34:38 -07:00
embeddings.py Studio: add configurable model download location (#7274) 2026-07-23 01:34:38 -07:00
ingestion.py Studio: customizable RAG embedding model with HF search, settings tab reorganization (#6800) 2026-07-02 05:26:33 -07:00
locators.py Whole-document context for RAG chat attachments (#6693) 2026-06-30 15:55:23 +02:00
parsers.py Studio: extract text from PDF web results (#7154) 2026-07-16 19:48:12 -03:00
retrieval.py Studio: customizable RAG embedding model with HF search, settings tab reorganization (#6800) 2026-07-02 05:26:33 -07:00
store.py Studio: Data settings tab, uploaded files manager, quant pinning, and chat image preview fix (#7029) 2026-07-20 04:57:44 -07:00
tool.py Whole-document context for RAG chat attachments (#6693) 2026-06-30 15:55:23 +02:00
web_rank.py Studio: read Deep Research synthesis context from the inference orchestrator 2026-07-21 11:05:19 +00:00