giers10/Heimgeist
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Update library embedding and context handling to use pipeline metadata

Browse Source
This commit is contained in:
Victor Giers
2026-03-19 22:34:47 +01:00
parent b2e487262c
commit f86989e722
1 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
backend/local_rag.py
View File

@@ -1138,9 +1138,11 @@ async def enrich_library(slug: str):
async def embed_library(slug: str, req: EmbedLibraryRequest): async def embed_library(slug: str, req: EmbedLibraryRequest):
data = read_library(slug) data = read_library(slug)
payload = library_payload(data) payload = library_payload(data)
pipeline = _pipeline_meta(data)
paths = _collect_library_paths(slug) paths = _collect_library_paths(slug)
if not payload["states"].get("has_corpus"): if not payload["states"].get("has_corpus"):
raise HTTPException(status_code=400, detail="Build the corpus before indexing.") raise HTTPException(status_code=400, detail="Build the corpus before indexing.")
embed_model = req.embed_model or pipeline.get("embed_model") or DEFAULT_EMBED_MODEL
lock = LIB_LOCKS.setdefault(slug, asyncio.Lock()) lock = LIB_LOCKS.setdefault(slug, asyncio.Lock())
async with lock: async with lock:
if _has_active_job(slug): if _has_active_job(slug):
@@ -1152,7 +1154,7 @@ async def embed_library(slug: str, req: EmbedLibraryRequest):
enhanced=None, enhanced=None,
shadow=paths["shadow"] if paths["shadow"].exists() else None, shadow=paths["shadow"] if paths["shadow"].exists() else None,
out_dir=paths["indexes"], out_dir=paths["indexes"],
embed_model=req.embed_model, embed_model=embed_model,
ollama=req.ollama, ollama=req.ollama,
target_chars=req.target_chars, target_chars=req.target_chars,
overlap_chars=req.overlap_chars, overlap_chars=req.overlap_chars,
@@ -1194,9 +1196,11 @@ def get_job(job_id: str):
@router.post("/libraries/{slug}/context") @router.post("/libraries/{slug}/context")
def library_context(slug: str, req: LibraryContextRequest): def library_context(slug: str, req: LibraryContextRequest):
payload = library_payload(read_library(slug)) payload = library_payload(read_library(slug))
pipeline = payload.get("pipeline") or {}
paths = _collect_library_paths(slug) paths = _collect_library_paths(slug)
if not payload["states"].get("is_indexed"): if not payload["states"].get("is_indexed"):
raise HTTPException(status_code=400, detail="Prepare the library before using it in chat.") raise HTTPException(status_code=400, detail="Prepare the library before using it in chat.")
embed_model = req.embed_model or pipeline.get("embed_model") or DEFAULT_EMBED_MODEL
try: try:
run_query = _load_pipeline_fn("unified_rag", "run_query") run_query = _load_pipeline_fn("unified_rag", "run_query")
result = run_query( result = run_query(
@@ -1207,7 +1211,7 @@ def library_context(slug: str, req: LibraryContextRequest):
query=req.prompt, query=req.prompt,
answer=False, answer=False,
ollama=req.ollama, ollama=req.ollama,
embed_model=req.embed_model, embed_model=embed_model,
gen_model=req.gen_model, gen_model=req.gen_model,
no_rerank=True, no_rerank=True,
k=max(1, req.top_k), k=max(1, req.top_k),