Initial commit

backend/app/api/routes_search.py

@@ -0,0 +1,84 @@
+"""Search endpoints for full-text and metadata document discovery."""
+
+from fastapi import APIRouter, Depends, Query
+from sqlalchemy import Text, cast, func, select
+from sqlalchemy.orm import Session
+
+from app.api.routes_documents import _apply_discovery_filters
+from app.db.base import get_session
+from app.models.document import Document, DocumentStatus
+from app.schemas.documents import DocumentResponse, SearchResponse
+
+
+router = APIRouter()
+
+
+@router.get("", response_model=SearchResponse)
+def search_documents(
+    query: str = Query(min_length=2),
+    offset: int = Query(default=0, ge=0),
+    limit: int = Query(default=50, ge=1, le=200),
+    include_trashed: bool = Query(default=False),
+    only_trashed: bool = Query(default=False),
+    path_filter: str | None = Query(default=None),
+    tag_filter: str | None = Query(default=None),
+    type_filter: str | None = Query(default=None),
+    processed_from: str | None = Query(default=None),
+    processed_to: str | None = Query(default=None),
+    session: Session = Depends(get_session),
+) -> SearchResponse:
+    """Searches documents using PostgreSQL full-text ranking plus metadata matching."""
+
+    vector = func.to_tsvector(
+        "simple",
+        func.coalesce(Document.original_filename, "")
+        + " "
+        + func.coalesce(Document.logical_path, "")
+        + " "
+        + func.coalesce(Document.extracted_text, "")
+        + " "
+        + func.coalesce(cast(Document.tags, Text), ""),
+    )
+    ts_query = func.plainto_tsquery("simple", query)
+    rank = func.ts_rank_cd(vector, ts_query)
+
+    search_filter = (
+        vector.op("@@")(ts_query)
+        | Document.original_filename.ilike(f"%{query}%")
+        | Document.logical_path.ilike(f"%{query}%")
+        | cast(Document.tags, Text).ilike(f"%{query}%")
+    )
+
+    statement = select(Document).where(search_filter)
+    if only_trashed:
+        statement = statement.where(Document.status == DocumentStatus.TRASHED)
+    elif not include_trashed:
+        statement = statement.where(Document.status != DocumentStatus.TRASHED)
+    statement = _apply_discovery_filters(
+        statement,
+        path_filter=path_filter,
+        tag_filter=tag_filter,
+        type_filter=type_filter,
+        processed_from=processed_from,
+        processed_to=processed_to,
+    )
+    statement = statement.order_by(rank.desc(), Document.created_at.desc()).offset(offset).limit(limit)
+
+    items = session.execute(statement).scalars().all()
+
+    count_statement = select(func.count(Document.id)).where(search_filter)
+    if only_trashed:
+        count_statement = count_statement.where(Document.status == DocumentStatus.TRASHED)
+    elif not include_trashed:
+        count_statement = count_statement.where(Document.status != DocumentStatus.TRASHED)
+    count_statement = _apply_discovery_filters(
+        count_statement,
+        path_filter=path_filter,
+        tag_filter=tag_filter,
+        type_filter=type_filter,
+        processed_from=processed_from,
+        processed_to=processed_to,
+    )
+    total = session.execute(count_statement).scalar_one()
+
+    return SearchResponse(total=total, items=[DocumentResponse.model_validate(item) for item in items])