Benjamin_Boenisch/breakpilot-lehrer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(ocr-pipeline): invalidate downstream results when steps are re-run

Browse Source 
When columns change (Step 3), invalidate row_result and word_result.
When rows change (Step 4), invalidate word_result.
This ensures Step 5 always uses the latest row boundaries instead of
showing stale cached word_result from a previous run.

Applies to both auto-detection and manual override endpoints.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Benjamin Admin
2026-03-01 12:24:44 +01:00
parent 97d4355aa9
commit 4970ca903e
1 changed files with 14 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
klausur-service/backend/ocr_pipeline_api.py
View File

@@ -708,15 +708,19 @@ async def detect_columns(session_id: str):
"duration_seconds": round(duration, 2), "duration_seconds": round(duration, 2),
} }
# Persist to DB # Persist to DB — also invalidate downstream results (rows, words)
await update_session_db( await update_session_db(
session_id, session_id,
column_result=column_result, column_result=column_result,
row_result=None,
word_result=None,
current_step=3, current_step=3,
) )
# Update cache # Update cache
cached["column_result"] = column_result cached["column_result"] = column_result
cached.pop("row_result", None)
cached.pop("word_result", None)
col_count = len([c for c in columns if c["type"].startswith("column")]) col_count = len([c for c in columns if c["type"].startswith("column")])
logger.info(f"OCR Pipeline: columns session {session_id}: " logger.info(f"OCR Pipeline: columns session {session_id}: "
@@ -737,10 +741,13 @@ async def set_manual_columns(session_id: str, req: ManualColumnsRequest):
"method": "manual", "method": "manual",
} }
await update_session_db(session_id, column_result=column_result) await update_session_db(session_id, column_result=column_result,
row_result=None, word_result=None)
if session_id in _cache: if session_id in _cache:
_cache[session_id]["column_result"] = column_result _cache[session_id]["column_result"] = column_result
_cache[session_id].pop("row_result", None)
_cache[session_id].pop("word_result", None)
logger.info(f"OCR Pipeline: manual columns session {session_id}: " logger.info(f"OCR Pipeline: manual columns session {session_id}: "
f"{len(req.columns)} columns set") f"{len(req.columns)} columns set")
@@ -919,14 +926,16 @@ async def detect_rows(session_id: str):
"duration_seconds": round(duration, 2), "duration_seconds": round(duration, 2),
} }
# Persist to DB # Persist to DB — also invalidate word_result since rows changed
await update_session_db( await update_session_db(
session_id, session_id,
row_result=row_result, row_result=row_result,
word_result=None,
current_step=4, current_step=4,
) )
cached["row_result"] = row_result cached["row_result"] = row_result
cached.pop("word_result", None)
logger.info(f"OCR Pipeline: rows session {session_id}: " logger.info(f"OCR Pipeline: rows session {session_id}: "
f"{len(rows)} rows detected ({duration:.2f}s): {type_counts}") f"{len(rows)} rows detected ({duration:.2f}s): {type_counts}")
@@ -947,10 +956,11 @@ async def set_manual_rows(session_id: str, req: ManualRowsRequest):
"method": "manual", "method": "manual",
} }
await update_session_db(session_id, row_result=row_result) await update_session_db(session_id, row_result=row_result, word_result=None)
if session_id in _cache: if session_id in _cache:
_cache[session_id]["row_result"] = row_result _cache[session_id]["row_result"] = row_result
_cache[session_id].pop("word_result", None)
logger.info(f"OCR Pipeline: manual rows session {session_id}: " logger.info(f"OCR Pipeline: manual rows session {session_id}: "
f"{len(req.rows)} rows set") f"{len(req.rows)} rows set")