Fix: Update all old-style imports inside packages to new paths

65 files in klausur-service packages + 3 in backend-lehrer packages
had stale imports referencing deleted shim modules.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

klausur-service/backend/ocr/__init__.py

@@ -2,7 +2,7 @@
 OCR package — restructured from cv_* flat modules.
 
 Backward-compatible re-exports: consumers can still use
-``from cv_layout import ...`` etc. via the shim files in backend/.
+``from ocr.layout.layout import ...`` etc. via the shim files in backend/.
 """
 
 from .types import *  # noqa: F401,F403

klausur-service/backend/ocr/detect/box_layout.py

@@ -149,7 +149,7 @@ def build_box_zone_grid(
 
     Returns the same format as _build_zone_grid (columns, rows, cells, header_rows).
     """
-    from grid_editor_helpers import _build_zone_grid
+    from grid.editor.helpers import _build_zone_grid
 
     if not zone_words:
         return {

klausur-service/backend/ocr/labeling/api.py

@@ -48,7 +48,7 @@ except ImportError:
     pass
 
 try:
-    from training_export_service import (  # noqa: F401
+    from training.export_service import (  # noqa: F401
         TrainingExportService,
         TrainingSample,
         get_training_export_service,

klausur-service/backend/ocr/labeling/helpers.py

@@ -57,7 +57,7 @@ except ImportError:
 
 # Try to import Training Export Service
 try:
-    from training_export_service import (
+    from training.export_service import (
         TrainingExportService,
         TrainingSample,
         get_training_export_service,

klausur-service/backend/ocr/labeling/routes.py

@@ -20,7 +20,7 @@ from typing import Optional, List
 from datetime import datetime
 import uuid
 
-from metrics_db import (
+from metrics.db import (
     create_ocr_labeling_session,
     get_ocr_labeling_sessions,
     get_ocr_labeling_session,

klausur-service/backend/ocr/labeling/upload_routes.py

@@ -17,7 +17,7 @@ from typing import Optional, List
 import uuid
 import os
 
-from metrics_db import (
+from metrics.db import (
     get_ocr_labeling_session,
     add_ocr_labeling_item,
     get_ocr_labeling_item,
@@ -42,7 +42,7 @@ except ImportError:
     pass
 
 try:
-    from training_export_service import TrainingSample, get_training_export_service
+    from training.export_service import TrainingSample, get_training_export_service
 except ImportError:
     pass
 
@@ -271,7 +271,7 @@ async def run_ocr_for_item(item_id: str):
     if ocr_text is None:
         raise HTTPException(status_code=500, detail="OCR failed")
 
-    from metrics_db import get_pool
+    from metrics.db import get_pool
     pool = await get_pool()
     if pool:
         async with pool.acquire() as conn:

klausur-service/backend/ocr/pipeline/auto.py

@@ -15,7 +15,7 @@ from .reprocess import router as _reprocess_router
 from .auto_steps import router as _steps_router
 
 # Combine both sub-routers into a single router for backwards compatibility.
-# The consumer imports `from ocr_pipeline_auto import router as _auto_router`.
+# The consumer imports `from ocr.pipeline.auto import router as _auto_router`.
 router = APIRouter(prefix="/api/v1/ocr-pipeline", tags=["ocr-pipeline"])
 router.include_router(_reprocess_router)
 router.include_router(_steps_router)

klausur-service/backend/ocr/pipeline/auto_steps.py

@@ -17,7 +17,7 @@ import numpy as np
 from fastapi import APIRouter, HTTPException, Request
 from fastapi.responses import StreamingResponse
 
-from cv_vocab_pipeline import (
+from ocr.cv_pipeline import (
     OLLAMA_REVIEW_MODEL,
     PageRegion,
     RowGeometry,

klausur-service/backend/ocr/pipeline/columns.py

@@ -14,7 +14,7 @@ from typing import Dict, List
 import cv2
 from fastapi import APIRouter, HTTPException
 
-from cv_vocab_pipeline import (
+from ocr.cv_pipeline import (
     _detect_header_footer_gaps,
     _detect_sub_columns,
     classify_column_types,

klausur-service/backend/ocr/pipeline/deskew.py

@@ -12,7 +12,7 @@ from datetime import datetime
 import cv2
 from fastapi import APIRouter, HTTPException
 
-from cv_vocab_pipeline import (
+from ocr.cv_pipeline import (
     create_ocr_image,
     deskew_image,
     deskew_image_by_word_alignment,

klausur-service/backend/ocr/pipeline/dewarp.py

@@ -17,7 +17,7 @@ from typing import Any, Dict
 import cv2
 from fastapi import APIRouter, HTTPException, Query
 
-from cv_vocab_pipeline import (
+from ocr.cv_pipeline import (
     _apply_shear,
     create_ocr_image,
     dewarp_image,

klausur-service/backend/ocr/pipeline/geometry.py

@@ -8,7 +8,7 @@ This module was split into:
   - ocr_pipeline_columns.py    (Column detection + ground truth)
 
 The `router` object is assembled here by including all sub-routers.
-Importers that did `from ocr_pipeline_geometry import router` continue to work.
+Importers that did `from ocr.pipeline.geometry import router` continue to work.
 """
 
 from fastapi import APIRouter

klausur-service/backend/ocr/pipeline/htr_api.py

@@ -243,7 +243,7 @@ async def recognize_from_session(req: HTRSessionRequest):
     Set use_clean=true to prefer the clean image (after handwriting removal step).
     This is useful when you want to do HTR on isolated handwriting regions.
     """
-    from ocr_pipeline_session_store import get_session_db, get_session_image
+    from ocr.pipeline.session_store import get_session_db, get_session_image
 
     session = await get_session_db(req.session_id)
     if not session:

klausur-service/backend/ocr/pipeline/llm_review.py

@@ -15,7 +15,7 @@ from typing import Dict, List
 from fastapi import APIRouter, HTTPException, Request
 from fastapi.responses import StreamingResponse
 
-from cv_vocab_pipeline import (
+from ocr.cv_pipeline import (
     OLLAMA_REVIEW_MODEL,
     llm_review_entries,
     llm_review_entries_streaming,

klausur-service/backend/ocr/pipeline/ocr_merge.py

@@ -15,7 +15,7 @@ import cv2
 import numpy as np
 from fastapi import APIRouter, HTTPException
 
-from cv_words_first import build_grid_from_words
+from ocr.words_first import build_grid_from_words
 from .common import _cache, _append_pipeline_log
 from .session_store import get_session_image, update_session_db
 
@@ -136,7 +136,7 @@ async def paddle_kombi(session_id: str):
     img_png, img_bgr = await _load_session_image(session_id)
     img_h, img_w = img_bgr.shape[:2]
 
-    from cv_ocr_engines import ocr_region_paddle
+    from ocr.engines.engines import ocr_region_paddle
 
     t0 = time.time()
 
@@ -202,8 +202,8 @@ async def rapid_kombi(session_id: str):
     img_png, img_bgr = await _load_session_image(session_id)
     img_h, img_w = img_bgr.shape[:2]
 
-    from cv_ocr_engines import ocr_region_rapid
-    from cv_vocab_types import PageRegion
+    from ocr.engines.engines import ocr_region_rapid
+    from ocr.types import PageRegion
 
     t0 = time.time()
 

klausur-service/backend/ocr/pipeline/orientation_api.py

@@ -9,7 +9,7 @@ from typing import Any, Dict
 import cv2
 from fastapi import APIRouter, HTTPException
 
-from cv_vocab_pipeline import detect_and_fix_orientation
+from ocr.cv_pipeline import detect_and_fix_orientation
 from .page_crop import detect_page_splits
 from .session_store import update_session_db
 

klausur-service/backend/ocr/pipeline/overlay_structure.py

@@ -17,8 +17,8 @@ from fastapi.responses import Response
 
 from .common import _get_base_image_png
 from .session_store import get_session_db
-from cv_color_detect import _COLOR_HEX, _COLOR_RANGES
-from cv_box_detect import detect_boxes, split_page_into_zones
+from ocr.detect.color_detect import _COLOR_HEX, _COLOR_RANGES
+from ocr.detect.box_detect import detect_boxes, split_page_into_zones
 
 logger = logging.getLogger(__name__)
 

klausur-service/backend/ocr/pipeline/postprocess.py

@@ -19,7 +19,7 @@ from .reconstruction import router as _reconstruction_router
 from .validation import router as _validation_router
 
 # Composite router — drop-in replacement for the old monolithic router.
-# ocr_pipeline_api.py imports ``from ocr_pipeline_postprocess import router``.
+# ocr_pipeline_api.py imports ``from ocr.pipeline.postprocess import router``.
 router = APIRouter()
 router.include_router(_llm_review_router)
 router.include_router(_reconstruction_router)

klausur-service/backend/ocr/pipeline/regression_endpoints.py

@@ -14,7 +14,7 @@ from typing import Any, Dict, Optional
 
 from fastapi import APIRouter, HTTPException, Query
 
-from grid_editor_api import _build_grid_core
+from grid.editor.api import _build_grid_core
 from .session_store import (
     get_session_db,
     list_ground_truth_sessions_db,

klausur-service/backend/ocr/pipeline/rows.py

@@ -17,7 +17,7 @@ import cv2
 import numpy as np
 from fastapi import APIRouter, HTTPException
 
-from cv_vocab_pipeline import (
+from ocr.cv_pipeline import (
     create_ocr_image,
     detect_column_geometry,
     detect_row_geometry,
@@ -120,7 +120,7 @@ async def detect_rows(session_id: str):
     # merges rows.  Word-grouping directly clusters words by Y proximity,
     # which is more robust for these cases.
     if is_sub_session and word_dicts:
-        from cv_layout import _build_rows_from_word_grouping
+        from ocr.layout.layout import _build_rows_from_word_grouping
         rows = _build_rows_from_word_grouping(
             word_dicts, left_x, right_x, top_y, bottom_y,
             right_x - left_x, bottom_y - top_y,

klausur-service/backend/ocr/pipeline/sessions_crud.py

@@ -15,7 +15,7 @@ import cv2
 import numpy as np
 from fastapi import APIRouter, File, Form, HTTPException, Query, UploadFile
 
-from cv_vocab_pipeline import render_image_high_res, render_pdf_high_res
+from ocr.cv_pipeline import render_image_high_res, render_pdf_high_res
 from .common import (
     VALID_DOCUMENT_CATEGORIES,
     UpdateSessionRequest,

klausur-service/backend/ocr/pipeline/sessions_images.py

@@ -17,7 +17,7 @@ import numpy as np
 from fastapi import APIRouter, HTTPException, Query
 from fastapi.responses import Response
 
-from cv_vocab_pipeline import create_ocr_image, detect_document_type
+from ocr.cv_pipeline import create_ocr_image, detect_document_type
 from .common import (
     VALID_DOCUMENT_CATEGORIES,
     _append_pipeline_log,

klausur-service/backend/ocr/pipeline/structure.py

@@ -15,9 +15,9 @@ import numpy as np
 from fastapi import APIRouter, HTTPException
 from pydantic import BaseModel
 
-from cv_box_detect import detect_boxes
-from cv_color_detect import _COLOR_RANGES, _COLOR_HEX
-from cv_graphic_detect import detect_graphic_elements
+from ocr.detect.box_detect import detect_boxes
+from ocr.detect.color_detect import _COLOR_RANGES, _COLOR_HEX
+from ocr.detect.graphic_detect import detect_graphic_elements
 from .session_store import (
     get_session_db,
     update_session_db,
@@ -100,7 +100,7 @@ async def detect_structure(session_id: str):
     )
 
     # --- Zone splitting ---
-    from cv_box_detect import split_page_into_zones as _split_zones
+    from ocr.detect.box_detect import split_page_into_zones as _split_zones
     zones = _split_zones(content_x, content_y, content_w_px, content_h_px, boxes)
 
     # --- Color region sampling ---
@@ -123,7 +123,7 @@ async def detect_structure(session_id: str):
             med_s = float(np.median(roi_hsv[:, :, 1]))
             med_v = float(np.median(roi_hsv[:, :, 2]))
             if med_s > 15:
-                from cv_color_detect import _hue_to_color_name
+                from ocr.detect.color_detect import _hue_to_color_name
                 bg_name = _hue_to_color_name(med_h)
                 bg_hex = _COLOR_HEX.get(bg_name, "#6b7280")
             else:

klausur-service/backend/ocr/pipeline/words.py

@@ -23,7 +23,7 @@ import numpy as np
 from fastapi import APIRouter, HTTPException
 from pydantic import BaseModel
 
-from cv_words_first import build_grid_from_words
+from ocr.words_first import build_grid_from_words
 from .session_store import (
     get_session_db,
     get_session_image,
@@ -72,7 +72,7 @@ async def paddle_direct(session_id: str):
 
     img_h, img_w = img_bgr.shape[:2]
 
-    from cv_ocr_engines import ocr_region_paddle
+    from ocr.engines.engines import ocr_region_paddle
 
     t0 = time.time()
     word_dicts = await ocr_region_paddle(img_bgr, region=None)

klausur-service/backend/ocr/pipeline/words_detect.py

@@ -17,7 +17,7 @@ import numpy as np
 from fastapi import APIRouter, HTTPException, Request
 from fastapi.responses import StreamingResponse
 
-from cv_vocab_pipeline import (
+from ocr.cv_pipeline import (
     PageRegion,
     RowGeometry,
     _cells_to_vocab_entries,
@@ -27,7 +27,7 @@ from cv_vocab_pipeline import (
     create_ocr_image,
     detect_column_geometry,
 )
-from cv_words_first import build_grid_from_words
+from ocr.words_first import build_grid_from_words
 from .session_store import (
     get_session_db,
     update_session_db,
@@ -223,7 +223,7 @@ async def _words_first_path(
     img_h, img_w = dewarped_bgr.shape[:2]
 
     if engine == "paddle":
-        from cv_ocr_engines import ocr_region_paddle
+        from ocr.engines.engines import ocr_region_paddle
         wf_word_dicts = await ocr_region_paddle(dewarped_bgr, region=None)
         cached["_paddle_word_dicts"] = wf_word_dicts
     else:

klausur-service/backend/ocr/pipeline/words_stream.py

@@ -15,7 +15,7 @@ from typing import Any, Dict, List
 import numpy as np
 from fastapi import Request
 
-from cv_vocab_pipeline import (
+from ocr.cv_pipeline import (
     PageRegion,
     RowGeometry,
     _cells_to_vocab_entries,

klausur-service/backend/ocr/review/spell.py

@@ -219,7 +219,7 @@ def spell_review_entries_sync(entries: List[Dict]) -> Dict:
     # Use SmartSpellChecker if available
     _smart = None
     try:
-        from smart_spell import SmartSpellChecker
+        from ocr.spell.smart_spell import SmartSpellChecker
         _smart = SmartSpellChecker()
         logger.debug("spell_review: using SmartSpellChecker")
     except Exception:

klausur-service/backend/ocr/spell/core.py

@@ -159,7 +159,7 @@ class _SmartSpellCoreBase:
             return True
         # Also accept known abbreviations (sth, sb, adj, etc.)
         try:
-            from cv_ocr_engines import _KNOWN_ABBREVIATIONS
+            from ocr.engines.engines import _KNOWN_ABBREVIATIONS
             if w in _KNOWN_ABBREVIATIONS:
                 return True
         except ImportError:

klausur-service/backend/ocr/spell/text.py

@@ -35,7 +35,7 @@ class SmartSpellChecker(_SmartSpellCoreBase):
         """
         # Import known abbreviations for vocabulary context
         try:
-            from cv_ocr_engines import _KNOWN_ABBREVIATIONS
+            from ocr.engines.engines import _KNOWN_ABBREVIATIONS
         except ImportError:
             _KNOWN_ABBREVIATIONS = set()
 
@@ -149,7 +149,7 @@ class SmartSpellChecker(_SmartSpellCoreBase):
         # --- Pass 1: Boundary repair between adjacent unknown words ---
         # Import abbreviations for the heuristic below
         try:
-            from cv_ocr_engines import _KNOWN_ABBREVIATIONS as _ABBREVS
+            from ocr.engines.engines import _KNOWN_ABBREVIATIONS as _ABBREVS
         except ImportError:
             _ABBREVS = set()