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/vocab/__init__.py

@@ -2,5 +2,5 @@
 Vocab package — restructured from vocab_* flat modules.
 
 Backward-compatible re-exports: consumers can still use
-``from vocab_worksheet_api import ...`` etc. via the shim files in backend/.
+``from vocab.worksheet.api import ...`` etc. via the shim files in backend/.
 """

klausur-service/backend/vocab/worksheet/analysis_api.py

@@ -27,7 +27,7 @@ from .generation import convert_pdf_page_to_image
 
 # Try to import Tesseract extractor
 try:
-    from tesseract_vocab_extractor import (
+    from ocr.engines.tesseract_extractor import (
         extract_bounding_boxes, TESSERACT_AVAILABLE,
     )
 except ImportError:
@@ -264,7 +264,7 @@ async def extract_with_boxes(session_id: str, page_number: int):
     # Deskew image before OCR
     deskew_angle = 0.0
     try:
-        from cv_vocab_pipeline import deskew_image_by_word_alignment, CV2_AVAILABLE
+        from ocr.cv_pipeline import deskew_image_by_word_alignment, CV2_AVAILABLE
         if CV2_AVAILABLE:
             image_data, deskew_angle = deskew_image_by_word_alignment(image_data)
             logger.info(f"Deskew: {deskew_angle:.2f}° for page {page_number}")

klausur-service/backend/vocab/worksheet/compare_api.py

@@ -23,7 +23,7 @@ from .generation import convert_pdf_page_to_image
 
 # Try to import Tesseract extractor
 try:
-    from tesseract_vocab_extractor import (
+    from ocr.engines.tesseract_extractor import (
         run_tesseract_pipeline,
         match_positions_to_vocab, TESSERACT_AVAILABLE,
     )
@@ -32,7 +32,7 @@ except ImportError:
 
 # Try to import CV Pipeline
 try:
-    from cv_vocab_pipeline import run_cv_pipeline, CV_PIPELINE_AVAILABLE
+    from ocr.cv_pipeline import run_cv_pipeline, CV_PIPELINE_AVAILABLE
 except ImportError:
     CV_PIPELINE_AVAILABLE = False
 
@@ -328,7 +328,7 @@ async def analyze_grid(session_id: str, page_number: int, use_tesseract: bool =
             # Run Tesseract if not already cached
             if not tess_page_data:
                 logger.info("Running Tesseract for grid analysis (not cached)")
-                from tesseract_vocab_extractor import run_tesseract_pipeline as _run_tess
+                from ocr.engines.tesseract_extractor import run_tesseract_pipeline as _run_tess
                 tess_page_data = await _run_tess(image_data, lang="eng+deu")
                 session[f"tesseract_page_{page_number}"] = tess_page_data
                 session["tesseract_words"] = tess_page_data.get("words", [])

klausur-service/backend/vocab/worksheet/ocr.py

@@ -39,7 +39,7 @@ except ImportError:
 
 # CV pipeline helpers
 try:
-    from cv_vocab_pipeline import (
+    from ocr.cv_pipeline import (
         deskew_two_pass,
         dewarp_image,
         detect_and_fix_orientation,
@@ -54,7 +54,7 @@ except ImportError:
     _fix_phonetic_brackets = None  # type: ignore[assignment]
 
 try:
-    from cv_cell_grid import (
+    from ocr.cell_grid.cell_grid import (
         _merge_wrapped_rows,
         _merge_phonetic_continuation_rows,
         _merge_continuation_rows,
@@ -65,17 +65,17 @@ except ImportError:
     _merge_continuation_rows = None  # type: ignore[assignment]
 
 try:
-    from cv_ocr_engines import ocr_region_rapid
+    from ocr.engines.engines import ocr_region_rapid
 except ImportError:
     ocr_region_rapid = None  # type: ignore[assignment]
 
 try:
-    from cv_vocab_types import PageRegion
+    from ocr.types import PageRegion
 except ImportError:
     PageRegion = None  # type: ignore[assignment]
 
 try:
-    from ocr_pipeline_ocr_merge import (
+    from ocr.pipeline.ocr_merge import (
         _split_paddle_multi_words,
         _merge_paddle_tesseract,
         _deduplicate_words,
@@ -86,12 +86,12 @@ except ImportError:
     _deduplicate_words = None  # type: ignore[assignment]
 
 try:
-    from cv_words_first import build_grid_from_words
+    from ocr.words_first import build_grid_from_words
 except ImportError:
     build_grid_from_words = None  # type: ignore[assignment]
 
 try:
-    from ocr_pipeline_session_store import (
+    from ocr.pipeline.session_store import (
         create_session_db as create_pipeline_session_db,
         update_session_db as update_pipeline_session_db,
     )
@@ -173,7 +173,7 @@ async def _run_ocr_pipeline_for_page(
     # 5. Content crop (removes scanner borders, gutter shadows)
     t0 = _time.time()
     try:
-        from page_crop import detect_and_crop_page
+        from ocr.pipeline.page_crop import detect_and_crop_page
         cropped_bgr, crop_result = detect_and_crop_page(dewarped_bgr)
         if crop_result.get("crop_applied"):
             dewarped_bgr = cropped_bgr
@@ -186,7 +186,7 @@ async def _run_ocr_pipeline_for_page(
     # 5b. Scan quality assessment
     scan_quality_report = None
     try:
-        from scan_quality import score_scan_quality
+        from ocr.pipeline.scan_quality import score_scan_quality
         scan_quality_report = score_scan_quality(dewarped_bgr)
     except Exception as e:
         logger.warning(f"  scan quality: failed ({e})")
@@ -200,7 +200,7 @@ async def _run_ocr_pipeline_for_page(
     is_degraded = scan_quality_report.is_degraded if scan_quality_report else False
     if is_degraded and enable_enhance:
         try:
-            from ocr_image_enhance import enhance_for_ocr
+            from ocr.image_enhance import enhance_for_ocr
             dewarped_bgr = enhance_for_ocr(dewarped_bgr, is_degraded=True)
             logger.info("  enhancement: applied (degraded scan)")
         except Exception as e:
@@ -212,8 +212,8 @@ async def _run_ocr_pipeline_for_page(
 
     # RapidOCR (local ONNX)
     try:
-        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
         full_region = PageRegion(type="full_page", x=0, y=0, width=img_w, height=img_h)
         rapid_words = ocr_region_rapid(dewarped_bgr, full_region) or []
     except Exception as e:
@@ -243,8 +243,8 @@ async def _run_ocr_pipeline_for_page(
         })
 
     # Merge dual-engine results
-    from ocr_pipeline_ocr_merge import _split_paddle_multi_words, _merge_paddle_tesseract, _deduplicate_words
-    from cv_words_first import build_grid_from_words
+    from ocr.pipeline.ocr_merge import _split_paddle_multi_words, _merge_paddle_tesseract, _deduplicate_words
+    from ocr.words_first import build_grid_from_words
 
     rapid_split = _split_paddle_multi_words(rapid_words) if rapid_words else []
     if rapid_split or tess_words:
@@ -300,7 +300,7 @@ async def _run_ocr_pipeline_for_page(
     # 8. Run full grid-build (with pipe-autocorrect, word-gap merge, etc.)
     t0 = _time.time()
     try:
-        from grid_editor_api import _build_grid_core
+        from grid.editor.api import _build_grid_core
         session_data = {
             "word_result": word_result,
         }

klausur-service/backend/vocab/worksheet/upload_api.py

@@ -36,7 +36,7 @@ LOCAL_STORAGE_PATH = os.getenv("VOCAB_STORAGE_PATH", "/app/vocab-worksheets")
 
 try:
     import numpy as np
-    from cv_preprocessing import render_pdf_high_res, detect_and_fix_orientation
+    from ocr.preprocessing.preprocessing import render_pdf_high_res, detect_and_fix_orientation
     OCR_PIPELINE_AVAILABLE = True
 except ImportError:
     np = None  # type: ignore[assignment]