- Reduce left-side threshold from 35% to 20% of content width
- Strong language signal (eng/deu > 0.3) now prevents page_ref assignment
- Increase column_ignore word threshold from 3 to 8 for edge columns
- Apply language guard to Level 1 and Level 2 classification
Fixes: column with deu=0.921 was misclassified as page_ref because
reference score check ran before language analysis.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add REGULATIONS_IN_RAG Set tracking all 42 regulations currently in Qdrant
- Add 4 new regulation entries: E-Commerce-RL, Verbraucherrechte-RL,
Digitale-Inhalte-RL, DMA (all ingested Feb 2026)
- Add RAG column to regulations table with green check/red x indicators
- Update Landkarte tab: green/x on industry cards, thematic clusters,
and regulation matrix
- Replace old "Integrated Regulations" section with full RAG coverage overview
- Update hardcoded chunk counts (Templates: 7689, NiBiS: 7996)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Address 5 weaknesses found via ground-truth comparison on session df3548d1:
- Add column_ignore for edge columns with < 3 words (margin detection)
- Absorb tiny clusters (< 5% width) into neighbors post-merge
- Restrict page_ref to left 35% of content area across all 3 levels
- Loosen marker thresholds (width < 6%, words <= 15) and add strong
marker score for very narrow non-edge columns (< 4%)
- Add EN/DE position tiebreaker when language signals are both weak
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Side-by-side view: auto result (readonly) vs GT editor where teacher
draws correct columns. Diff table shows Auto vs GT with IoU matching.
GT data persisted per session for algorithm tuning.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Sub-alignments within a column (indented words, etc.) were 60-90px apart
and not getting merged at 3%. On a typical 5-col page (~1500px), 6% = ~90px
merges sub-alignments while keeping real column boundaries (~300px) separate.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Clusters now track Y-positions of their words and filter by vertical
coverage (>=30% primary, >=15%+5words secondary) to reject noise from
indentations or page numbers. Merge distance widened to 3% content width.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- ManualColumnEditor now uses grid-cols-2 layout (image left, controls right)
matching the normal view size so the image doesn't zoom in
- StepColumnDetection only runs auto-detection when no cached result exists;
revisiting step 3 loads cached columns without re-running detection
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Ersetzt hardcodierte Positionsregeln durch ein zweistufiges System:
Phase A erkennt Spaltengeometrie (Clustering), Phase B klassifiziert
Typen per Inhalt (Sprache/Rolle) mit 3-stufiger Fallback-Kette.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace projection-profile layout analysis with Tesseract word bounding
box clustering to detect 5-column vocabulary layouts (page_ref, EN, DE,
markers, examples). Falls back to projection profiles when < 3 clusters.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Sessions werden jetzt in PostgreSQL gespeichert statt in-memory.
Neue Session-Liste mit Name, Datum, Schritt. Sessions ueberleben
Browser-Refresh und Container-Neustart. Step 3 nutzt analyze_layout()
fuer automatische Spaltenerkennung mit farbigem Overlay.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The old displacement-map approach shifted entire rows by a parabolic
profile, creating a circle/barrel distortion. The actual problem is
a linear vertical shear: after deskew aligns horizontal lines, the
vertical column edges are still tilted by ~0.5°.
New approach:
- Detect shear angle from strongest vertical edge slope (not curvature)
- Apply cv2.warpAffine shear to straighten vertical features
- Manual slider: -2.0° to +2.0° in 0.05° steps
- Slider initializes to auto-detected shear angle
- Ground truth question: "Spalten vertikal ausgerichtet?"
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The old -3.0 to +3.0 scale multiplied the full displacement map (up to ~79px)
directly, causing extreme distortion at values >1. New slider:
- 0% = no correction
- 100% = auto-detected correction (default)
- 200% = double correction
- Step size: 5%
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Tesseract OCR + 70 Debian packages + pip dependencies are now in a
separate base image (klausur-base:latest) that is built once and reused.
A --no-cache build now only rebuilds the code layer (~seconds) instead
of re-downloading 33 MB of system packages (~9 minutes).
Rebuild base when requirements.txt or system deps change:
docker build -f klausur-service/Dockerfile.base -t klausur-base:latest klausur-service/
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix dewarp method selection: prefer methods with >5px curvature over
higher confidence (vertical_edge 79px was being ignored for text_baseline 2px)
- Add grid overlay on left image in Dewarp step for side-by-side comparison
- Add GET /sessions/{id} endpoint to reload session data
- StepDeskew accepts sessionId prop to restore state when navigating back
- SessionInfo type extended with optional deskew_result and dewarp_result
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
"Korrekt ausgerichtet?" → "Rotation korrekt?" mit Hinweis,
dass Woelbung/Verzerrung im naechsten Schritt korrigiert wird.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Neue Route /ai/ocr-pipeline mit schrittweiser Begradigung (Deskew),
Raster-Overlay und Ground Truth. Schritte 2-6 als Platzhalter.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- edu-search-service von breakpilot-pwa nach breakpilot-lehrer kopiert (ohne vendor)
- opensearch + edu-search-service in docker-compose.yml hinzugefuegt
- voice-service aus docker-compose.yml entfernt (jetzt in breakpilot-core)
- geo-service aus docker-compose.yml entfernt (nicht mehr benoetigt)
- CI/CD: edu-search-service zu Gitea Actions und Woodpecker hinzugefuegt
(Go lint, test mit go mod download, build, SBOM)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The formula int(17 - grade*3) was off by one for grades 1.0 and 1.3
due to integer truncation. Added +1 adjustment for grades < 2.0 to
match the intended mapping (1.0=15, 1.3=14).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The act_runner cannot create /home/act_runner cache dir inside
container images. Replace actions/checkout@v4 with manual
git clone using GITHUB_SERVER_URL and GITHUB_REPOSITORY env vars.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Adds .gitea/workflows/ci.yaml with lint and test jobs.
Runs on gitea.meghsakha.com with Gitea Actions runner.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Der Geo-Service (DSGVO-konforme Geografie-Lernplattform) war ein
nie genutzter Prototyp und wird nicht mehr benoetigt. Entfernt aus:
- Quellcode (geo-service/)
- CI/CD Pipeline (test, build, sbom, deploy steps)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- session_manager: add session to _local_cache in create_session()
so get_session() and get_active_sessions() work without Redis/Postgres
- message_bus: use asyncio.create_task() in request() for publish
so handler runs concurrently and timeout actually fires
All 76 tests pass now.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Add complete service table with containers, ports, and tech stack
- Add Core dependency table
- Add URLs section with Lehrer-Tools
- Add deployment and git instructions
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- DSB Portal, Industry Templates, Multi-Tenant, SSO frontend pages
- All SDK API proxy routes (academy, crawler, incidents, vendors, whistleblower, etc.)
- Blog section with compliance articles
- BYOEH system documentation
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Remove communication, infrastructure, and development categories from
navigation (now in Admin Core on port 3008). Rename Admin v2 to
Admin Lehrer KI in sidebar, header, and browser title.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
