2 Commits

Author	SHA1	Message	Date
Benjamin Admin	fac9280716	feat(pipeline): Block D5+-E complete session — 20k+ new chunks ... Session 02-03.05.2026 accomplishments: - D5+: NIST/ENISA PDF quality fix (0%→45% section rate) - D5+: 4 lost NIST PDFs restored (11k chunks) - D5+: Text normalization + section detection for NIST/BSI - D6: Citation backfill (3,651 controls updated, old archived) - E2: 8 DE laws ingested (ArbZG, MuSchG, GmbHG, AktG, InsO...) - E3: 5 EU regulations (CSRD, CSDDD, Taxonomy, eIDAS, Pay Trans.) - E4: Standards (GoBD, BAIT, VAIT) - E6: 3 CH + 4 AT laws (OR, DSV, ArG, ArbVG, AngG, AZG, NISG) - E7: 9 court judgments as full text (Schrems II 154 chunks, Meta 101, BVerfG 161, DSK OH 119, Planet49 42, SCHUFA 41, Schadenersatz 29, BAG 48, Google Fonts 14) - Infra: Qdrant snapshot mechanism, upload-before-delete safety Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-05-03 22:31:57 +02:00
Benjamin Admin	2f4a3f2ea2	fix(embedding): add NIST control IDs to _SECTION_NUMBER_RE ... _SECTION_NUMBER_RE only had patterns for §/Art/Section/Kapitel/Annex but missed NIST-style identifiers (AC-1, GV.OC-01, 3.1, A01:2021). This caused 0% section rate for all NIST/BSI/ENISA documents even though sections were correctly detected — the section NUMBER wasn't extracted from the header. Also adds: - reupload_legal_strategy.py: re-upload with legal chunking - extract_and_upload_nist.py: local PDF extraction workaround - qdrant-snapshot.sh: backup mechanism for Qdrant collections Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-05-03 07:42:06 +02:00