Commit Graph

53 Commits

Author SHA1 Message Date
e431c50347 FEAT-8c: mark conflict animals in animals.json for the API loader
Adds a machine-readable conflict:bool to each merged animal (true when files
disagree on genotype/farbschlag/death) so the import loader can quarantine the
32 conflicts without parsing the German review report.
2026-06-06 07:55:08 +02:00
7ea2f97e72 FEAT-8b: bake Julian's authoritative Wurfchronik semantics into the extractor
- survivedToGoHome: unlabeled Tabelle1 col E detected positionally; value-
  adaptive row parsing recovers it from schema-shifted Tabelle2 rows too
  (118 recovered, 79 confirmed by the E=F-TG-s identity)
- breakdown G -> males/females/stillborn/diedLater ('s' = died before Abgabe)
- validation: E = F - TG - s; 113 mismatches as German review-report warnings
- (name, Zucht) canonicalisation: [brackets] == of/von suffix ([ZdkC] ==
  von den Kleinen Chaoten); Zucht = dedup discriminator (0 splits in data)
- animal->litter matching via DOB+(Vater,Mutter): 95 high-confidence,
  40 date-only, 9 ambiguous; litterRef in animals.json
- regenerated report: 889 raw -> 574 unique (279 dated), 32 conflicts

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-06 00:56:14 +02:00
1b776cd994 FEAT-8b: spreadsheet import tooling (stages 1-2) + review report
tools/import/ (Python, zero-dep migration tooling, not product code):
- xlsx_util.py: dependency-free .xlsx reader (shared strings, cells, drawing anchors)
- genotype.py: notation -> frozen 8-locus mapping + verbatim rawGenotype + unmappedTokens; '-' -> '?'
- extract.py: 10 Stammbaum charts + Wurfchronik -> animals.json/litters.json + anchor-mapped photos;
  dedup on normalise(name)+DOB -> German review-report.md (no DB load)

Run: 889 raw -> 587 unique animals, 24 conflicts, 310 ambiguous, 123 photos, 752 litters.
Output gitignored except review-report.md. Re-runnable per file (Wurfchronik Teil2+).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-06 00:40:54 +02:00