Commit Graph
8 Commits
Author SHA1 Message Date
Jesse_ChenandClaude Opus 5.5 1dd6203883 fix(report): parity yoga category for ayur reads Constitution; only the boundary scan may be prose (#91 fix2)
"Longevity" in the canonical yoga table broke the BUG-1210 lifespan wording
test. Sections other than the dasha boundary scan must carry a table, so a
fixed sentence cannot pass as packet data.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HwMNZEzdzmrbPhuRNmZe3C
2026-10-22 22:01:33 +08:00
Jesse_ChenandClaude Opus 5.5 ac3dccbac8 fix(yoga): Chinese combination text from a Chinese source, not 未判定 (#91 fix2 F3, BUG-1418)
Removing the English-to-Chinese back-translation left every BVR semantic
note and every yogas_doshas raj/dhana "H*_Lord(...) in H*" sentence as
未判定 in the Chinese yoga table and pages.

- yoga_engine: semantic conditions return BiText(Chinese, old note). The
  Chinese comes from references/yoga_semantic_notes_zh.json, one sentence per
  note, so dedup keeps the same equality classes. English-only custom
  templates get the same lookup. combination_en is unchanged.
- yogas_doshas: raj/dhana rows add combination_zh; combination is unchanged.
- chinese_combination prefers a clean combination_zh.

Which yogas fire, their strength and category are unchanged (fictional
charts 0-19, same rows as staging; only set-iteration order differs).

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HwMNZEzdzmrbPhuRNmZe3C
2026-10-22 21:55:02 +08:00
Jesse_ChenandClaude Opus 5.5 00c2b4d498 fix(report): six parity chapters read real packet records and sit at ## (#91 fix2 F1/F2/F4)
Eleven sections printed one-column tables of 未判定 and packet key names
(summary, p5, stats ...). They now read the nested records: KP profile and
ruling planets, Tara/Chandra combined rows, lagna vargottamamsa flags, LL/7L
P5/P8/exchange checks, Yoga Pinda planets and summary, Sodhita BAV totals and
Ekadhipatya audit, friendship method and bodies, multi-reference houses,
double-transit PAC targets, congregation houses, seven/eight karaka tables.
Argala, Birth Calendar, Tithi, Tajika candidates, Goel dignity and yoga
categories no longer print bare lower-case keys or enums either.

The six chapters are now ## and their sections ###, so the reader counts
six more chapters instead of seventy.

Tests: a section passes only with a real value outside the label column; no
table cell may be a raw key; heading counts equal the base report plus six.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HwMNZEzdzmrbPhuRNmZe3C
2026-10-22 21:44:51 +08:00
jesse-uxGrok (agent loop)Cursor
92b4e220f2 新六章有 12 节读错公开包字段,整节只剩未判定,而且六章绕过了口径替换。按资料包改读路径,先拼六章再统一替换措辞,并把中文附录实际丢掉的组合句数出来。
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-10-22 18:51:54 +08:00
jesse-uxGrok (agent loop)Cursor
4020c3958a 按六章补上已有数据的小附录,并把 Goel 强度与两种 Narayana 口径分开印出。
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-10-22 16:22:45 +08:00
jesse-uxGrok (agent loop)Cursor
8444660f36 中文完整数据附录改为按中文源和结构化字段渲染,不再把英文附录回译成中文。
Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-10-22 16:22:25 +08:00
jesse-uxGrok (agent loop)Cursor
adfa98bf97 英文漏译改在渲染侧查表,测试不再放行 PyJHora 子进程。
冻结的 jaimini.py 恢复成 staging 原样。缺译单元格写成固定标记,不再留空。

Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-10-22 09:58:09 +08:00
jesse-uxandGrok (agent loop) dbba127e71 拆分报告渲染函数的第一阶段,把 34 个无外层依赖的辅助函数搬到 report_render。
同一份真实盘包在屏蔽与不屏蔽 PyJHora 下新旧渲染 sha 都一致,各 24 项且退出码 0。指定 pytest 233 通过、3 跳过。

Co-Authored-By: Grok (agent loop)
2026-10-15 20:59:11 +08:00