root/Jyotisha
1
0
Fork 0
Code Issues 56 Pull Requests Actions Packages Projects Releases Wiki Activity

fix(report): keep the English edition when one phrase was untranslated

Browse Source 
The English personal report was discarded whenever a Chinese fragment survived the trailing glossary. English sections now render the English fields that were already calculated.

Co-Authored-By: Grok (agent loop)
Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
jesse-ux
2026-10-22 02:56:00 +08:00
co-authored by Grok (agent loop) Cursor
parent 9812f31737
commit ba1ffeb35f
18 changed files with 815 additions and 53 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/report_english_fictional.py
+205
View File
@@ -0,0 +1,205 @@
"""Fixed fictional charts for the English-edition sweep (issue #89).
Not collected by pytest. Birth data stays inside the generator; callers report
only the chart index and leftover fragments.
"""
from __future__ import annotations
import builtins
import importlib
import os
import random
import re
import sys
from pathlib import Path
HAN = re.compile(r"[㐀-鿿豈-﫿]")
GLUE = (
"scoreandtimeline",
"marriagequality",
"inperiod start",
"Dashasignpresent",
"aturninperiod",
"atpresent",
"planetsinbetween",
"whetherinunderway",
"withSaturn",
"careerplanetAmK",
"careerdirection",
"supportingstrongdegree",
"beneficfactor count",
"comparativelystrong",
"occupiesplanetcount",
"lordandstrong",
"lordstrongand",
"weakafflicted",
"partialreducedweak",
"debilitatedsign",
"planetenterssign",
"planetsdistributed",
"beneficlocated",
"house 5 level",
"house seven-round",
"condition planets",
"[source text retained",
)
SEED = 20261021
TODAY = "2026-10-21"
CHART_COUNT = 200
def fictional_case(index: int) -> dict:
"""Chart ``index`` from the fixed sequence. Replaying 0..index keeps the seed stable."""
rng = random.Random(SEED)
chosen = None
for i in range(index + 1):
year = rng.randint(1940, 2015)
month = rng.randint(1, 12)
day = rng.randint(1, 28)
hour = rng.randint(0, 23)
minute = rng.randint(0, 59)
lat = round(rng.uniform(-45, 60), 4)
lon = round(rng.uniform(-120, 150), 4)
tz = max(-12.0, min(14.0, round(lon / 7.5) * 0.5))
chosen = {
"name": f"chart-{i:03d}",
"year": year,
"month": month,
"day": day,
"hour": hour,
"minute": minute,
"second": 0,
"lat": lat,
"lon": lon,
"tz": tz,
"node_mode": rng.choice(("mean", "true")),
"ayanamsa": rng.choice(("lahiri", "raman", "kp")),
"today": TODAY,
"transit_date": TODAY,
"target_year": 2026,
"age": 2026 - year,
"birth_time_accuracy": "confirmed",
}
return chosen
def block_jhora() -> None:
"""Fail closed if this process imports PyJHora. Call once per process."""
real_import = builtins.__import__
def guarded(name, globals=None, locals=None, fromlist=(), level=0):
if str(name).split(".", 1)[0] == "jhora":
raise ImportError("jhora blocked")
return real_import(name, globals, locals, fromlist, level)
builtins.__import__ = guarded
real_import_module = importlib.import_module
def guarded_module(name, package=None):
if str(name).split(".", 1)[0] == "jhora":
raise ImportError("jhora blocked")
return real_import_module(name, package)
importlib.import_module = guarded_module
for key in list(sys.modules):
if key == "jhora" or key.startswith("jhora."):
del sys.modules[key]
def _prepare_process() -> None:
root = Path(__file__).resolve().parents[1]
for entry in (str(root), str(root / "scripts")):
if entry not in sys.path:
sys.path.insert(0, entry)
block_jhora()
def _residue(text: str) -> dict:
found = HAN.findall(text or "")
glue = [word for word in GLUE if word in (text or "")]
return {"han": len(found), "han_chars": "".join(sorted(set(found)))[:40], "glue": glue}
def render_english(index: int) -> dict:
"""Build one fictional packet and run ``_english_edition``. No birth data in the result."""
_prepare_process()
from scripts.jyotish_engine import build_professional_report_reference_packet, cmd_full_reading
from scripts.professional_report_reference import _english_edition, _export_args
birth = fictional_case(index)
reading = cmd_full_reading(type("Args", (), dict(birth))())
packet = dict(build_professional_report_reference_packet(reading, _export_args(dict(birth)), []))
packet["report_version"] = "pl9_personal_long_report.v3"
edition = _english_edition(packet, True)
markdown = edition.get("markdown_en") or ""
flags = _flags_from_packet(packet)
applicability = edition.get("reader_dasha_applicability_en") or ""
body = _residue(markdown)
extra = _residue(applicability)
dump_dir = os.environ.get("REPORT_ENGLISH_SWEEP_DUMP")
if dump_dir and markdown:
path = Path(dump_dir)
path.mkdir(parents=True, exist_ok=True)
(path / f"chart-{index:03d}.md").write_text(markdown, encoding="utf-8")
return {
"i": index,
"status": "ok" if markdown and not edition.get("english_unavailable") else edition.get("english_unavailable") or "empty",
"han": body["han"] + extra["han"],
"han_chars": (body["han_chars"] + extra["han_chars"])[:40],
"glue": sorted(set(body["glue"] + extra["glue"])),
**flags,
}
def _flags_from_packet(packet: dict) -> dict:
rath = (((packet.get("worksheets") or {}).get("timing_and_predictive_systems") or {}).get("narayana_dasha") or {}).get("rath") or {}
rule = rath.get("current_md_antardasha_rule") or {}
direction = rule.get("direction") if isinstance(rule, dict) else {}
advanced = (packet.get("worksheets") or {}).get("advanced_systems") or {}
doshas = advanced.get("yogas_doshas") or {}
kaal = doshas.get("kaal_sarp_dosha") or {}
birth_sheet = (packet.get("worksheets") or {}).get("birth_and_parameters") or {}
pack = birth_sheet.get("panchanga") or {}
panchanga = pack.get("raw") if isinstance(pack, dict) and isinstance(pack.get("raw"), dict) else pack
yoga = advanced.get("yoga") if isinstance(advanced.get("yoga"), dict) else {}
rows = [row for row in (yoga.get("detected_yogas") or []) if isinstance(row, dict)]
return {
"level": rath.get("strength_level_decided"),
"direction": rath.get("direction"),
"ad_rule": direction.get("rule") if isinstance(direction, dict) else None,
"overall": panchanga.get("overall_quality") if isinstance(panchanga, dict) else None,
"kaal": kaal.get("type") if isinstance(kaal, dict) else None,
"punarphoo": any(str(row.get("strength") or "") == "观察" or "Punarphoo" in str(row.get("name") or "") for row in rows),
}
def chart_flags(index: int) -> dict:
"""Class flags used to prove the smoke set covers L1–L6. No birth data."""
_prepare_process()
from scripts.jyotish_engine import build_professional_report_reference_packet, cmd_full_reading
from scripts.professional_report_reference import _export_args
birth = fictional_case(index)
reading = cmd_full_reading(type("Args", (), dict(birth))())
packet = dict(build_professional_report_reference_packet(reading, _export_args(dict(birth)), []))
rath = (((packet.get("worksheets") or {}).get("timing_and_predictive_systems") or {}).get("narayana_dasha") or {}).get("rath") or {}
rule = rath.get("current_md_antardasha_rule") or {}
direction = rule.get("direction") if isinstance(rule, dict) else {}
advanced = (packet.get("worksheets") or {}).get("advanced_systems") or {}
doshas = advanced.get("yogas_doshas") or {}
kaal = doshas.get("kaal_sarp_dosha") or {}
birth_sheet = (packet.get("worksheets") or {}).get("birth_and_parameters") or {}
pack = birth_sheet.get("panchanga") or {}
panchanga = pack.get("raw") if isinstance(pack, dict) and isinstance(pack.get("raw"), dict) else pack
yoga = advanced.get("yoga") if isinstance(advanced.get("yoga"), dict) else {}
names = [str(row.get("name") or "") for row in (yoga.get("detected_yogas") or []) if isinstance(row, dict)]
strengths = [str(row.get("strength") or "") for row in (yoga.get("detected_yogas") or []) if isinstance(row, dict)]
return {
"i": index,
"level": rath.get("strength_level_decided"),
"direction": rath.get("direction"),
"ad_rule": direction.get("rule") if isinstance(direction, dict) else None,
"overall": panchanga.get("overall_quality") if isinstance(panchanga, dict) else None,
"kaal": kaal.get("type") if isinstance(kaal, dict) else None,
"punarphoo": "观察" in strengths or any("Punarphoo" in name for name in names),
}
tests/test_narayana_legacy_label.py
+15 -12
View File
@@ -91,26 +91,29 @@ def test_full_reading_carries_the_rath_view_and_leaves_the_legacy_result_alone()
def test_reference_edition_shows_both_and_english_terms_cover_the_rath_lines() -> None:
from scripts.narayana_legacy_label import rath_report_lines_zh
from scripts.pl9_reader_english_terms import PHRASES_EN, REGEX_EN
from scripts.narayana_legacy_label import rath_report_lines_en
packet = _packet(CASES["day"])
# 原值：reference 版同时有旧算法句和 Rath 表
# 新值：full_data 版同时有这两段
# 原因：用户看到的是完整数据版；Rath 并列表必须跟着走（BUG-1218）
# 原值：Rath 中文行经 PHRASES_EN 可译干净
# 新值：完整数据版英文渲染的 Rath 段无汉字且用英文函数
# 原因：完整数据版不走 PHRASES_EN，旧断言测的不是生产路径
zh = _pl9_export_markdown_for_edition(copy.deepcopy(packet), "full_data")
legacy = zh.index(f"- 算法：{LEGACY_NOTE_ZH}")
rath = zh.index("- 并列：Sanjay Rath 版（并列参考，未进打分与时间轴结论；按原书 p.43 正文（第七轮裁定））：起运宫 ")
assert legacy < rath
assert "| Rath MD | 星座 | 宫主 | 起始年龄 | 结束年龄 | 基本年数 | 修正 | 最终 |" in zh and "| Rath AD | 星座 | 起始年龄 | 结束年龄 |" in zh
view = packet["worksheets"]["timing_and_predictive_systems"]["narayana_dasha"]["rath"]
for line in rath_report_lines_zh(view):
text = line
for pattern, target in REGEX_EN:
text = pattern.sub(target, text)
for source, target in PHRASES_EN:
text = text.replace(source, target)
assert not re.search(r"[一-鿿]", text), text
english_packet = copy.deepcopy(packet)
english_packet["report_language"] = "en"
en = _pl9_export_markdown_for_edition(english_packet, "full_data")
lines = [line for line in rath_report_lines_en(view) if line]
assert lines and lines[0].startswith("- Alongside:")
assert not re.search(r"[一-鿿]", "\n".join(lines))
start = en.index("- Alongside:")
section = en[start:start + 2500]
assert "level " in section and "[source text retained" not in section
assert not re.search(r"[一-鿿]", section)
assert "house " not in section.split("\n", 1)[0]
def test_consultation_modules_carry_the_rath_view() -> None:
tests/test_report_english_han_sources.py
+148
View File
@@ -0,0 +1,148 @@
"""Source English for the leaks named in issue #89. Fictional inputs only."""
from __future__ import annotations
import re
from scripts.muhurta import calc_panchanga
from scripts.narayana_legacy_label import (
DIRECTION_EN,
STRENGTH_LEVEL_EN,
rath_report_lines_en,
rath_report_lines_zh,
rath_summary_line_zh,
)
from scripts.pl9_reader_export import render_pl9_parity_markdown
from scripts.yogas_doshas import calc_kaal_sarp_dosha, calc_sade_sati
from scripts.yoga_engine import STRENGTH_EN
HAN = re.compile(r"[㐀-鿿豈-﫿]")
def _rath_view(*, level: int, direction: str, with_divergence: bool) -> dict:
rule_en = "Ketu in the MD sign: reversed (Table 14)"
direction_block = {
"rule_text": "大运星座有计都：反向（Table 14）",
"rule_text_en": rule_en,
"direction": "reverse",
"book_divergence": [
{
"rule_text": "p.47（J.S. 1.1.34）看大运星座奇偶",
"rule_text_en": "p.47 (J.S. 1.1.34) by the odd/even of the MD sign",
"direction": "zodiacal",
}
] if with_divergence else [],
}
return {
"seed_sign": "Libra",
"strength_level_decided": level,
"strength_level_text": {1: "占据星数", 5: "宫主星座内度数"}.get(level, f"级别{level}"),
"strength_level_text_en": STRENGTH_LEVEL_EN[level],
"direction": direction,
"direction_text": {key: "计都在起运宫：反向（Table 12）" for key in DIRECTION_EN}.get(direction, "中文顺序"),
"direction_text_en": DIRECTION_EN[direction],
"mahadasha": [{
"sign": "Libra", "lord": "Venus", "start_age": 0, "end_age": 7,
"base_years": 7, "dignity_adjustment": 0, "years": 7,
}],
"current_dasha": {"md": {"sign": "Libra", "lord": "Venus"}},
"current_md_antardasha": [{"sign": "Libra", "start_age": 0, "end_age": 0.5}],
"current_md_antardasha_rule": {
"compared": ["Libra", "Aries"],
"stronger_sign": "Libra",
"strength_level_decided": level,
"lord": "Venus",
"start_sign": "Libra",
"direction": direction_block,
},
}
def test_rath_english_lines_cover_levels_tables_and_book_divergence() -> None:
before = rath_summary_line_zh({"seed_sign": "Libra", "strength_level_decided": 5, "strength_level_text": "宫主星座内度数"})
for level in range(1, 8):
for direction in ("table10_general", "table11_saturn_zodiacal", "table12_ketu_reversed"):
view = _rath_view(level=level, direction=direction, with_divergence=True)
lines = "\n".join(rath_report_lines_en(view))
assert not HAN.search(lines)
assert "house " not in lines
assert f"level {level}: {STRENGTH_LEVEL_EN[level]}" in lines
assert DIRECTION_EN[direction] in lines
assert "the book itself disagrees, not applied:" in lines
ketu = "\n".join(rath_report_lines_en(_rath_view(level=5, direction="table12_ketu_reversed", with_divergence=True)))
assert "level 5: lord's degree in sign" in ketu
assert "Ketu in the start sign: reversed (Table 12)" in ketu
assert rath_summary_line_zh({"seed_sign": "Libra", "strength_level_decided": 5, "strength_level_text": "宫主星座内度数"}) == before
zh = "\n".join(rath_report_lines_zh(_rath_view(level=5, direction="table12_ketu_reversed", with_divergence=True)))
assert "- 并列：" in zh and "计都在起运宫：反向（Table 12）" in zh
def test_panchanga_overall_quality_keeps_chinese_and_adds_english() -> None:
seen = {}
for sun in range(0, 360, 12):
for moon in range(0, 360, 15):
row = calc_panchanga(float(sun), float(moon), 3, 6.0)
seen[row["overall_quality"]] = row["overall_quality_en"]
assert row["overall_quality"] in {"吉（Subha）", "中（Mixed）", "凶（Asubha）"}
assert seen == {
"吉（Subha）": "Auspicious (Subha)",
"中（Mixed）": "Mixed",
"凶（Asubha）": "Inauspicious (Asubha)",
}
def _planets(houses: dict[str, int]) -> dict:
return {name: {"house": house, "sign": "Aries"} for name, house in houses.items()}
def test_kaal_sarp_english_summaries_count_planets_in_words() -> None:
names = ("Sun", "Moon", "Mars", "Mercury", "Jupiter", "Venus", "Saturn")
full = _planets({name: 1 + (index % 6) for index, name in enumerate(names)})
full["Rahu"] = {"house": 1, "sign": "Aries"}
full["Ketu"] = {"house": 7, "sign": "Libra"}
full_row = calc_kaal_sarp_dosha(full)
assert full_row["summary"].startswith("完全Kaal Sarp Dosha")
assert full_row["summary_en"] == "Full Kaal Sarp Dosha — all 7 planets lie between the Rahu (1)-Ketu axis"
assert not HAN.search(full_row["summary_en"])
partial = _planets({name: (1 if index < 5 else 8) for index, name in enumerate(names)})
partial["Rahu"] = {"house": 1, "sign": "Aries"}
partial["Ketu"] = {"house": 7, "sign": "Libra"}
partial_row = calc_kaal_sarp_dosha(partial)
assert partial_row["summary"].startswith("部分Kaal Sarp Dosha")
assert partial_row["summary_en"] == "Partial Kaal Sarp Dosha — 5 of 7 planets lie between the axis"
assert "condition planets" not in partial_row["summary_en"]
none = _planets({name: 8 for name in names})
none["Rahu"] = {"house": 1, "sign": "Aries"}
none["Ketu"] = {"house": 7, "sign": "Libra"}
none_row = calc_kaal_sarp_dosha(none)
assert none_row["summary"] == "无Kaal Sarp Dosha"
assert none_row["summary_en"] == "No Kaal Sarp Dosha"
def test_sade_sati_english_summary_has_no_glued_words() -> None:
row = calc_sade_sati("Leo")
assert row["summary"].startswith("Sade Sati阶段框架：")
assert "judge from Saturn's actual position whether it is underway" in row["summary_en"]
assert not HAN.search(row["summary_en"])
for glued in ("atpresent", "whetherinunderway", "withSaturn"):
assert glued not in row["summary_en"]
def test_punarphoo_strength_column_uses_english() -> None:
packet = {
"report_language": "en",
"worksheets": {"advanced_systems": {"yoga": {"detected_yogas": [{
"name": "Punarphoo Yoga",
"category": "relationship_observation",
"strength": "观察",
"strength_en": "observational",
"combination": "月亮与土星合相",
"combination_en": "Moon conjunct Saturn",
"effects_en": ["Watch the relationship"],
}]}}},
}
text = render_pl9_parity_markdown(packet)
assert "| Punarphoo Yoga | relationship_observation | observational | Moon conjunct Saturn |" in text
assert "观察" not in text
assert STRENGTH_EN["观察"] == "Watch"
tests/test_report_english_sweep.py
+27
View File
@@ -0,0 +1,27 @@
"""200 fixed fictional charts: the English edition must come back whole (issue #89).
This file is not on the quick gate. Windows timing is recorded in
docs/tasks/PROGRESS-report-english-han-leak-20261021.md. Set
REPORT_ENGLISH_SWEEP_DUMP to a directory to write the English markdown for
the acceptance pass. Workers stay at 4.
"""
from __future__ import annotations
import os
from concurrent.futures import ProcessPoolExecutor, as_completed
from tests.report_english_fictional import CHART_COUNT, render_english
def test_two_hundred_fictional_charts_have_clean_english() -> None:
workers = min(4, os.cpu_count() or 1, CHART_COUNT)
failures = []
with ProcessPoolExecutor(max_workers=workers) as pool:
futures = [pool.submit(render_english, index) for index in range(CHART_COUNT)]
for future in as_completed(futures):
row = future.result()
if row["status"] != "ok" or row["han"] or row["glue"]:
failures.append(
f"#{row['i']} status={row['status']} han={row['han']} chars={row['han_chars']} glue={row['glue']}"
)
assert failures == []
tests/test_report_english_sweep_smoke.py
+46
View File
@@ -0,0 +1,46 @@
"""Smoke slice of the English sweep, on the quick gate (issue #89 / BUG-1410).
Indices are from tests.report_english_fictional.fictional_case. Together they
hit Rath level 5, an inauspicious Panchanga, Ketu in the start sign, a Table 14
sub-period, Punarphoo, and full Kaal Sarp. Replacing the English Rath lines
with the Chinese ones makes this file fail.
"""
from __future__ import annotations
from concurrent.futures import ProcessPoolExecutor, as_completed
from tests.report_english_fictional import render_english
# Filled once the flag scan names a chart for every class. The test checks the
# flags, so a stale index fails instead of silently dropping a class.
# 1 table12, 4/10/12 level 5, 5 table14 + inauspicious, 6 table12,
# 9/12 partial Kaal Sarp, 11/15 table 11, 14 inauspicious,
# 32 full Kaal Sarp, 34 Punarphoo. Cheap positions matched the
# 0-15 packet Kaal Sarp types before these two were added.
SMOKE_INDEXES = (1, 4, 5, 6, 9, 10, 11, 12, 14, 15, 32, 34)
def test_smoke_charts_cover_every_leak_class_and_render_clean_english() -> None:
rows = []
failures = []
with ProcessPoolExecutor(max_workers=min(4, len(SMOKE_INDEXES))) as pool:
futures = [pool.submit(render_english, index) for index in SMOKE_INDEXES]
for future in as_completed(futures):
row = future.result()
rows.append(row)
if row["status"] != "ok" or row["han"] or row["glue"]:
failures.append(
f"#{row['i']} status={row['status']} han={row['han']} chars={row['han_chars']} glue={row['glue']}"
)
levels = {row["level"] for row in rows}
directions = {row["direction"] for row in rows}
ad_rules = {row["ad_rule"] for row in rows}
overall = {row["overall"] for row in rows}
kaal = {row["kaal"] for row in rows}
assert 5 in levels
assert "table12_ketu_reversed" in directions
assert "rath_ketu_table14_v1" in ad_rules
assert "凶（Asubha）" in overall
assert any(row["punarphoo"] for row in rows)
assert "full" in kaal
assert failures == []