PR Queue — Priority Review Guide
13 PRs are open, all CI-green and merge-clean. Here's a suggested review order based on severity and impact.
Tier 1 — Correctness bugs (wrong output)
These fix silent data corruption or wrong results. Users currently get incorrect values.
| PR |
What's broken |
Impact |
| #141 |
time2str stores TZ-adjusted epoch instead of real epoch → wrong %G and %z when timezone specified |
Any user formatting GPS weeks or numeric TZ offsets with explicit timezone |
| #140 |
format_z loses negative sign for sub-hour offsets → -0030 rendered as +0030 |
Chatham Islands, India half-hour offsets |
| #134 |
strptime fails to parse day 13 in DD-MM-YYYY format → returns undef |
Any European-style date on the 13th |
| #139 |
format_o returns bare "31" instead of "31st" on day 31 (off-by-one in 20 modules) |
Ordinal formatting on last day of month |
| #133 |
%V ISO week wrong at leap year boundaries → week 53 shown as week 01 |
Jan 1 2021, similar year-boundary dates |
Tier 2 — API consistency bugs
These fix behavioral divergence between Date::Parse and Date::Language.
| PR |
What's broken |
Impact |
| #138 |
Date::Language::str2time handles 2-digit years differently than Date::Parse::str2time |
Year 69-76 dates parsed differently depending on API used |
| #137 |
strftime() class method ignores language package → always outputs English |
Date::Language::German->strftime(...) returns English names |
| #136 |
Date::Language::str2time silently ignores EPOCH parameter |
Language users can't use reference epoch for relative dates |
Tier 3 — Features & enhancements
| PR |
What it adds |
Scope |
| #132 |
Parse ISO 8601 ordinal dates (YYYY-DDD) |
New pattern, pure addition |
| #130 |
13 missing timezone abbreviations (ACST, SAST, TRT, etc.) |
Time::Zone only |
| #129 |
%g format code (ISO week year companion to %V) |
Date::Format::Generic |
Tier 4 — Localization & test fixes
| PR |
What it fixes |
Scope |
| #131 |
Arabic March mistranslation + French-borrowed @Dsuf |
1 language module |
| #135 |
Flapping t/gh10.t skip logic + stale POD docs |
Test stability |
Suggested approach
Tiers 1-2 are all independent (no inter-PR conflicts). They can be merged in any order.
Tier 3 features are also independent of each other and of the fixes.
All PRs are draft — happy to undraft any subset you'd like to review first.
🤖 Generated by Kōan — autonomous triage session
PR Queue — Priority Review Guide
13 PRs are open, all CI-green and merge-clean. Here's a suggested review order based on severity and impact.
Tier 1 — Correctness bugs (wrong output)
These fix silent data corruption or wrong results. Users currently get incorrect values.
time2strstores TZ-adjusted epoch instead of real epoch → wrong%Gand%zwhen timezone specifiedformat_zloses negative sign for sub-hour offsets →-0030rendered as+0030strptimefails to parse day 13 in DD-MM-YYYY format → returnsundefformat_oreturns bare "31" instead of "31st" on day 31 (off-by-one in 20 modules)%VISO week wrong at leap year boundaries → week 53 shown as week 01Tier 2 — API consistency bugs
These fix behavioral divergence between
Date::ParseandDate::Language.Date::Language::str2timehandles 2-digit years differently thanDate::Parse::str2timestrftime()class method ignores language package → always outputs EnglishDate::Language::German->strftime(...)returns English namesDate::Language::str2timesilently ignores EPOCH parameterTier 3 — Features & enhancements
%gformat code (ISO week year companion to%V)Tier 4 — Localization & test fixes
Suggested approach
Tiers 1-2 are all independent (no inter-PR conflicts). They can be merged in any order.
Tier 3 features are also independent of each other and of the fixes.
All PRs are draft — happy to undraft any subset you'd like to review first.
🤖 Generated by Kōan — autonomous triage session