Proof Portal
Omarchy
ProbeLabs65 findings · 95 requirementsA proof layer — requirements, tests and verified fixes — for two of Omarchy's subsystems: the application menu (launcher scripts, QML model, JSONC config, search and selection) and the lock screen (lock scripts, QML, PAM authentication). Scope is deliberately limited to those components of omacom/omarchy; the rest of the distribution is not covered.
Calendar grid is off by one day in timezones whose DST transition is at midnight
Reported by mquezada · filed Sep 14, 2026Closed upstream on 2026-09-21 by bjarneo via an automated message, as a duplicate of #11126 (R-734). GitHub close reason: completed.
At a glance
- 1 · What is claimed?
- Only the report text below — no validator has restated it as checkable claims yet.
- 2 · What does it affect?
- Not yet mapped to any requirement or component.
- 3 · Is it reproducible?
- No validation has been run yet.
- 4 · What remains unclear?
- Not validated yet: no claims, no reproducer.
- 5 · What happens next?
- Our staff decide whether the report is valid. Where this project allows it, agents decide whether it duplicates another report or a known issue. No decision has been recorded yet.
Original report
The clock panel's calendar grid repeats a day and shifts every following day one column to the left, for the whole month, in any timezone whose DST transition happens at midnight.
<img width="564" height="463" alt="Image" src="https://github.com/user-attachments/assets/80bfe0dd-8d2c-4bef-8e1d-3abc2aaf1d92" />
Seen on America/Santiago, September 2026. Chile springs forward at midnight on Sat Sep 5 -> Sun Sep 6, so 2026-09-06 00:00 does not exist locally:
$ zdump -v America/Santiago | grep 2026
America/Santiago Sun Sep 6 03:59:59 2026 UT = Sat Sep 5 23:59:59 2026 -04 isdst=0
America/Santiago Sun Sep 6 04:00:00 2026 UT = Sun Sep 6 01:00:00 2026 -03 isdst=1What happens: the second row starts at 5 instead of 6, so the 5th appears twice (Saturday and Sunday) and everything after it sits under the wrong weekday. Today, Monday Sep 14, is drawn under TUE. Sep 30 is drawn under THU when it is a Wednesday. The bar clock itself is correct, so the bar and the calendar it opens disagree.
What I expected: the grid to match the calendar, with each day under its own weekday, as it does in any month without a midnight DST transition.
Reproduction
sudo timedatectl set-timezone America/Santiago
# open the clock panel, step to September 2026The row numbering (ISO week column) stays correct, because isoWeek() works in UTC. Only the day cells are wrong.
Cause
shell/plugins/panels/clock/Model.js, monthGrid(): the cursor is built at local midnight and advanced with setDate(getDate() + 1).
var leading = (new Date(year, month, 1).getDay() - start + 7) % 7
var cursor = new Date(year, month, 1 - leading)
...
cursor.setDate(cursor.getDate() + 1)When the resulting local time falls in a DST gap, QML's JS engine resolves it backwards, to 23:00 on the previous day, rather than forwards past the gap. That is the difference from V8, where the same code is correct, so this does not reproduce in a browser:
qml6: 8th cell after Aug 30 -> day 5 (wrong, should be 6)
node: 8th cell after Aug 30 -> Sun Sep 06 01:00 (correct)Suggested fix
Anchor the cells at midday, which cannot fall in a DST gap since transitions are an hour:
var leading = (new Date(year, month, 1, 12).getDay() - start + 7) % 7
var cursor = new Date(year, month, 1 - leading, 12)With that change the grid is correct across the September 2026 transition, checked in the QML engine itself (0 wrong cells over the two weeks spanning the gap).
stepMonth() builds new Date(year, month + delta, 1) the same way and has the same latent problem: in a timezone where a month's 1st is the transition day, stepping would land on the previous month. Chile transitions on a Sunday in September so it does not bite there, but the same midday anchor fixes it.
Environment
- omarchy 4.0.3-1
- quickshell 0.3.1-1, qt6-declarative 6.11.2-1
- TZ America/Santiago
Filed by Claude Opus 5 via Claude Code.
View raw text
The clock panel's calendar grid repeats a day and shifts every following day one column to the left, for the whole month, in any timezone whose DST transition happens at midnight. <img width="564" height="463" alt="Image" src="https://github.com/user-attachments/assets/80bfe0dd-8d2c-4bef-8e1d-3abc2aaf1d92" /> Seen on America/Santiago, September 2026. Chile springs forward at midnight on Sat Sep 5 -> Sun Sep 6, so `2026-09-06 00:00` does not exist locally: ``` $ zdump -v America/Santiago | grep 2026 America/Santiago Sun Sep 6 03:59:59 2026 UT = Sat Sep 5 23:59:59 2026 -04 isdst=0 America/Santiago Sun Sep 6 04:00:00 2026 UT = Sun Sep 6 01:00:00 2026 -03 isdst=1 ``` **What happens:** the second row starts at 5 instead of 6, so the 5th appears twice (Saturday and Sunday) and everything after it sits under the wrong weekday. Today, Monday Sep 14, is drawn under TUE. Sep 30 is drawn under THU when it is a Wednesday. The bar clock itself is correct, so the bar and the calendar it opens disagree. **What I expected:** the grid to match the calendar, with each day under its own weekday, as it does in any month without a midnight DST transition. ## Reproduction ```bash sudo timedatectl set-timezone America/Santiago # open the clock panel, step to September 2026 ``` The row numbering (ISO week column) stays correct, because `isoWeek()` works in UTC. Only the day cells are wrong. ## Cause `shell/plugins/panels/clock/Model.js`, `monthGrid()`: the cursor is built at local midnight and advanced with `setDate(getDate() + 1)`. ```js var leading = (new Date(year, month, 1).getDay() - start + 7) % 7 var cursor = new Date(year, month, 1 - leading) ... cursor.setDate(cursor.getDate() + 1) ``` When the resulting local time falls in a DST gap, QML's JS engine resolves it *backwards*, to 23:00 on the previous day, rather than forwards past the gap. That is the difference from V8, where the same code is correct, so this does not reproduce in a browser: ``` qml6: 8th cell after Aug 30 -> day 5 (wrong, should be 6) node: 8th cell after Aug 30 -> Sun Sep 06 01:00 (correct) ``` ## Suggested fix Anchor the cells at midday, which cannot fall in a DST gap since transitions are an hour: ```js var leading = (new Date(year, month, 1, 12).getDay() - start + 7) % 7 var cursor = new Date(year, month, 1 - leading, 12) ``` With that change the grid is correct across the September 2026 transition, checked in the QML engine itself (0 wrong cells over the two weeks spanning the gap). `stepMonth()` builds `new Date(year, month + delta, 1)` the same way and has the same latent problem: in a timezone where a month's 1st is the transition day, stepping would land on the previous month. Chile transitions on a Sunday in September so it does not bite there, but the same midday anchor fixes it. ## Environment - omarchy 4.0.3-1 - quickshell 0.3.1-1, qt6-declarative 6.11.2-1 - TZ America/Santiago --- Filed by Claude Opus 5 via Claude Code.
- 80bfe0dd-8d2c-4bef-8e1d-3abc2aaf1d92
Validation
Not checked yet. A validator will restate this report as checkable claims, map it to the project's requirements and write a reproducer that the proof engine runs.
Linked
- Fixed bypr/12943· confirmed
- Fixed bypr/10476· confirmed
- Fixed bypr/13186· confirmed
- Fixed bypr/13751· confirmed
- Fixed bypr/13096· confirmed
- Fixed bypr/13127· confirmed
Decision
Decisions are recorded by our staff — or, where this project allows it, by agents: checked by a second, independent agent, or by a single agent where the project's policy says so. Any decision reached so far is shown below.
No decision has been recorded yet.