chore: track with mossaic action v0.7.0 - #13
Merged
Merged
Conversation
The pin was `@v0.6.0` while `version:` is left on `latest`, so this
repository was already running **mossaic-art 0.7.0** from crates.io through
the **0.6.0 action wrapper** — and the two fixes that matter here live in
`action.yml`, not in the tool:
- **The midnight split.** The 0.6.0 wrapper runs the tracker twice,
`--format json` then `--format markdown`, and each run makes its own
GraphQL query and its own `Local::now()`. Across a local midnight the two
halves land on different days: the gate reads `today-short` from the first
call and the issue comment carries the body of the second. This repository
splits them by design — track.yml gates on `steps.art.outputs.today-short`
and posts `steps.art.outputs.markdown` — and its cron is 23:30 local, with
a `cargo install` ahead of it, so a delayed run lands on the boundary. The
0.7.0 wrapper pins the markdown call to the date the json call reported.
- **The empty subject.** `fail-on: holed` interpolated the `text:` input,
which is empty by construction for a `matrix:` plan like this one, so the
annotation read " can no longer be drawn cleanly this year". `fail-on` is
`never` here today, but it is one line from not being.
Verified against this repository's own `heart.art` and plan inputs before
bumping: the json parses, `start_week` is 35 and `columns` is 11, and the
posted markdown now carries `### Heart · 2026 · week 35, 11 columns` — the
placement guard README.md describes and no report line used to carry. The
holed sentence also reads as a sentence again ("14 days inside the letters
are already lit, and nothing takes them away"); 0.6.3 had moved the verb so
it described the letters rather than the days.
The `("0.5.0")` example in the `version:` comment goes to `("0.7.0")` in the
same pass — it was illustrating a pin two releases behind the one in use.
Signed-off-by: Vyncint Ng <115854244+vyncint@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bumps the action pin from
@v0.6.0to@v0.7.0.Why this is not cosmetic
version:is left onlatestdeliberately — "this repository is where a mossaic release gets exercised against a real contribution graph". So this repository was already running mossaic-art 0.7.0 from crates.io, through the 0.6.0 action wrapper. The two fixes that matter here live inaction.yml, not in the tool, so the bump is what actually delivers them:The midnight split (mossaic#79). The 0.6.0 wrapper runs the tracker twice,
--format jsonthen--format markdown, and each run makes its own GraphQL query and its ownLocal::now(). This repository splits the halves by design —track.ymlgates onsteps.art.outputs.today-short(the json call) and postssteps.art.outputs.markdown(the second). The cron is 23:30 local with acargo installahead of it, so a delayed run lands on the boundary, and then the gate and the message are about different days with nothing in the message saying so. The 0.7.0 wrapper pins the markdown call to the date the json call reported.The empty subject (mossaic#91).
fail-on: holedinterpolated thetext:input, which is empty by construction for amatrix:plan like this one, so the annotation read::error:: can no longer be drawn cleanly this year.fail-onisneverhere today — but it is one line from not being, and this plan's verdict is alreadyholed.Verified before bumping
Run against this repository's own
heart.artand plan inputs with 0.7.0:Two visible changes in what gets posted, both improvements:
· week 35, 11 columns— the placement guard README.md describes ("the plan is these inputs, and changing one compares against a different plan"), which no report line used to carry;Also
The
("0.5.0")example in theversion:comment goes to("0.7.0")— it was illustrating a pin two releases behind the one actually in use.Not done here
start-weekandcolumnsare now outputs (mossaic#88), which is what would turn README's placement guard into an actual assertion — a step that fails ifsteps.art.outputs.start-week != '35'. That is a new CI step rather than a version bump, so it is left for a follow-up.zizmor --persona=pedanticreports 20 findings on.github/workflows/here, byte-identical before and after this change — pre-existing, and this repository has no zizmor job. Out of scope for a bump, but worth its own issue: mossaic fixed the equivalent in its shipped example template in 0.7.0.