Repository navigation
chore(main): release 0.0.1-alpha.88 - #569
Merged
fluxopt-release-bot[bot] merged 1 commit intoAug 11, 2026
Merged
fluxopt-release-bot[bot] merged 1 commit into
fluxopt-release-bot[bot] merged 1 commit into
Conversation
fluxopt-release-bot
Bot
added
the
autorelease: pending
Merging this PR will degrade performance by 2.07%
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Memory | test_emit[sector-s-lpspec-highs] |
3.9 MB | 4.4 MB | -9.95% |
| ⚡ | Memory | test_emit[profiled-s-lpspec-highs] |
27 MB | 25.4 MB | +6.49% |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing release-please--branches--main--components--lpspec (0037a3d) with main (6847419)
Footnotes
-
38 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩
Contributor
Author
|
🤖 Created releases: 🌻 |
fluxopt-release-bot
Bot
added
autorelease: tagged
FBumann
added a commit
that referenced
this pull request
Aug 11, 2026
`cancel-in-progress` was unconditional, and on `main` that quietly punches holes in the series the metered job exists to produce. Four consecutive commits lost their walltime on 2026-08-11 — #511, #567, #568, #569, merged four minutes apart, each cancelling the one before. Only #565's own run survived, which is two engine changes older than the tip. That is not a cosmetic gap. CodSpeed stores one number per commit and compares a pull request against its own merge base, so a cancelled run is a base the next PR cannot be read against: the first one opened after that burst had nothing to compare to but a commit predating the bound attach, which would have flattered it on exactly the case the attach speeds up. On a branch the behaviour is right and stays: the newest push is the only one anybody wants a number for. The cost is stated in the file rather than hidden. A burst of N merges now takes N metered macro runs where it took one — the price of a complete series, which is the whole reason this job is fenced to merges.
FBumann
added a commit
that referenced
this pull request
Aug 11, 2026
`cancel-in-progress` was unconditional, and on `main` that punches holes in the series the metered job exists to produce. Four consecutive commits lost their walltime on 2026-08-11 — #511, #567, #568, #569, merged four minutes apart, each cancelling the one before. Only #565's own run survived, two engine changes older than the tip. That is not cosmetic. CodSpeed keeps one number per commit and reads a pull request against its own merge base, so a cancelled run is a base the next PR cannot be read against: the first one opened after that burst would have been compared to a commit predating the bound attach, flattering it on exactly the case the attach speeds up. Keyed on the **ref**, not the event. `main` is the only ref whose runs are a series; every other one — a pull request, or a `workflow_dispatch` CodSpeed fires at some branch to backfill — wants its newest push to supersede the run in flight, which is the behaviour that was already right. An event-name test would have caught `push` and let dispatch stop cancelling as a side effect. The cost is that a burst of N merges takes N metered runs where it took one.
FBumann
added a commit
that referenced
this pull request
Aug 11, 2026
`cancel-in-progress` was unconditional, and on `main` that punches holes in the series the metered job exists to produce. Four consecutive commits lost their walltime on 2026-08-11 — #511, #567, #568, #569, merged four minutes apart, each cancelling the one before. Only #565's own run survived, two engine changes older than the tip. That is not cosmetic. CodSpeed keeps one number per commit and reads a pull request against its own merge base, so a cancelled run is a base the next PR cannot be read against: the first one opened after that burst would have been compared to a commit predating the bound attach, flattering it on exactly the case the attach speeds up. Keyed on the **ref**, not the event. `main` is the only ref whose runs are a series; every other one — a pull request, or a `workflow_dispatch` CodSpeed fires at some branch to backfill — wants its newest push to supersede the run in flight, which is the behaviour that was already right. An event-name test would have caught `push` and let dispatch stop cancelling as a side effect. The cost is that a burst of N merges takes N metered runs where it took one.
FBumann
pushed a commit
that referenced
this pull request
Sep 19, 2026
…hey run along math-spec#569 renames the axis key on `piecewise:` and `sos:` from `over:` to `along:`, and lpspec follows it: every model, example, fixture and documented model takes the new key, and `curves.py`, `assembly.py` and the linopy builder read `along` off the declarations. The language already separated `over=`, which consumes a dimension, from `along=`, which keeps it and keeps it ordered. A breakpoint axis is ordered and survives into the weights, so it was on the wrong side; `sos:` moves with it because `method: sos2` hands one block's axis straight to the other's. `Buildable` widens to `Mapping[str, object]`, which is what `to_spec` now takes, so `test_the_model_argument_is_what_the_language_takes_minus_the_lowered_form` holds again. Two test expectations follow upstream wording rather than behaviour: the sos dim refusal now says `along 'other' is not a dim`, and a stray values dim is judged against its own link's row frame rather than against every link expression. The pin is PROVISIONAL — a math-spec branch, because #569 is unreleased. Repin to the first tag carrying it before merge. Ran `pytest` (3778 passed, 10 failed, 337 skipped, 1 xfailed — the 10 are the gurobi and xpress extras, not installed here, each failing with the package's own message saying so), `ruff check`, `ruff format --check` and `pyrefly check`. pyrefly is 6 errors: the base branch with this same pin installed reports 15, the extra nine being the `.over` reads this commit renames, and the 6 that remain are the identical set the base reports on its own pin. Did not run `pixi run check` as a whole, the differential harness or the benchmarks. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Hh3uK8SqVcXKuLnYHAjrVW
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.
🤖 I have created a release beep boop
0.0.1-alpha.88 (2026-08-11)
Performance
This PR was generated with Release Please. See documentation.