Update openblas-experimental from 0.3.35.dev0 to 0.3.36.0dev0 - #6698
Open
emscripten-forge-bot wants to merge 1 commit into
Open
emscripten-forge-bot wants to merge 1 commit into
emscripten-forge-bot wants to merge 1 commit into
Conversation
emscripten-forge-bot
added
the
Automerge
IsabelParedes
added
DO NOT MERGE
Member
|
Do not close this PR. This should prevent the bot from trying to update again. |
emscripten-forge-bot
added
the
Needs Human Review
Contributor
Author
|
Either the CI is failing, or the recipe is not tested. I need help from a human. |
MMesch
added a commit
that referenced
this pull request
Sep 15, 2026
Refactor the version-bump bot into a plan/execute pipeline with four named stages, and expose them as a --mode flag so operators can run any prefix of the pipeline instead of only "do everything or nothing." emci/bot/bump_recipes_versions.py: - Introduce Mode enum (check/plan/edit/submit) and four stage functions with a 1:1 correspondence to modes: check_bump, plan_bump, edit_bump, submit_bump. Each stage strictly adds one action to the previous. - Introduce composed dataclasses: Candidate (produced by check_bump) and BumpAction (Candidate + sha256, produced by plan_bump). - Rename get_new_version -> find_new_version_url. It now returns the URL (not the sha256), returns the recipe's current version even when no newer release is found, and skips candidate versions whose rendered URL equals the current one -- stops the bot from opening no-source- change "bump" PRs on commit-pinned recipes whose URL template doesn't reference version (#6698, #6570). - Move the previously-inlined skip list to module scope as SKIP_RECIPES; add discover_recipes() helper (single source of truth for both real and preview runs). - Extract _process_existing_bot_prs() and _checkout_target_branch() so the dispatcher clearly separates side-effect setup from planning. - Stage functions are print-silent; the dispatcher owns all output. Per-recipe status lines: "BUMP" / "no bump" / "error". emci/__main__.py: - Add --mode {check,plan,edit,submit}, default check. - Add --limit N (default 20). Counts successful completions of the terminal stage, so --limit 20 in submit mode means 20 real PRs. .github/workflows/new_versions.yaml: - Pass --mode submit explicitly (default flipped to check). No other behavior change: --limit default of 20 matches the old pr_limit=20.
MMesch
added a commit
that referenced
this pull request
Sep 15, 2026
Refactor the version-bump bot into a plan/execute pipeline with four named stages, and expose them as a --mode flag so operators can run any prefix of the pipeline instead of only "do everything or nothing." emci/bot/bump_recipes_versions.py: - Introduce Mode enum (check/plan/edit/submit) and four stage functions with a 1:1 correspondence to modes: check_bump, plan_bump, edit_bump, submit_bump. Each stage strictly adds one action to the previous. - Introduce composed dataclasses: Candidate (produced by check_bump) and BumpAction (Candidate + sha256, produced by plan_bump). - Rename get_new_version -> find_new_version_url. It now returns the URL (not the sha256), returns the recipe's current version even when no newer release is found, and skips candidate versions whose rendered URL equals the current one -- stops the bot from opening no-source- change "bump" PRs on commit-pinned recipes whose URL template doesn't reference version (#6698, #6570). - Move the previously-inlined skip list to module scope as SKIP_RECIPES; add discover_recipes() helper (single source of truth for both real and preview runs). - Extract _process_existing_bot_prs() and _checkout_target_branch() so the dispatcher clearly separates side-effect setup from planning. - Stage functions are print-silent; the dispatcher owns all output. Per-recipe status lines: "BUMP" / "no bump" / "error". emci/__main__.py: - Add --mode {check,plan,edit,submit}, default check. - Add --limit N (default 20). Counts successful completions of the terminal stage, so --limit 20 in submit mode means 20 real PRs. .github/workflows/new_versions.yaml: - Pass --mode submit explicitly (default flipped to check). No other behavior change: --limit default of 20 matches the old pr_limit=20.
MMesch
added a commit
that referenced
this pull request
Sep 15, 2026
Refactor the version-bump bot into a plan/execute pipeline with four named stages, and expose them as a --mode flag so operators can run any prefix of the pipeline instead of only "do everything or nothing." emci/bot/bump_recipes_versions.py: - Introduce Mode enum (check/plan/edit/submit) and four stage functions with a 1:1 correspondence to modes: check_bump, plan_bump, edit_bump, submit_bump. Each stage strictly adds one action to the previous. - Introduce composed dataclasses: Candidate (produced by check_bump) and BumpAction (Candidate + sha256, produced by plan_bump). - Rename get_new_version -> find_new_version_url. It now returns the URL (not the sha256), returns the recipe's current version even when no newer release is found, and skips candidate versions whose rendered URL equals the current one -- stops the bot from opening no-source- change "bump" PRs on commit-pinned recipes whose URL template doesn't reference version (#6698, #6570). - Move the previously-inlined skip list to module scope as SKIP_RECIPES; add discover_recipes() helper (single source of truth for both real and preview runs). - Extract _process_existing_bot_prs() and _checkout_target_branch() so the dispatcher clearly separates side-effect setup from planning. - Stage functions are print-silent; the dispatcher owns all output. Per-recipe status lines: "BUMP" / "no bump" / "error". emci/__main__.py: - Add --mode {check,plan,edit,submit}, default check. - Add --limit N (default 20). Counts successful completions of the terminal stage, so --limit 20 in submit mode means 20 real PRs. .github/workflows/new_versions.yaml: - Pass --mode submit explicitly (default flipped to check). No other behavior change: --limit default of 20 matches the old pr_limit=20.
Collaborator
|
Either the CI is failing, or the recipe is not tested. I need help from a human. |
MMesch
added a commit
that referenced
this pull request
Sep 18, 2026
Refactor the version-bump bot into a plan/execute pipeline with four named stages, and expose them as a --mode flag so operators can run any prefix of the pipeline instead of only "do everything or nothing." emci/bot/bump_recipes_versions.py: - Introduce Mode enum (check/plan/edit/submit) and four stage functions with a 1:1 correspondence to modes: check_bump, plan_bump, edit_bump, submit_bump. Each stage strictly adds one action to the previous. - Introduce composed dataclasses: Candidate (produced by check_bump) and BumpAction (Candidate + sha256, produced by plan_bump). - Rename get_new_version -> find_new_version_url. It now returns the URL (not the sha256), returns the recipe's current version even when no newer release is found, and skips candidate versions whose rendered URL equals the current one -- stops the bot from opening no-source- change "bump" PRs on commit-pinned recipes whose URL template doesn't reference version (#6698, #6570). - Move the previously-inlined skip list to module scope as SKIP_RECIPES; add discover_recipes() helper (single source of truth for both real and preview runs). - Extract _process_existing_bot_prs() and _checkout_target_branch() so the dispatcher clearly separates side-effect setup from planning. - Stage functions are print-silent; the dispatcher owns all output. Per-recipe status lines: "BUMP" / "no bump" / "error". emci/__main__.py: - Add --mode {check,plan,edit,submit}, default check. - Add --limit N (default 20). Counts successful completions of the terminal stage, so --limit 20 in submit mode means 20 real PRs. .github/workflows/new_versions.yaml: - Pass --mode submit explicitly (default flipped to check). No other behavior change: --limit default of 20 matches the old pr_limit=20.
MMesch
added a commit
that referenced
this pull request
Sep 18, 2026
Restructures the version-bump bot around two ideas: a stage ladder for "how far the run goes" and a shared ops model for "execute vs preview." The two phases the bot used to run together (merging its own open PRs and opening new bumps) are split into independent subcommands. Mode ladder for `bump-recipes-versions` --------------------------------------- Each mode does its own stage plus every earlier one; default is `check`. check HEAD candidate URLs until one exists. No downloads. plan + download the tarball and compute sha256. edit + create a local branch, write recipe.yaml, commit (no push). submit + push the branch and open the PR. CI passes this explicitly. Dry-run (orthogonal flag) ------------------------- --dry-run turns every write / git / gh mutation into a printed line and shows a unified diff of the recipe.yaml change instead of writing it. Reads (HEADs, gh pr list, gh pr checks, gh pr view) still run so decisions are made against real state. `--mode submit --dry-run` is a full rehearsal that touches nothing. Ops model --------- Stage functions (edit_bump, submit_bump, merge processing, target checkout) build a list of SubprocessCmd / FileWrite dataclasses instead of invoking side effects directly. A single execute() runs them for real; print_ops() prints them without running. Dry-run picks the printer — same op list, no drift possible. Data classes ------------ Candidate (post-check) and BumpAction (BumpAction inherits from Candidate, adds sha256 + edit_ops + edit_cleanup + submit_ops). plan_bump builds the whole plan up front; edit_bump/submit_bump are thin runners. URL-unchanged skip ------------------ In find_new_version_url, skip candidate versions whose rendered URL matches the current one. Stops the bot from opening no-source-change "bump" PRs on commit-pinned recipes whose URL template doesn't reference version (#6698, #6570). Split subcommands ----------------- - bump-recipes-versions: open new bump PRs. In submit mode, fetches open bot PRs read-only to skip recipes with an in-flight PR (skipped entirely when --recipe is given). - merge-open-prs: NEW subcommand for the merge/label pass over the bot's already-open PRs. Same --dry-run semantics. - CI (.github/workflows/new_versions.yaml) runs the two as separate steps for independent failure signal. Other flags on bump-recipes-versions ------------------------------------ - --limit N (default 20): stop after N successful completions of the terminal stage — 20 real PRs in submit, 20 hashed candidates in plan. - --recipe NAME (repeatable): scope to specific recipes. Misc ---- - Normalise recipe.yaml block sequences to 2-space dashes on write (ruamel default; standardises across recipes as bumps land). - Bot-authored commits use `git -c user.name=... -c user.email=...` on the one commit call instead of `git config --global`, so --mode edit works on read-only $HOME. - shlex.join on dry-run argv output — copy-pasteable command lines. - One "Would" line per PR grouped with that PR's header. - Loud DRY RUN banner top and bottom of a dry-run run. - emci/README.md updated with both subcommands and worked examples. Removed helpers no longer needed in this file: bot_github_user_ctx, set_bot_user (still used by update_matplotlib_fontcache.py via git_utils.py; that module is untouched).
This branch has not been deployed
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
3 participants
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.
Beep-boop-beep! Whistle-whistle-woo!