feat(ci): read merge-queue metadata as a generic value, tolerating key renames#1723
Merged
Conversation
…y renames `queue_metadata::parse_yaml_block` modeled the MQ draft-PR-body payload with a rigid serde struct whose `draft_pr_number` was required. The engine (MRGFY-8075) renamed that key to `batch_pr_number` and dual-emits both for one release window before dropping the old spelling. A missing required field fails the *whole document*, and the caller's `.ok()` swallows that into `None`, so once the alias is gone every batch body with a non-empty `previous_failed_batches` would silently cost `git-refs` its `checking_base_sha` and diff against the wrong base, with no error surfaced anywhere. That struct has been dead code since #1600: its only reader, `git_refs`, uses just `checking_base_sha`; every other field is parsed and discarded. So rather than a `#[serde(alias)]` band-aid on one key, parse the payload into a generic `serde_json::Value` — exactly how the git-note twin (`git::read_note`) already reads the identical payload — and pull `checking_base_sha` out with the existing shared helper. The CLI is now immune to any engine key rename in this block, with no schema left to drift and no call-site churn. While here: warn instead of silently falling through to the PR base when MQ metadata is present but carries no `checking_base_sha`, so a future rename of that key can't degrade the base silently either; and share the YAML-mapping parse between the note and PR-body readers so they can't diverge in what they accept. Fixes MRGFY-8104 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Change-Id: I6732b7057db0c98a7bb275d95fbd6470e39bad8b
Contributor
Merge Protections🟢 All 6 merge protections satisfied — ready to merge. Show 6 satisfied protections🟢 🤖 Continuous Integration
🟢 👀 Review Requirements
🟢 Enforce conventional commitMake sure that we follow https://www.conventionalcommits.org/en/v1.0.0/
🟢 🔎 Reviews
🟢 📕 PR description
🟢 🚦 Auto-queueWhen all merge protections are satisfied, this pull request will be queued automatically. |
There was a problem hiding this comment.
Pull request overview
This PR hardens mergify-ci’s merge-queue metadata extraction against engine-side YAML key renames by parsing the fenced PR-body block (and git-note payload) into a generic serde_json::Value, then extracting only checking_base_sha via a shared helper. It also makes silent “wrong base” fallthroughs observable by emitting a warning when MQ metadata exists but lacks checking_base_sha.
Changes:
- Replace rigid
serdestructs for PR-body MQ metadata with a genericserde_json::Valueparse, mirroring the existing git-note reader approach. - Share a single YAML-mapping parser (
parse_yaml_mapping) between git-note and PR-body readers to keep accepted shapes in sync. - Warn (human-mode only) when MQ metadata is present but missing
checking_base_sha, and add regression tests for the reported failure mode.
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
| crates/mergify-ci/src/queue_metadata.rs | Switch PR-body fenced YAML parsing to a generic value and expand tests to cover key renames/unknown keys/non-mapping payloads. |
| crates/mergify-ci/src/git.rs | Introduce shared parse_yaml_mapping used by both note and PR-body parsing paths, keeping behavior aligned. |
| crates/mergify-ci/src/git_refs.rs | Extract checking_base_sha from generic metadata, warn on missing key, and add tests for the reported regression and warning behavior. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
sileht
marked this pull request as ready for review
July 17, 2026 17:05
JulianMaurin
approved these changes
Jul 18, 2026
kozlek
approved these changes
Jul 20, 2026
Contributor
Merge Queue Status
This pull request spent 7 minutes 9 seconds in the queue, including 6 minutes 52 seconds running CI. Required conditions to merge
|
40 tasks
mergify
Bot
deleted the
devs/sileht/mrgfy-8104-cli-read-batch-pr-number/read-merge-queue-metadata-generic-value--6732b705
branch
July 20, 2026 07:52
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.
queue_metadata::parse_yaml_blockmodeled the MQ draft-PR-body payloadwith a rigid serde struct whose
draft_pr_numberwas required. Theengine (MRGFY-8075) renamed that key to
batch_pr_numberand dual-emitsboth for one release window before dropping the old spelling. A missing
required field fails the whole document, and the caller's
.ok()swallows that into
None, so once the alias is gone every batch bodywith a non-empty
previous_failed_batcheswould silently costgit-refsits
checking_base_shaand diff against the wrong base, with no errorsurfaced anywhere.
That struct has been dead code since #1600: its only reader,
git_refs,uses just
checking_base_sha; every other field is parsed and discarded.So rather than a
#[serde(alias)]band-aid on one key, parse the payloadinto a generic
serde_json::Value— exactly how the git-note twin(
git::read_note) already reads the identical payload — and pullchecking_base_shaout with the existing shared helper. The CLI is nowimmune to any engine key rename in this block, with no schema left to
drift and no call-site churn.
While here: warn instead of silently falling through to the PR base when
MQ metadata is present but carries no
checking_base_sha, so a futurerename of that key can't degrade the base silently either; and share the
YAML-mapping parse between the note and PR-body readers so they can't
diverge in what they accept.
Fixes MRGFY-8104
Co-Authored-By: Claude Opus 4.8 (1M context) noreply@anthropic.com