-
Notifications
You must be signed in to change notification settings - Fork 72
Comparing changes
Open a pull request
base repository: DataDog/datadog-ci
base: v5.20.1
head repository: DataDog/datadog-ci
compare: v5.21.0
- 16 commits
- 37 files changed
- 9 contributors
Commits on Jun 29, 2026
-
Configuration menu - View commit details
-
Copy full SHA for 569002c - Browse repository at this point
Copy the full SHA 569002cView commit details
Commits on Jun 30, 2026
-
Configuration menu - View commit details
-
Copy full SHA for 0d7f957 - Browse repository at this point
Copy the full SHA 0d7f957View commit details -
fix(coverage): fetch base branch before merge-base for shallow-clone CIs
Some CI environments (e.g. Bitbucket Pipelines) only clone the current branch by default, leaving the base branch ref unavailable locally. This causes git merge-base to fail silently and patch coverage to be missing from the uploaded event. Explicitly fetch origin/<baseBranch> before computing the merge-base, guarded in a try/catch so the fix degrades gracefully on fetch failure. GitHub Actions and GitLab CI are unaffected — they supply a base SHA directly and take the earlier code path. Azure/CircleCI/Jenkins hit this path but already have all refs available, so the fetch is a no-op.
Configuration menu - View commit details
-
Copy full SHA for fb4f452 - Browse repository at this point
Copy the full SHA fb4f452View commit details -
Configuration menu - View commit details
-
Copy full SHA for 2d433a0 - Browse repository at this point
Copy the full SHA 2d433a0View commit details -
Configuration menu - View commit details
-
Copy full SHA for 87d3ea0 - Browse repository at this point
Copy the full SHA 87d3ea0View commit details -
fix(ci): map BITBUCKET_PR_DESTINATION_COMMIT to base branch SHA
Bitbucket Pipelines merges the destination branch into the source at build setup, so destination-branch commits are reachable in PR pipeline history. However origin/<baseBranch> is absent as a remote-tracking ref, causing git merge-base to fail silently and patch coverage to be missing. Rather than fetching the ref (previous approach, reverted), map BITBUCKET_PR_DESTINATION_COMMIT to GIT_PULL_REQUEST_BASE_BRANCH_SHA in the CI detection layer. datadog-ci then takes the direct baseSha code path — same as GitHub Actions (pull_request.base.sha) and GitLab CI (CI_MERGE_REQUEST_DIFF_BASE_SHA) — with no git operations needed. Non-PR pipelines are unaffected: BITBUCKET_PR_DESTINATION_COMMIT is only set in PR pipeline contexts, where patch coverage is meaningful.
Configuration menu - View commit details
-
Copy full SHA for bbc8de6 - Browse repository at this point
Copy the full SHA bbc8de6View commit details -
Configuration menu - View commit details
-
Copy full SHA for 1df71b2 - Browse repository at this point
Copy the full SHA 1df71b2View commit details -
Merge pull request #2394 from DataDog/alba/fix-coverage-fetch-base-br…
…anch fix(ci): map BITBUCKET_PR_DESTINATION_COMMIT to base branch SHA for patch coverage
Configuration menu - View commit details
-
Copy full SHA for db28094 - Browse repository at this point
Copy the full SHA db28094View commit details -
Add uk1.datadoghq.com as a supported Datadog site
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for de581b7 - Browse repository at this point
Copy the full SHA de581b7View commit details
Commits on Jul 2, 2026
-
Configuration menu - View commit details
-
Copy full SHA for 0a6276f - Browse repository at this point
Copy the full SHA 0a6276fView commit details -
Merge pull request #2395 from DataDog/tal-usvyatsky/add-uk1.datadoghq…
….com Add uk1.datadoghq.com as a supported Datadog site
Configuration menu - View commit details
-
Copy full SHA for d158a63 - Browse repository at this point
Copy the full SHA d158a63View commit details -
feat(tag): comma-separated
--level(e.g.--level pipeline,job) (#……2388) * feat(tag,measure): support additive comma-separated --level Allow `--level pipeline,job` so the same tags/measures are applied to multiple span levels in one invocation. Levels are deduped and validated fail-fast; one request is sent per level. Adds a shared per-level loop (`processLevels`) used by both `tag` and `measure`, and a per-execution cache so the GitHub Actions diagnostic logs are located and read once even when several levels need them (e.g. `--level job,step`). Single-level usage and `--no-fail` semantics are unchanged: `--no-fail` still tolerates a failed submission but not setup/config errors, which are reported with the offending level. Refs #2387 Co-Authored-By: Claude Opus 4.8 <[email protected]> * fix(tag,measure): deterministically mock fs in fatal-level tests These tests assumed the host filesystem lacks GitHub Worker_*.log diagnostic files under the well-known fallback dirs to force the `step` enrichment to fail. That assumption doesn't hold on real GitHub Actions runners, causing flaky CI failures. Mock fs.readdirSync to throw ENOENT explicitly instead. * fix(tag,measure): drop multi-level support from measure, polish tag README Reviewer feedback on #2388: measure semantics for multi-level are unclear (should measures be aggregated/overwritten per level instead of duplicated?), so revert measure back to single-level --level. tag keeps additive comma-separated --level. Also applies 3 README wording nits for tag. Co-Authored-By: Claude Sonnet 5 <[email protected]> * Clean up and remove technical details --------- Co-authored-by: Claude Opus 4.8 <[email protected]> Co-authored-by: Corentin Girard <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 3bac124 - Browse repository at this point
Copy the full SHA 3bac124View commit details
Commits on Jul 6, 2026
-
feat(trace): add --span-id and --parent-id to
trace spanLet callers assign an explicit span ID and nest a custom span under another custom span, instead of every span being flat under the job. --span-id defaults to a generated ID when omitted; --parent-id references another span's ID and attaches to the job/step as before when omitted. Both are validated as hexadecimal. This makes it possible to assemble arbitrary span trees (e.g. from an OpenTelemetry dump) via repeated `trace span` calls. Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e583ac4 - Browse repository at this point
Copy the full SHA e583ac4View commit details -
Merge pull request #2397 from DataDog/rodrigo.roca/span-command-span-…
…id-parent-id Add --span-id and --parent-id options to `trace span`
Configuration menu - View commit details
-
Copy full SHA for d80a1cf - Browse repository at this point
Copy the full SHA d80a1cfView commit details -
docs(trace): surface the
trace spansubcommand (#2396)* docs(trace): surface the `trace span` subcommand The `trace span` subcommand (report a standalone custom span with a name and a duration or start/end time) ships and has its own README, but it was not discoverable from the top-level command index: the `trace` section only described the command-wrapping form and never mentioned or linked `span`. - List `span` under the `trace` section in the root README, linking its README. - Add a cross-reference from the `trace` README to the `trace span` subcommand. Docs only; no behavior change. Co-Authored-By: Claude Opus 4.8 <[email protected]> * docs(span): make epoch timestamps explicit about UTC `--start-time` / `--end-time` are epoch milliseconds, i.e. an absolute UTC instant (serialized via `toISOString()`), but "unix epoch" alone can read as ambiguous. Spell out UTC. Co-Authored-By: Claude Opus 4.8 <[email protected]> * Make front readme less verbose --------- Co-authored-by: Claude Opus 4.8 <[email protected]> Co-authored-by: Corentin Girard <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 2912bfd - Browse repository at this point
Copy the full SHA 2912bfdView commit details -
Configuration menu - View commit details
-
Copy full SHA for 4d1587a - Browse repository at this point
Copy the full SHA 4d1587aView commit details
This comparison is taking too long to generate.
Unfortunately it looks like we can’t render this comparison for you right now. It might be too big, or there might be something weird with your repository.
You can try running this command locally to see the comparison on your machine:
git diff v5.20.1...v5.21.0