chore: complete the shigola rebrand (MAPCO-11504) - #37
Merged
Merged
Conversation
atlas passed a literal "tegola" to Map.Collectors, so the PostGIS pool and query metrics were the one family the metrics rename missed. The grafana dashboard carried a provider_prefix variable to work around it; that goes too. Refs MAPCO-11504 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
debug.go read TEGOLA_SQL_DEBUG with os.Getenv, so the documented SHIGOLA_SQL_DEBUG did nothing. It now goes through the env shim like every other setting: the new name works and the legacy one still works with a deprecation warning. Refs MAPCO-11504 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ests Mechanical, case-preserving: comments, CLI help text, log and metric help strings, test fixtures and the file-cache fixture directory. Test hostnames go to example.com rather than a made-up shigola domain. Comments that described the removed native routes as this server's now say they were removed. Left alone on purpose: upstream issue and PR links, the env shim's legacy prefix, and comments that name Tegola as the project this descends from. Refs MAPCO-11504 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The CITE harness accepted TEGOLA_PORT as a legacy spelling of SHIGOLA_PORT. It is a CI script, not a deployment setting, so there is nothing to keep compatible. The S3 test bucket keeps its tegola-ci name because an existing bucket cannot be renamed; the workflow now says so. Refs MAPCO-11504 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The migration dropped a leftover "tegola" database and role on every run, for volumes created before the fixture was renamed to shigola. That rename landed with MAPCO-11479, so the cleanup and its README note go. Refs MAPCO-11504 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The bulk rename turned "the grid tegola has always served" into a claim about Shigola; those comments describe Tegola's history and now say so. A config test also assembled cdn.shigola.io from parts, a made-up domain the example.com swap missed. Refs MAPCO-11504 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Coverage Report for CI Build 7Coverage increased (+0.09%) to 56.828%Details
Uncovered Changes
Coverage RegressionsNo coverage regressions found. Coverage Stats
💛 - Coveralls |
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.
Removes the last stale
tegolareferences from code, comments, config, tests and workflows. The deliberate exceptions stay. Jira: MAPCO-11504. Docs PR: MapColonies/shigola-docs#18.Base is
development, because feature branches are cut from it and target it.What changed and why
Two changes alter behaviour. Both were bugs the rename had papered over.
shigola_*.atlas.SetObservabilitypassed a literal"tegola"toMap.Collectors, so the PostGIS pool and query metrics were the one family still published astegola_*. README's breaking-changes table already says "nothing emits the old names"; this makes that true. Any dashboard or alert ontegola_postgres_*ortegola_*provider_sql_query_secondsneeds updating. The Grafana dashboard'sprovider_prefixworkaround is removed. Pinned byTestProviderCollectorsArePrefixed.SHIGOLA_SQL_DEBUGnow works. The docs document it, butdebug.goreadTEGOLA_SQL_DEBUGdirectly withos.Getenv, so the documented name did nothing. It now goes through theenv.Getenvshim, so the legacy name still works and warns. Pinned byTestSQLDebugFlags.The rest is a rename:
cache/file/testfiles/tegola-cachedirectory. Test hostnames go toexample.comrather than a made-up shigola domain.serve --helpno longer advertises the removed/maps/...route.TEGOLA_PORT. It is a CI script, not a deployment setting.tegoladatabase. MAPCO-11479 retired that name.Deliberately kept
LICENSE.md,NOTICE.md,CHANGELOG.md,tms/LICENSE-morecantile, and the provenance sections of README and CONTRIBUTING. These are the licence obligation.postgis.go,server.go,seed_purge_generator_test.go).TEGOLA_legacy prefix. It still accepts the old names and still warns.AWS_TEST_BUCKET: tegola-ci. It is an existing S3 bucket, which cannot be renamed. A comment in the workflow now records it as the remaining exception.internal/build/stamp_path_test.gonamesgithub.com/go-spatial/tegola/...on purpose. It is the input for the "path still naming upstream" case.Verification
RUN_POSTGIS_TESTS=yesagainst the fixture DB, under both-raceandCGO_ENABLED=0, on arm64. The only failures were three pre-existing ones that also fail on unmodifieddevelopment:TestValidateTileInGridandTestTile/a_matrix_the_scheme_does_not_havefail on arm64 only.TestTracedRequestPublishesNoNewMetricsis a known flake.go vetis clean, and thegofmt -sgate is clean.development.go-spatialreference survives, except the six ininternal/envtest data, which are URL-parsing fixtures and not attribution.git grep -iE 'shigola\.io|go-spatial/shigola|go-spatial\.org/shigola|MapColonies/tegola'returns nothing.Not done here
shigola_*prefixes, separately from the newatlas.metricPrefix. Sharing one constant would be a small follow-up, but it is outside the scope of a rename.🤖 Generated with Claude Code