README: the GitLab component was never real, and we vendor 1.0.3 - #7
Closed
pavel-te wants to merge 1 commit into
Closed
README: the GitLab component was never real, and we vendor 1.0.3#7pavel-te wants to merge 1 commit into
pavel-te wants to merge 1 commit into
Conversation
…r 1.0.3 Three claims on the front page were false as of v1.1.1. "Both pin ptc-cli v1.0.2" — we vendor 1.0.3 since #6. The number was never going to stay right by hand; it now points at the tree the action actually ships, and the self-test already asserts the vendored VERSION. "GitLab: a CI/CD Catalog component" with an include: component: recipe. That address can never resolve for anyone: GitLab resolves it against the reader's own instance, so a component we publish on one server is unreachable from gitlab.com and from every self-hosted install. ci18-7363 established this and moved `ptc init` to printing an inline job — but the README kept advertising the component, and templates/translate/template.yml kept sitting in a GitHub repo where no GitLab could read it. Both are gone. The GitLab section now carries the same job ptc init prints, including the PTC_GIT_PUSH_TOKEN fallback and the [skip ci] guard, plus the sha256 for anyone who wants the integrity check vendoring gives the GitHub side for free. "[skip translations] marker ... as a second guard" — GitHub honours only [skip ci], [ci skip], [no ci], [skip actions] and [actions skip]. The marker is a label, not a guard; loop safety comes from the paths: trigger and the fact that ptc/translations is not a trigger branch. Said so, rather than letting a reader design around a token that does nothing. Also drops config-file from the quick start: ci18-7277 made api-token the only input you must pass, and a root .ptc-config.yml is picked up on its own. The self-test step that parsed templates/ now asserts the directory stays gone, so a component cannot come back by accident.
pavel-te
changed the title
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
2 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.
Three claims on the front page were false as of v1.1.1.
"Both pin ptc-cli v1.0.2" — we vendor 1.0.3 since #6. The number was never
going to stay right by hand; it now points at the tree the action actually
ships, and the self-test already asserts the vendored VERSION.
"GitLab: a CI/CD Catalog component" with an include: component: recipe. That
address can never resolve for anyone: GitLab resolves it against the reader's
own instance, so a component we publish on one server is unreachable from
gitlab.com and from every self-hosted install.
ptc initmoved to printing aninline job for that reason — but the README kept advertising the component,
and templates/translate/template.yml kept sitting in a GitHub repo where no
GitLab could read it. Both are gone. The GitLab section now carries the same
job ptc init prints, including the PTC_GIT_PUSH_TOKEN fallback and the
[skip ci] guard, plus the sha256 for anyone who wants the integrity check
vendoring gives the GitHub side for free.
"[skip translations] marker ... as a second guard" — GitHub honours only
[skip ci], [ci skip], [no ci], [skip actions] and [actions skip]. The marker
is a label, not a guard; loop safety comes from the paths: trigger and the
fact that ptc/translations is not a trigger branch. Said so, rather than
letting a reader design around a token that does nothing.
Also drops config-file from the quick start: api-token is the only input you
must pass, and a root .ptc-config.yml is picked up on its own.
The self-test step that parsed templates/ now asserts the directory stays
gone, so a component cannot come back by accident.