Add a changelog, build improvements #227
Conversation
| group 'io.github.eisopux' | ||
| version '0.2.0' | ||
|
|
There was a problem hiding this comment.
Thanks, looks good, will this downloader be used as a dependency in the future?
There was a problem hiding this comment.
Also, do you need to keep all version numbers consistent? Or just different numbers for different repos?
There was a problem hiding this comment.
Thanks, looks good, will this downloader be used as a dependency in the future?
What do you mean with "this downloader" and "as a dependency"?
There was a problem hiding this comment.
Also, do you need to keep all version numbers consistent? Or just different numbers for different repos?
I was thinking whether we should use one version for all three projects, but there is no need to keep them in sync - cf-vscode uses the cf-ls-downloader to download the latest release of eisop/cf and cf-ls. The cf-ls-downloader likely won't need many updates. The cf-ls and the cf-vscode projects hopefully can evolve separately and, ideally, we'll have other IDEs that use the same LSP also. So I think having separate versions is OK.
What do you think?
There was a problem hiding this comment.
Ok, I agree now. Each extension should have a separate version number for developing. And if we think the downloader is stable, we can make a 1.0.0 release.
There was a problem hiding this comment.
Thanks, looks good, will this downloader be used as a dependency in the future?
What do you mean with "this downloader" and "as a dependency"?
I am just wondering why adding metadata in the project.
There was a problem hiding this comment.
I am just wondering why adding metadata in the project.
In general, I would say it's good to put this metadata into your build files.
Here, it results in the built jar file being named ./build/libs/checker-framework-languageserver-downloader-0.3.0-all.jar instead of ./build/libs/checker-framework-languageserver-downloader-all.jar.
So the build artifact also nicely contains the version number.
In the releases, we need to make sure that the first artifact is the built jar file, because that's an assumption of the cf-ls-downloader (we should document that better).
Is there any downside in including this metadata in the build script?
There was a problem hiding this comment.
Thanks for the clarification. I looked online, it says this is standard practice when developing libraries or modules intended to be published and shared with other projects. We only need to be mindful when evolving and remember to change it.
No description provided.