asyncapi v0.11.2 (new formula) #92115
Conversation
BrewTestBot
added
new formula
|
I haven't been able to test it locally because of error I tried to find some answers online but haven't found any answers 😞 |
|
I wonder if it is because in |
|
I just checked another example, apollo-cli formula and they also use oclify to build CLI, and they also have any ideas? should I open a separate issue? |
|
Co-authored-by: Carlo Cabrera <30379873+carlocab@users.noreply.github.com>
|
@chrmoritz, any chance you could look at the |
|
That's an issue with language node. The error comes from It can be worked around by disabling the def install
inreplace "package.json", "postpack", "postpack_disabled"
system "npm", "install", *Language::Node.std_npm_install_args(libexec)
bin.install_symlink Dir["#{libexec}/bin/*"]
rm libexec/"lib/node_modules/@asyncapi/cli/oclif.manifest.json"
end |
|
@carlocab @chrmoritz thanks a lot folks, I applied the workaround, locally installation works and tests pass 💪🏼 Thanks a bunch! tbh I still do not get why this script is fired after all, like, when I install locally with verbose flag, these is nothing about postpack script being triggered. It is triggered by |
Formula/asyncapi.rb
Outdated
| inreplace "package.json", "postpack", "postpack_disabled" | ||
| system "npm", "install", *Language::Node.std_npm_install_args(libexec) | ||
| bin.install_symlink Dir["#{libexec}/bin/*"] | ||
| rm libexec/"lib/node_modules/@asyncapi/cli/oclif.manifest.json" |
There was a problem hiding this comment.
Can we add some comments explaining why we're doing the inreplace and rm here?
|
The reason is, that homebrew doesn't do a normal |
|
It still puzzles me why it is needed 😄 Take apollo CLI and their formula. They use oclif, also have manifest file but do not need a workaround In their |
|
|
We use but locally when I test, |
|
still CI fails 🤔 but locally workaround proved to work |
|
Oh my mistake, I didn't remembered correctly. The issue is that The remaining CI issue is about shipping a universal binary fixed by adding def install
# Call rm -f instead of rimraf, because devDeps aren't present in Homebrew at postpack time
inreplace "package.json", "rimraf oclif.manifest.json", "rm -f oclif.manifest.json"
system "npm", "install", *Language::Node.std_npm_install_args(libexec)
bin.install_symlink Dir["#{libexec}/bin/*"]
# Replace universal binaries with their native slices
deuniversalize_machos
endThe other issue is that the |
|
Nice! now it is all clear 😆 thanks a bunch, and 🍻 for responsiveness 🙇🏼 Yes, CLI is just a repo, it is for tooling related to https://github.com/asyncapi/spec (almost 1.9k ⭐ ) and one of the tools that it integrates is https://github.com/asyncapi/generator (409 ⭐ ). We are just not fans of monorepo there. |
|
rimraf issues solved 🥳 what would be the next steps? |
carlocab
added
the
CI-no-fail-fast
It would be nice if you could do this. We regularly reject new formulae for the notability requirement, and I'd really rather not give the contributors whose PRs we reject the excuse of "but you accepted x, and they don't meet the notability requirement". (These arguments are rarely successful, but that doesn't make dealing with them any better.) |
|
@carlocab Hey, so this is what we got
Is it enough for a proof? I know you expect 30 watches but tbh I do not want to push the community to do it, even I limit my These are the numbers from our main repo And also from previous CLI for code generation I hope we won't suffer just because we do not like monorepos 😄 |
There was a problem hiding this comment.
Apologies for the delay here, @derberg. I missed your last message -- I have way too many GitHub notifications.
No need for additional "watches" -- the notability criteria is actually "any of" and not "all of", so the stars are enough.
Thanks for your contribution!
brew install --build-from-source <formula>, where<formula>is the name of the formula you're submitting?brew test <formula>, where<formula>is the name of the formula you're submitting?brew audit --strict <formula>(after doingbrew install --build-from-source <formula>)? If this is a new formula, does it passbrew audit --new <formula>?formula generated with https://github.com/zmwangx/homebrew-npm-noob, only tests added, inspired by https://github.com/Homebrew/homebrew-core/blob/master/Formula/angular-cli.rb