fall back to classifiers if [pypi] license text is really long #8690
Conversation
chris48s
added
bug
| @@ -50,7 +50,7 @@ function getLicenses(packageData) { | |||
| const { | |||
| info: { license }, | |||
| } = packageData | |||
| if (license) { | |||
| if (license && license.length < 40) { | |||
There was a problem hiding this comment.
As noted in #8689 (comment) we're going to have to make some kind of assumption here. The longest SPDX identifier in https://raw.githubusercontent.com/jslicense/spdx-license-ids/main/index.json is 36 characters so this seems like a reasonable-ish threshold.
My hunch is that the license meta-data field is more commonly used than the classifiers, so I think switching to prefer the classifiers in all cases is probably going to break more badges out in the wild than doing this.
There isn't really a perfect solution here.
There was a problem hiding this comment.
Agreed with the approach. I do think it might be worth adding an inline comment explaining this though so that 40 doesn't appear like a magic number to readers of the code.
I recognize that one could use git blame info to backtrack to this PR and explanation, but I'm a fan of having things be as self-explanatory as possible within the code
Closes #8689