[GitHub] Top language, languages count and code size badges #1160
Conversation
server.js
Outdated
| } | ||
| badgeData.text[0] = 'code size'; | ||
| badgeData.text[1] = metric(sumBytes) + 'B'; | ||
| badgeData.colorscheme = 'blue'; |
There was a problem hiding this comment.
I'm really glad to have a repo size badge! Though, it would be more reliable to get this information from the Repositories endpoint. Would you be up for implementing it that way instead, under a different URL path?
There was a problem hiding this comment.
How would you get the code size directly using that endpoint? The size parameter seemed to be a good candidate, but must actually relate to something else, as it indicates only 4945 for this repository for instance.
There was a problem hiding this comment.
I think that is right. It's in kilobytes. Here's a StackOverflow answer that talks about it.
There was a problem hiding this comment.
The more I think about it, the repo size is more about how many revisions and binary blobs are checked in. The size of the identifiable code could be useful, distinct from the repo size.
Do they provide lines of code? That would also be really cool. 😃
There was a problem hiding this comment.
Yes, I think it's the size on disk rather than the code size. I would be inclined to keep the code size badge as it is, but I can also provide a repo size badge through a new pull request, if you want!
They do not provide lines of code directly, but this StackOverflow answer suggests a nice approach to get it. Yet another pull request coming up? 😄
There was a problem hiding this comment.
Awesome! Yes, a separate badge + PR for repo size makes sense to me.
Wow, if that SO answer works, that would be cool!
There was a problem hiding this comment.
Okay, I'll take care of that!
There was a problem hiding this comment.
Actually the solutions proposed to calculate the lines of code using the API are dodgy according to a comment on the answer I had previously linked. I don't think a line count badge can be implemented reliably and efficiently with the current API possibilities.
service-tests/github.js
Outdated
| .get('/languages/bytes/badges/shields.json') | ||
| .expectJSONTypes(Joi.object().keys({ | ||
| name: Joi.equal('code size'), | ||
| value: Joi.string().regex(/^[0-9]*(k|M|G|T|P|E|Z|Y)B$/), |
There was a problem hiding this comment.
You can use the isFileSize helper from service-tests/helpers/validators.js.
There was a problem hiding this comment.
My bad, I had forked the repository before the helpers were added, this is now fixed.
service-tests/github.js
Outdated
| .get('/languages/count/badges/shields.json') | ||
| .expectJSONTypes(Joi.object().keys({ | ||
| name: Joi.equal('languages'), | ||
| value: Joi.string().regex(/^[0-9]*$/), |
There was a problem hiding this comment.
Joi.number().integer().positive() should work here.
service-tests/github.js
Outdated
| .expectJSONTypes(Joi.object().keys({ | ||
| name: Joi.equal('language'), | ||
| value: Joi.equal('none'), | ||
| })); |
There was a problem hiding this comment.
You can tighten this up a bit since these are literals:
.expectJSON({ name: 'language', value: 'none' })
server.js
Outdated
| sumBytes += parseInt(parsedData[language]); | ||
| } | ||
| badgeData.text[0] = 'code size'; | ||
| badgeData.text[1] = metric(sumBytes) + 'B'; |
There was a problem hiding this comment.
Could you use prettyBytes instead?
server.js
Outdated
| badgeData.text[1] = Object.keys(parsedData).length; | ||
| badgeData.colorscheme = 'blue'; | ||
| break; | ||
| case 'bytes': |
There was a problem hiding this comment.
How about code-size or size to match the label?
|
Awesome 👍 I'd like to see this available! 😃 |
|
Github service tests passing locally with the patch from #1118. |
This pull request is related to issue #1139 .
In addition to the main programming language badge discussed there, these commits also add two additional badges making use of the same languages API endpoint, but parsing the received data slightly differently.
The resulting badges should look similar to the following:
Hope I have followed the guidelines correctly and looking forward to any review comments and feedback! 👍
Cheers,
Pyves