Reviewer Badges :)

[sneakers-the-rat]

Fix: #1385

here she is:

review_badge.mp4

and on mobile:

review_badge_mobile.mp4

OK this was more complicated than i thought because it turns out you need to like calculate the widths of all the characters for a given font to make an SVG badge, but i did it because i have wanted to have something like this in my own websites without having to use shields.io or w/e. Elsewhere the SVGs are just hardcoded, but I thought it would be nice to generate them for any key and color because we might want to use them for other things :)

I basically transcribed and simplified shields.io's code, it's mostly self explanatory except don't ask me what's going on with those scaling parameters. I grabbed an array of codepoint widths and made it a constant, use that to calculate the width, and the rest is in the controller.

I tried to reuse as much as i could from the rest of the site, but it looks like we're not using sass variables so i just hunted out values that seemed like they were the right ones to use.

The badges will be at joss.theoj.org/badges/reviewed_by/:username. I imagine re-rendering them from scratch every time is pointlessly expensive, so plz lmk how caching works here because they should only rarely change, but it's hard to detect the review count being stale because we don't have a Reviewer model.

I am holding off on tests for now until i get signoff on the basic approach, and then i'll write em!

[arfon]

Hey! Thanks for getting going on this @sneakers-the-rat ⚡. This is definitely the sort of thing I'm thinking about. One thought, do you think as part of this we might want to consider consolidating the other SVG rendering (e.g.,

joss/app/controllers/papers_controller.rb

Lines 307 to 321 in c447b2a

	def status
	if params[:doi] && valid_doi?
	@paper = Paper.find_by_doi(params[:doi])
	else
	@paper = Paper.find_by_sha(params[:id])
	end
	# TODO: Remove these SVGs from the controller
	if @paper
	svg = @paper.status_badge
	else
	prefix = setting(:abbreviation)
	svg = "<svg xmlns='http://www.w3.org/2000/svg' width='102' height='20'><linearGradient id='b' x2='0' y2='100%'><stop offset='0' stop-color='#bbb' stop-opacity='.1'/><stop offset='1' stop-opacity='.1'/></linearGradient><mask id='a'><rect width='102' height='20' rx='3' fill='#fff'/></mask><g mask='url(#a)'><path fill='#555' d='M0 0h40v20H0z'/><path fill='#9f9f9f' d='M40 0h62v20H40z'/><path fill='url(#b)' d='M0 0h102v20H0z'/></g><g fill='#fff' text-anchor='middle' font-family='DejaVu Sans,Verdana,Geneva,sans-serif' font-size='11'><text x='20' y='15' fill='#010101' fill-opacity='.3'>#{prefix}</text><text x='20' y='14'>#{prefix}</text><text x='70' y='15' fill='#010101' fill-opacity='.3'>Unknown</text><text x='70' y='14'>Unknown</text></g></svg>"
	end

) into a common controller?

[sneakers-the-rat]

I think it would make sense to do that, but didnt want to change it all until I got a nod to go ahead :)

[sneakers-the-rat]

reviewing my open PRs and thought i'd ping on this?

The only outstanding question i think is caching, and from what I can tell from the docs, rails should automatically cache rendering of the template for a given set of parameters. since the parameters given to the template are all generic across reviewer identity (i.e. they just have the text of the badge and the numbers) that should kick in. Ideally we would also cache the badge per reviewer so we don't need to do a db query at all, but doing that would require us to add reviewers to the schema so we can do cache invalidation when a reviewer does another review, and that should probably be a different PR.

Another approach would be to make the badge endpoint totally generic, someting like example.com/api/badges?key=JOSS%20Reviews&value=20&value_color=blue so that every badge could be cached on its own, and then the badges/reviewed_by/:username endpoint could just forward to the relevant badge page, so rather than rendering a badge per user we are rendering a badge per parameterization.

I also think swapping the existing badges to this controller should probably be another PR.

Anything blocking this?