Port MIME type detection over from HttpServer.jl

[rdeits]

Fixes JuliaGizmos/WebIO.jl#181

This was broken in the transition from HttpServer.jl to HTTP.jl. There's a TODO in the code to switch to HTTP.sniff() but that doesn't quite work. The problem is that HTTP.sniff() only operates on the file contents and not the extension, so it always returns text/plain for things like .css files, while old implementation correctly returned text/css.

To avoid a new dependency, I just copied over the whole mimetypes.jl file from HttpServer into Mux. It's not an elegant solution, but given that the current behavior is broken, I think it's necessary. In the future, we could consider moving the mimetype list elsewhere.

[rdeits]

Here's the old implementation that I'm essentially restoring:

Mux.jl/src/examples/files.jl

Line 20 in 6aec344

fileheaders(f) = d("Content-Type" => get(mimetypes, extension(f), "application/octet-stream"))

[codecov-io]

Codecov Report

Merging #68 into master will increase coverage by 0.34%.
The diff coverage is 100%.

@@            Coverage Diff            @@
##           master     #68      +/-   ##
=========================================
+ Coverage   90.56%   90.9%   +0.34%     
=========================================
  Files           5       6       +1     
  Lines          53      55       +2     
=========================================
+ Hits           48      50       +2     
  Misses          5       5

Impacted Files	Coverage Δ
src/Mux.jl	100% <ø> (ø)	⬆️
src/examples/files.jl	100% <100%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 962b54f...4bbd9e5. Read the comment docs.

[piever]

Maybe this could be moved to HTTP eventually, but for now this PR is a good practical solution.

[hustf]

Thank you for the great work you are putting in. This may belong in a new HttpCommon in the end, but for now I will refer Mux.jl in private repositories.

Would you consider including some additional extensions? I always add these lines to HttpServer (sorry for layout)

Add some mime types, which some browsers support.

HttpServer.mimetypes["jl"]= "text/plain"
HttpServer.mimetypes["md"] = "text/markdown; charset=UTF-8"
HttpServer.mimetypes["vec"] = "text/plain"

[rdeits]

I don't actually know much about the details of MIME types. The .jl one seems fine, but why is the charset specified only for markdown? And what is a .vec file? There seem to be a bunch of formats using that extension, some of which are binary.

Maybe we can decide whether to add those as part of a separate PR? I'd like to get the original issue fixed as soon as possible.

[piever]

Maybe we can decide whether to add those as part of a separate PR? I'd like to get the original issue fixed as soon as possible.

Yes! Not loading any css library right now in WebIO is pretty bad, I think we can discuss in a separate PR which other mimetype detection we want to add and whether we want to allow a mechanism to customize this detection on the user side.

[hustf]

Not adding these is of course just fine. The more one adds, the less general the code becomes.