Remove charset=utf-8 default

[BigBlueHat]

Fixes #220

This also fixes image loading issues when using fetch() + createImageBitmap() in Firefox:
http-party/http-server#296 (comment)

[codecov-io]

Codecov Report

Merging #223 into master will decrease coverage by 0.19%.
The diff coverage is 100%.

@@            Coverage Diff            @@
##           master     #223     +/-   ##
=========================================
- Coverage   75.86%   75.67%   -0.2%     
=========================================
  Files           9        9             
  Lines         518      518             
  Branches      117      117             
=========================================
- Hits          393      392      -1     
  Misses         45       45             
- Partials       80       81      +1

Impacted Files	Coverage Δ
lib/ecstatic.js	73.3% <100%> (-0.46%)	⬇️

---

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 609f1c9...e8ac599. Read the comment docs.

[jfhbrook]

Thanks for trying to make ecstatic better! I agree that it's alarming that images break because of this. A few minor concerns:

1. This would be a major version bump as-is, which is fine in and of itself but I dunno that I'd consider this a patchfix of the main line. If this was feature flagged somehow (or made configurable s.t. you can set your own charsets) I could justify it being a minor, and then flip the defaults for a major later.
2. You mentioned being interested in upgrading the mime package--would that fix this "better"?

[BigBlueHat]

made configurable s.t. you can set your own charsets

I wouldn't want to encourage a blanket charset setting option as that puts us back right where we are (passing charset on all the things).

Maybe an --auto-charset option would do the trick? It could become the default in a future major version and the defaulting to utf-8 would be gone.

You mentioned being interested in upgrading the mime package--would that fix this "better"?

See #220 (comment) :)

[athei]

I also stumbled on this because it prevents Firefox from using wasm streaming instantiation. Why is this a major version bump? We are not breaking the API. It is a bug fix isn't it?

[jfhbrook]

One person's bugfix is another's expected behavior 🤷‍♂️

[athei]

Okay :D

So the progress on this is stalling because you need to evaluate if you skip this patch alltogether and just apply #218 instead?

Edit: Okay #218 alone does not fix this. The charset would still be applied.

[jfhbrook]

I'm not merging #218, at least not as-is, it's open for discussion and so that someone can take the opportunity to fix the issues outlined in that PR.

FWIW I /believe/ this ticket is still an issue even with something along the lines of #218 merged.

[jfhbrook]

Dredged up #44 which was the original issue--I think there are good reasons to specify the charset in at least some cases, hence why this exists in the first place.

[athei]

Yes and this patch delegates that charset decision to the mime library like it should. The charset should still be added to types that need it after that patch.

[jfhbrook]

A test that confirms that would be nice!

[jfhbrook]

Anyway, just to reiterate what I'm looking for around this and #218 :

• Some answer to cases where setting charset would be appropriate--either delegating to an autodetect library, or an optional way to override, with tests
• Some answer to passing in custom types--either a separate library for parsing types files with new mime, or the ability to pass in a custom mimetypes resolver in javascript (punting to the user)

[jfhbrook]

I get that you're frustrated, btw, with this long-standing issue that prevents a lot of cool things from working right. The problem is that I don't want to merge casual code, but don't have the time to implement the right thing myself. All I can say is, be the change you wish to see in the world.

[jfhbrook]

I rolled these changes into #240, which addresses the concerns I had by exposing a way to set a custom content-type lookup function. Thanks!