-
-
Notifications
You must be signed in to change notification settings - Fork 842
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
chore: Update idna to 1.0, drop unicode-normalization #18598
Conversation
We should definitely add something to (probably) |
The size of the vanilla WASM module after wasm-bindgen and wasm-opt:
|
Let's not forget to do a |
0abfc32
to
ce4538a
Compare
On top of adding bans in |
b1f1498
to
4597514
Compare
See this for testing: package {
import flash.display.Loader;
import flash.display.Sprite;
import flash.net.URLRequest;
public class Test extends Sprite {
var nextY = 0;
function loadFromDomain(domain: String) {
var imageURL:String = domain + "/img/navi/top.gif";
var loader:Loader = new Loader();
var req:URLRequest = new URLRequest(imageURL);
trace("Loading from " + req.url);
loader.load(req);
loader.y = this.nextY;
this.nextY += 50;
addChild(loader);
}
public function Test() {
loadFromDomain("https://www.schoen.de");
loadFromDomain("https://www.schön.de");
loadFromDomain("https://www.xn--schn-7qa.de");
}
}
} SWF here: test.zip On Linux, I see all 3 instances of the same image, on Windows, the middle one is missing (with the FP debugger showing an IO exception about it). |
This will also be needed to resolve this advisory: |
e65a830
to
241a6ad
Compare
To be fair, -200kB was a way bigger deal when our wasm bundle was like 2MB :( |
Anyway, as long as evilpie's concern is resolved, I think this is safe-ish? |
I don't think that would be necessary even, we can easily switch between the backends now via the |
da43e34
to
26a8750
Compare
Thanks to both of you! 😊 |
@dependabot unignore url |
Now that all of
url
,cookie_store
, andpublicsuffix
are out with releases depending onidna
1.0.x
, we can switch to that.With this, the
idna_adapter
crate is introduced, which lets us globally control (in a hacky, but workable way), which IDNA backend we want, and allows dropping support for it entirely.See: https://crates.io/crates/idna_adapter
It should probably be locked to
1.0.x
with some mechanism. (Adding it directly toCargo.toml
, and/or configuringcargo-deny
, since we already have it...)Cc: @adrian17, since I remember you complaining a lot about the binary size bloat caused by these encoding tables... 😅