1330245 - application/x-www-form-urlencoded parser should not strip BOM

Status: NEW

(Core :: DOM: Core & HTML, defect, P3)

Description

[Anne (:annevk)]

See https://github.com/w3c/web-platform-tests/pull/4517 for tests. (The other problem seen in those tests is that we don't properly detect EOF, but I think that's a known issue with the encoding library.)

Comment 1

[Andrew Overholt [:overholt]]

Henri, do you know where the code that handles this lives?

Comment 2

[Henri Sivonen (:hsivonen)]

(In reply to Andrew Overholt [:overholt] from comment #1)
> Henri, do you know where the code that handles this lives?

http://searchfox.org/mozilla-central/source/intl/uconv/nsUTF8ToUnicode.cpp#324

That is, uconv always removes the BOM when decoding UTF-8. (encoding_rs has entry points with all the BOM handling policies that the spec requires.)

Comment 3

[Andrew Overholt [:overholt]]

Given encoding_rs' "someday soon" landing in Gecko I'm tempted to wontfix this. Henri/Anne, WDYT?

Comment 4

[Anne (:annevk)]

Even if we landed that, presumably we'd still need to make it do the correct thing for application/x-www-form-urlencoded. That is, an initial landing of that code presumably wants to preserve the same behavior as much as possible. So I think we still want this bug for tracking purposes.

Comment 5

[Andrew Overholt [:overholt]]

OK, cool.

Comment 6

[Henri Sivonen (:hsivonen)]

I agree with annevk. Let's leave this open as a reminder to make this work right once encoding_rs lands.