792845 - WARNING: GetScanCodeWithExtendedFlat() returns without extended flag

Status: RESOLVED FIXED

(Core :: Widget: Win32, defect)

Description

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

WARNING: GetScanCodeWithExtendedFlat() returns without extended flag: file c:/proj/firefox/widget/windows/../../../mozilla-central/widget/windows/KeyboardLayout.cpp, line 554

spam in debug builds when I just hold down the control key.  Fairly clean Windows 7 install with latest updates, debug m-c build.

Comment 1

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

Attachment: remove this warning

Pretty sure this can just go away; if the NativeKey isn't extended, we don't need to set the flag, right?  So we shouldn't print the warning... and is it useful to print it on < VISTA if mIsExtended?

Comment 2

[Jim Mathies [:jimm]]

Comment on attachment 663246 [details] [diff] [review]
remove this warning

Masayuki a better reviewer for this. He's authored all our overhauled keyboard handling.

Comment 3

[Masayuki Nakano [:masayuki] (he/him)(JST, +0900)]

Comment on attachment 663246 [details] [diff] [review]
remove this warning

Ah, it's my mistake. I'd like to add NS_ASSERTION(), but it's not so urgent. I'll do it later. So, just removing it is okay for now.

Comment 4

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/02e75979ee3b

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/02e75979ee3b

Should this have a test?

Comment 6

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

Don't think so, it just nukes an assertion that was incorrectly-written in the first place (there's no reason to assert the condition as written, so nothing to test).