442627 - xpcom/tests/unit/test_bug364285-1.js fails when locale charset is not UTF-8

Status: RESOLVED WONTFIX

(Core :: XPCOM, defect)

Description

[Mike Hommey [:glandium]]

../../_tests/xpcshell-simple/xpcom/unit/test_bug364285-1.js: FAIL
../../_tests/xpcshell-simple/xpcom/unit/test_bug364285-1.js.log:
>>>>>>>
*** test pending
*** exiting
*** CHECK FAILED:  CAA:89 ==  CAA:89
JS frame :: ../../tools/test-harness/xpcshell-simple/head.js :: do_throw ::
line 99
JS frame :: ../../tools/test-harness/xpcshell-simple/head.js :: do_check_eq ::
line 114
JS frame :: ../../_tests/xpcshell-simple/xpcom/unit/test_bug364285-1.js ::
create_file :: line 33
JS frame :: ../../_tests/xpcshell-simple/xpcom/unit/test_bug364285-1.js ::
test_create :: line 40
JS frame :: ../../_tests/xpcshell-simple/xpcom/unit/test_bug364285-1.js ::
run_test :: line 49
JS frame :: ../../tools/test-harness/xpcshell-simple/tail.js :: _execute_test
:: line 41
JS frame :: ../../tools/test-harness/xpcshell-simple/execute_test.js ::
<TOP_LEVEL> :: line 38
2147500036
*** FAIL ***

<<<<<<<

It would probably go a little further with ru_RU.KOI8-R, but would then stop on the japanese text that can't be expressed in KOI8-R. (Or maybe it would stop even earlier because of the french ç)

Anyways, the test only passes when the locale charset is UTF-8.

FWIW, I spotted this because by build chroots don't set a particular locale, therefore using POSIX as locale.

Comment 1

[Simon Montagu :smontagu]

So I assume bug 364285 itself is not fixed on non-UTF-8 Linux locales?

Comment 2

[Mike Hommey [:glandium]]

Actually the problem is different than that of bug 364285. Here, the problem is that the filenames are UTF16 that can't be transcribed in "native" charset, where "native" charset is the one from the system locale, in some cases (depending on the system locale). The -n files are correctly named. The problem is that the test assumes the filenames can be converted in native charset losslessly.

Comment 3

[Simon Montagu :smontagu]

Attachment: Disable the test on non-UTF-8 Linux

Comment 4

[Benjamin Smedberg]

Comment on attachment 327386 [details] [diff] [review]
Disable the test on non-UTF-8 Linux

I don't think we want this. We assume that any modern Linux is going to have their system locale in UTF8... we should check this assumption at startup and do something sane (abort?) if it's not, but we shouldn't cripple our unit tests like this.

Comment 5

[Christian :Biesinger (don't email me, ping me on IRC)]

What about other unices?

Comment 6

[Dimitrios Apostolou]

FWIW, I am trying to build firefox3 on Irix, and I know that Irix is failing this test because it doesn't support UTF-8 locales.

Comment 7

[BugBot [:suhaib / :marco/ :calixte]]

The bug assignee didn't login in Bugzilla in the last 7 months.
:nika, could you have a look please?
For more information, please visit auto_nag documentation.

Comment 8

[Nika Layzell [:nika] (ni? for response)]

I don't think things have changed here in the last 14 years. If we still have problems we should open a new bug.