#4701 ActionPack, 1.9.2-head 3 failing tests regarding encodings - Ruby on Rails

Type	To find
responsible:me	tickets assigned to you
tagged:"@high"	tickets tagged @high
milestone:next	tickets in the upcoming milestone
state:invalid	tickets with the state invalid
created:"last week"	tickets created last week
sort:number, importance, updated	tickets sorted by #, importance or updated
Combine keywords for powerful searching.
Use advanced searching »

This project is archived and is in readonly mode.

#4701 ✓committed

ActionPack, 1.9.2-head 3 failing tests regarding encodings

Reported by Rohit Arondekar | May 26th, 2010 @ 07:19 AM | in 3.0.2

Rails Master ActionPack, ruby 1.9.2dev (2010-05-08 trunk 27665) [x86_64-linux] has 3 failing tests.

The stacktrace is as follows:

1) Failure: (TextHelperTest) [test/template/text_helper_test.rb:221]: <"...\x83ciency could not be..."> expected but was
<"...ﬃciency could not be...">.

2) Failure: test_truncate_multibyte(TextHelperTest) [test/template/text_helper_test.rb:87]:
<"\xEC\x95\x88\xEB\x85\x95\xED..."> expected but was
<"안녕하세요">.

3) Failure: test_scoping_by_partial_of_an_array(TranslationHelperTest) [test/template/translation_helper_test.rb:41]:
<"foobar"> expected but was
<"["foo", "bar"]">.

2704 tests, 11359 assertions, 3 failures, 0 errors, 0 skips

Note: It's best to run the tests yourself, because I don't think some of the chars got copied from my terminal properly.

Regarding the first 2 failures, it happens when force_encoding is not used.

The failing assert is as follows:

assert_equal("...\203ciency could not be...", excerpt("That's why e\357\254\203ciency could not be helped", 'could', 8))

Note that:

"...\203ciency could not be...".encoding => ASCII-8BIT excerpt("That's why e\357\254\203ciency could not be helped", 'could', 8).encoding => UTF-8

Similarly the 2nd test. I'm looking into the 3rd test now.

Comments and changes to this ticket

Rohit Arondekar May 26th, 2010 @ 07:31 AM
The 3rd failure seems like a mistake in the return value of t. I've attached a patch to fix this (if it's indeed the problem).
- 4701_fix_1_test.diff 1.1 KB
Rohit Arondekar May 26th, 2010 @ 04:32 PM
Bad formatting in the original message.

This is regarding the first 2 tests, asserts without force_encoding:

The string literals have encoding of ASCII-8BIT
"...\203ciency could not be...".encoding => ASCII-8BIT

Whereas, excerpt returns a UTF-8 string, and hence the tests fail.
excerpt("That's why e\357\254\203ciency could not be helped", 'could', 8).encoding => UTF-8
Santiago Pastorino May 26th, 2010 @ 04:56 PM
- Milestone cleared.
- State changed from “new” to “open”
That's not the proper fix Rohit.
This test is fixed here.
- translation_template_test_19.patch 748 Bytes
Rohit Arondekar May 27th, 2010 @ 03:12 AM
Santiago, thanks for the proper fix. I've never used translation so I didn't know what t() is supposed to return. Any idea how to fix the first 2 failures?
Rohit Arondekar May 27th, 2010 @ 03:32 AM
+1 on the patch, the translation test now passes on 1.8.7 and 1.9.2-head.
Repository May 29th, 2010 @ 08:21 PM
(from [e5a56eb297f41dae12e03cf03bad4a25068bba09]) fix translation test in actionpack [#4701 state:commited]

Signed-off-by: wycats wycats@gmail.com
http://github.com/rails/rails/commit/e5a56eb297f41dae12e03cf03bad4a...
Rohit Arondekar May 30th, 2010 @ 06:20 AM
I think it would be a good idea to mark this ticket resolved as the tests now pass after commit => http://github.com/rails/rails/commit/d7f6f2b596d00a7deb4f83592aaa78...

But from the comment in the commit, ".mb_chars always returns UTF-8, even in 1.9. This is not great, but it's how it works. Let's work this out." (http://bit.ly/aUsn1o), a new dedicated ticket can be opened?
Santiago Pastorino May 30th, 2010 @ 06:27 AM
- State changed from “open” to “committed”
[#OMG state:committed] with one t. So i'm closing by hand
Jeremy Kemper October 15th, 2010 @ 11:01 PM
- Milestone set to “3.0.2”
- Importance changed from “” to “Low”
[bulk edit]