Bug #6040: Transcoding test failure: Big5 to UTF8 not defined (MinGW)

• Status: Closed
• Assignee: naruse (Yui NARUSE)
• Target version: 1.9.3
• ruby -v: ruby 1.9.3p125 (2012-02-16) [i386-mingw32]

[luislavena (Luis Lavena)]

Hello,

While testing 1.9.3-p125 (released), discovered this bug which was not present in previous builds:

100) Error:
test_Big5(TestTranscode):
Encoding::UndefinedConversionError: "\xA1@" from Big5 to UTF-8
    C:/Users/Luis/Projects/oss/oneclick/rubyinstaller/sandbox/ruby_1_9/test/ruby/test_transcode.rb:54:in `encode'
    C:/Users/Luis/Projects/oss/oneclick/rubyinstaller/sandbox/ruby_1_9/test/ruby/test_transcode.rb:54:in `check_both_ways'
    C:/Users/Luis/Projects/oss/oneclick/rubyinstaller/sandbox/ruby_1_9/test/ruby/test_transcode.rb:1788:in `test_Big5'

I can't reproduce this when run in isolation (make test-all TESTS="ruby/test_transcode.rb")

Perhaps this is a fluke?

Thank you

[jeremyevans0 (Jeremy Evans)]

I can confirm that this error occurs on both OpenBSD amd64 and i386. This happens both when the whole test suite is run and when just test_transcode is run in isolation.

[nobu (Nobuyoshi Nakada)]

• Category changed from core to build
• Assignee set to naruse (Yui NARUSE)

I've seen this failure when I made enc/trans/*.c with ruby 1.8, but nothing wrong when made with ruby 1.9 or later.
Therefore, seems a bug in the conversion script.

Or a packaging issue.

[luislavena (Luis Lavena)]

Thank you Nobu and Jeremy, I've used 1.9.3-p0 as base ruby so shouldn't be
that.

I'll try build 1.9.3 branch again tomorrow just to confirm.

Sorry for top posting. Sent from mobile.

[naruse (Yui NARUSE)]

• Status changed from Open to Closed

This should be fixed r34650 in trunk / 1.9.3-p127.
Could you check and reopen this if not fixed?

[luislavena (Luis Lavena)]

On Thu, Feb 16, 2012 at 9:55 PM, Yui NARUSE [e-mail] wrote:

Issue #6040 has been updated by Yui NARUSE.

Status changed from Open to Closed

This should be fixed r34650 in trunk / 1.9.3-p127.
Could you check and reopen this if not fixed?

Sorry, redmine seems to be down so responding to ruby-core directly.

ruby_1_9_3, revision 34660 worked perfectly. Thank you.

This justifies another patchlevel release? Big5 encoding seems
important for eastern languages (chinese) and RubyInstaller seems to
have some userbase there (along with Japan).

Thank you.

Luis Lavena
AREA 17

Perfection in design is achieved not when there is nothing more to add,
but rather when there is nothing more to take away.
Antoine de Saint-Exupéry

[naruse (Yui NARUSE)]

Luis Lavena wrote:

On Thu, Feb 16, 2012 at 9:55 PM, Yui NARUSE [e-mail] wrote:

Issue #6040 has been updated by Yui NARUSE.

Status changed from Open to Closed

This should be fixed r34650 in trunk / 1.9.3-p127.
Could you check and reopen this if not fixed?

Sorry, redmine seems to be down so responding to ruby-core directly.

ruby_1_9_3, revision 34660 worked perfectly. Thank you.

This justifies another patchlevel release? Big5 encoding seems
important for eastern languages (chinese) and RubyInstaller seems to
have some userbase there (along with Japan).

This affect only on packaging.
So I repacked it.

[luislavena (Luis Lavena)]

On Fri, Feb 17, 2012 at 7:47 PM, Yui NARUSE [e-mail] wrote:

Issue #6040 has been updated by Yui NARUSE.

Luis Lavena wrote:

 This justifies another patchlevel release? Big5 encoding seems
 important for eastern languages (chinese) and RubyInstaller seems to
 have some userbase there (along with Japan).

This affect only on packaging.
So I repacked it.

That means I need to re-download and recompile source?

RubyInstaller 1.9.3-p125 was already released on the assumption the
package was final.

I can't re-release p125 to users, RubyForge do not allow same versions
to be uploaded and there is more than 3K downloads already for the
installers.

Please advice.

Luis Lavena
AREA 17

Perfection in design is achieved not when there is nothing more to add,
but rather when there is nothing more to take away.
Antoine de Saint-Exupéry

[naruse (Yui NARUSE)]

(2012/02/18 7:54), Luis Lavena wrote:

On Fri, Feb 17, 2012 at 7:47 PM, Yui NARUSE [e-mail] wrote:

Issue #6040 has been updated by Yui NARUSE.

Luis Lavena wrote:

This justifies another patchlevel release? Big5 encoding seems
important for eastern languages (chinese) and RubyInstaller seems to
have some userbase there (along with Japan).

This affect only on packaging.
So I repacked it.

That means I need to re-download and recompile source?

RubyInstaller 1.9.3-p125 was already released on the assumption the
package was final.

I can't re-release p125 to users, RubyForge do not allow same versions
to be uploaded and there is more than 3K downloads already for the
installers.

Please advice.

We have some additional patches and current ruby_1_9_3 is p133.
So releasing another patch release is an option.

--
NARUSE, Yui [e-mail]

[luislavena (Luis Lavena)]

If you could do that will be really appreciated.

Thank you.

Sorry for top posting. Sent from mobile.
On Feb 17, 2012 8:08 PM, "NARUSE, Yui" [e-mail] wrote:

(2012/02/18 7:54), Luis Lavena wrote:

On Fri, Feb 17, 2012 at 7:47 PM, Yui NARUSE [e-mail] wrote:

Issue #6040 has been updated by Yui NARUSE.

Luis Lavena wrote:

This justifies another patchlevel release? Big5 encoding seems
important for eastern languages (chinese) and RubyInstaller seems to
have some userbase there (along with Japan).

This affect only on packaging.
So I repacked it.

That means I need to re-download and recompile source?

RubyInstaller 1.9.3-p125 was already released on the assumption the
package was final.

I can't re-release p125 to users, RubyForge do not allow same versions
to be uploaded and there is more than 3K downloads already for the
installers.

Please advice.

We have some additional patches and current ruby_1_9_3 is p133.
So releasing another patch release is an option.

--
NARUSE, Yui [e-mail]