Your message dated Sun, 19 Jul 2009 20:56:15 +0200
with message-id <[email protected]>
and subject line Re: Bug#516526: Acknowledgement (po4a: <firstname>-tag not 
recognized in <author>)
has caused the Debian Bug report #516526,
regarding funny "# type: ...." line with wrong tag
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)


-- 
516526: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=516526
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: po4a
Version: 0.35-1
Severity: normal

I am testing experimental 0.35-1 now :-)

Let me present what has happened and my thought.  Here we go....

I have:

|    <bookinfo>
|      <title>Debian Reference</title>
|      <authorgroup>
|        <author>
|          <firstname>Osamu</firstname>
|          <surname>Aoki</surname>
|        </author>
|
|      </authorgroup>
|
|      <abstract>
|<xi:include href="abstract.en.xml"/>
|      </abstract>
|
|      <copyright>
|        <year>2007-2009</year>
|        <holder>Osamu Aoki</holder>
|      </copyright>
|      <legalnotice>
|
|<xi:include href="copy.en.xml"/>
|
|      </legalnotice>
|    </bookinfo>


Then the PO file with lenny version 0.34-2 was:

|# type: Content of: <book><bookinfo><title>
|#: ../en/debian-reference.en.xml:20
|msgid "Debian Reference"
|msgstr "Debianレファレンス"
|
|# type: Content of: <book><bookinfo><copyright><holder>
|#: ../en/debian-reference.en.xml:35
|msgid "Osamu Aoki"
|msgstr "青木 修 (Osamu Aoki)"

I see no translation for tags inside <author> .
As for copyright, I see only <holder> as po.
Since current local build script seems not to use <authorgroup> section
for html generation, this was fine.

I have insytalled 0.35-1 and run po4a-updatepo.  It created:

|# type: Content of: <book><bookinfo><title>
|#: ../en/debian-reference.en.xml:20
|msgid "Debian Reference"
|msgstr "Debianレファレンス"
|
|# type: Content of: <book><bookinfo><copyright><holder>
|#: ../en/debian-reference.en.xml:23
|#, fuzzy
|#| msgid "Osamu Aoki"
|msgid "Osamu"
|msgstr "青木 修 (Osamu Aoki)"
|
|# type: Content of: <book><bookinfo><authorgroup><author><surname>
|#: ../en/debian-reference.en.xml:24
|msgid "Aoki"
|msgstr ""
|
|# type: Content of: <book><bookinfo>
|#: ../en/debian-reference.en.xml:33
|msgid ""
|"<copyright> <year>2007-2009</year> <holder>Osamu Aoki</holder> </copyright>"
|msgstr ""

It seems to me that 
 * <authorgroup> section is correctly split into strings down to
 * <surname> <firstname> <holder> tag is correctly treated as inline
 * within <authorgroup> commented line with --previous option has funny 
   "# type: ...." line with wrong tag with respect to the latest
   original file while line number is correct.

I think po4a moving string due to new split scheme is right action but
comments needs to reflect it and to be understandable one.  If
--previous holds old information under "#|" line, I think po4a-updatepo
should have created something along:

|# type: Content of: <book><bookinfo><title>
|#: ../en/debian-reference.en.xml:20
|msgid "Debian Reference"
|msgstr "Debianレファレンス"
|
|# type: Content of: <book><bookinfo><authorgroup><author><firstname>
|#: ../en/debian-reference.en.xml:23
|#, fuzzy
|#| type: Content of: <book><bookinfo><copyright><holder>
|#|: ../en/debian-reference.en.xml:35
|#| msgid "Osamu Aoki"
|msgid "Osamu"
|msgstr "青木 修 (Osamu Aoki)"
|
|# type: Content of: <book><bookinfo><authorgroup><author><surname>
|#: ../en/debian-reference.en.xml:24
|msgid "Aoki"
|msgstr ""
|
|# type: Content of: <book><bookinfo>
|#: ../en/debian-reference.en.xml:33
|msgid ""
|"<copyright> <year>2007-2009</year> <holder>Osamu Aoki</holder> </copyright>"
|msgstr ""

I hope you get my point.  This should help user locate which string po4a
is converting.

Please note all quoted text are lead by "|".

As for practical build impact, there is no damage.  I just need to
update po for "# type: Content of: <book><bookinfo>".  But this is
somehow basic issue with po4a-updatepo handling of string move, I am
reporting as bug.

-- System Information:
Debian Release: 5.0
  APT prefers stable
  APT policy: (500, 'stable'), (20, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.26-1-amd64 (SMP w/2 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages po4a depends on:
ii  gettext                   0.17-6         GNU Internationalization utilities
ii  libsgmls-perl             1.03ii-32      Perl modules for processing SGML p
ii  perl                      5.10.0-19      Larry Wall's Practical Extraction 
ii  perl-modules              5.10.0-19      Core Perl modules
ii  sp                        1.3.4-1.2.1-47 James Clark's SGML parsing tools

Versions of packages po4a recommends:
ii  liblocale-gettext-perl        1.05-4     Using libc functions for internati
ii  libterm-readkey-perl          2.30-4     A perl module for simple terminal 
ii  libtext-wrapi18n-perl         0.06-6     internationalized substitute of Te

po4a suggests no packages.

-- no debconf information



--- End Message ---
--- Begin Message ---
Version: 0.36-1

On Sun, Feb 22, 2009 at 04:59:49PM +0100, Nicolas François wrote:
> 
> # type: Content of: <book><bookinfo><authorgroup><author><firstname>
> #: ../en/debian-reference.en.xml:23
> #, fuzzy
> #| type: Content of: <book><bookinfo><copyright><holder>
> #|: ../en/debian-reference.en.xml:35
> #| msgid "Osamu Aoki"
> msgid "Osamu"
> msgstr "青木 修 (Osamu Aoki)"
> 
> I will check if using a msgctxt would fix this without drawbacks (I fear
> It could break string reuse)

It was not possible in current po4a to introduce context.
Introducing context would also increase a lot the complexity for users.
If context were to be supported, I would propose to generate multiple PO
files, and merge them using contexts.

> Using an EXTRACTED-COMMENTS instead of TRANSLATOR-COMMENTS could be better
> (this will fix the last point, about outdated tag)

This was done in version 0.36.

Best Regards,
-- 
Nekral


--- End Message ---

Reply via email to