mki18n.py?

Hello,

I tried to use the Pierre Rouleau's mki18n.py script at
<http://wiki.wxpython.org/index.cgi/Internationalization> and I ran
into a couple of issues: (1) it supports only the two letter ISO
language code (for example, "pt"), and not a language code followed by
the country code ("pt_BR") and (2) things become weird if the app's
directory contains spaces.

I fixed both issues here (Python 2.3.5, wxPython 2.6.3.0 ansi, msw XP
(pt_BR)), and replaced the 'iso639_languageDict' by some code that
finds out what are the currently wx supported canonical forms of
current locale names. It seems to be working.

May I update the script on the wiki page? Or should it be tested on
other platforms before? Volunteers, anyone? 8^)

-- tacao

No bits were harmed during the making of this e-mail.

Hello,

I tried to use the Pierre Rouleau's mki18n.py script at
<http://wiki.wxpython.org/index.cgi/Internationalization&gt; and
I ran into a couple of issues: (1) it supports only the two
letter ISO language code (for example, "pt"), and not a
language code followed by the country code ("pt_BR") and (2)
things become weird if the app's directory contains spaces.

I fixed both issues here (Python 2.3.5, wxPython 2.6.3.0
ansi, msw XP (pt_BR)), and replaced the 'iso639_languageDict'
by some code that finds out what are the currently wx
supported canonical forms of current locale names. It seems
to be working.

May I update the script on the wiki page? Or should it be
tested on other platforms before? Volunteers, anyone? 8^)

I've gave up using this script since I couldn't figure out how to get it to
support Russian or Chinese. Perhaps someone more clever than I can figure
this out too.

(I know, it doesn't address your question, but if people are going to start
looking into the script, I thought I'd at least raise the issue.)

David

E. A. Tacao wrote:

Hello,

I tried to use the Pierre Rouleau's mki18n.py script at
<http://wiki.wxpython.org/index.cgi/Internationalization&gt; and I ran
into a couple of issues: (1) it supports only the two letter ISO
language code (for example, "pt"), and not a language code followed by
the country code ("pt_BR") and (2) things become weird if the app's
directory contains spaces.

I fixed both issues here (Python 2.3.5, wxPython 2.6.3.0 ansi, msw XP
(pt_BR)), and replaced the 'iso639_languageDict' by some code that
finds out what are the currently wx supported canonical forms of
current locale names. It seems to be working.

May I update the script on the wiki page?

I think it would be okay.

···

--
Robin Dunn
Software Craftsman
http://wxPython.org Java give you jitters? Relax with wxPython!

E. A. Tacao wrote:

Hello,

Hello. I just noticed your mail.

I tried to use the Pierre Rouleau's mki18n.py script at
<http://wiki.wxpython.org/index.cgi/Internationalization&gt; and I ran
into a couple of issues: (1) it supports only the two letter ISO
language code (for example, "pt"), and not a language code followed by
the country code ("pt_BR") and (2) things become weird if the app's
directory contains spaces.

Yep, when I wrote that, I wasn't even trying to support those.

I fixed both issues here (Python 2.3.5, wxPython 2.6.3.0 ansi, msw XP
(pt_BR)), and replaced the 'iso639_languageDict' by some code that
finds out what are the currently wx supported canonical forms of
current locale names. It seems to be working.

Great! Thanks for doing it.

May I update the script on the wiki page? Or should it be tested on
other platforms before? Volunteers, anyone? 8^)

I don't see any problem with updating. I am running Python 2.4.3, wxPython 2.6.3.2 unicode msw XP and I'll try your update. I just recently working over some wxPython stuff again so it'll fit right in.

···

--

Pierre Rouleau

E. A. Tacao wrote:

Hello,

Hi. I just noticed your mail.

I tried to use the Pierre Rouleau's mki18n.py script at
<http://wiki.wxpython.org/index.cgi/Internationalization&gt; and I ran
into a couple of issues: (1) it supports only the two letter ISO
language code (for example, "pt"), and not a language code followed by
the country code ("pt_BR") and (2) things become weird if the app's
directory contains spaces.

Yep, when I wrote that, I wasn't even trying to support those.

I fixed both issues here (Python 2.3.5, wxPython 2.6.3.0 ansi, msw XP
(pt_BR)), and replaced the 'iso639_languageDict' by some code that
finds out what are the currently wx supported canonical forms of
current locale names. It seems to be working.

Great! Thanks for doing it.

May I update the script on the wiki page? Or should it be tested on
other platforms before?

Nooo problem! Go ahead!

  Volunteers, anyone? 8^)

Sure, I can test on my platform. I am running Python 2.4.3, wxPython 2.6.3.2 unicode msw XP and I'll try your update. I just recently working over some wxPython stuff again so it'll fit right in.

···

--

Pierre Rouleau

Thursday, April 27, 2006, 6:18:37 PM, Pierre Rouleau wrote:

E. A. Tacao wrote:

Hello,

Hi. I just noticed your mail.

I tried to use the Pierre Rouleau's mki18n.py script at
<http://wiki.wxpython.org/index.cgi/Internationalization&gt; and I ran
into a couple of issues: (1) it supports only the two letter ISO
language code (for example, "pt"), and not a language code followed by
the country code ("pt_BR") and (2) things become weird if the app's
directory contains spaces.

Yep, when I wrote that, I wasn't even trying to support those.

I fixed both issues here (Python 2.3.5, wxPython 2.6.3.0 ansi, msw XP
(pt_BR)), and replaced the 'iso639_languageDict' by some code that
finds out what are the currently wx supported canonical forms of
current locale names. It seems to be working.

Great! Thanks for doing it.

May I update the script on the wiki page? Or should it be tested on
other platforms before?

Nooo problem! Go ahead!

  Volunteers, anyone? 8^)

Sure, I can test on my platform. I am running Python 2.4.3, wxPython
2.6.3.2 unicode msw XP and I'll try your update. I just recently
working over some wxPython stuff again so it'll fit right in.

Ok, updated. FWIW I'm also writing a kind of wxPython front-end for
the mki18n.py script and I hope to make it available in a couple of
days.

-- tacao

No bits were harmed during the making of this e-mail.

David Woods wrote:

  > I've gave up using this script since I couldn't figure out how to get it to

support Russian or Chinese. Perhaps someone more clever than I can figure
this out too.

David what problems did you run into? Do you have .po files in any of those languages? I am revamping the script a little with E.A.Tacao's additions. I'll see if I can help.

···

--
Pierre Rouleau

address = 'zNbOpS.PfAxMe@bMjEgSrAaMuIgGnOcSz vA@PhInPrYyThHbOeNc'
Decode it with: address[::-2].encode('nospam@31tor.com'[7:-4][::-1])
Python-ez bien!

David Woods wrote:

  > I've gave up using this script since I couldn't figure out how to
get it to
> support Russian or Chinese. Perhaps someone more clever than I can
> figure this out too.
>
David what problems did you run into? Do you have .po files
in any of
those languages? I am revamping the script a little with E.A.Tacao's
additions. I'll see if I can help.

--
Pierre Rouleau

Pierre,

The problem I run into with a Russian file is that msgmerge reports an
"iconv failure".

I've attached the *.po file for Transana's Russian translation, should you
care to explore it further. mki18n.py is wonderful for automating the
process of creating translation files, and I use it for those translations
it work with, but I end up using poEdit's *.mo file for Russian.

Thanks for looking into this.

David

Transana_ru.po (119 KB)