[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-gnubg] Untranslated items
From: |
Jim Segrave |
Subject: |
Re: [Bug-gnubg] Untranslated items |
Date: |
Sat, 6 Dec 2003 23:57:38 +0100 |
User-agent: |
Mutt/1.4.1i |
On Sat 06 Dec 2003 (19:41 +0100), Petr Kadlec wrote:
> Hi all!
>
> I have a few of translation-related issues again.
>
> First, a bit of good news: After all those problems with charset in
> typewrite-style text (sorry for the About bug...), gnubg is now able to
> display the analysis of the position in correct Czech! But, the news is not
> so good: the solution was to uninstall "OCR A Extended" font, which is a
> fixed-width font that does not have east-european charset. I made many
> attempts with various entries to gtkrc file, but, if the font is present,
> gnubg chooses it, whatever is in the gtkrc file. Without the font it works
> OK even without any gtkrc file. So that I removed the font for now, but if
> someone of you knows where the cause of the problem is, or what could be
> done to force gnubg not to choose the font, please, let me know.
That's very odd - I'd like to know the name of the font as GTK sees
it (eg. the ones that take a form like
-adobe-courier-medium-r-*-*-10-*-*-*-*-*-*-* ). I don't know enough
about how Windows does font mapping (I won't claim to be an expert at
how X does it either), but I wonder if there's a font aliases type
file which maps names like the above into actual font files and which
has some curious default setting. I don't see any obvious ocr fonts on
my Linux or FreeBSD machines.
> 2. To Jim Segrave: in an e-mail from 19.11., you wrote that you have fixed
> some untranslated strings. Well, I have checked it now, using both source
> code snapshot 20031203 and Windows build "031203 (build Dec 5 2003)"
> (quite strange name, isn't it :-o ). And the result is -- negative. All
> items are still untranslated. Have you committed the fixes to the CVS?
CVS is down at the moment, so I can't check, but I do remember
changing a bunch of string and regenerating po/gnubg.pot. Since the
ChangeLog files contain my comments, I assume I did commit changes. I
have failed to commit a change entirely, but I usually don't update
Changelog without having commited the changed files as well.
> 3. As my translation is practically done (I am just cleaning some doubtful
> terminology now), I would like to ask some general things:
> Where/whom should I put/send the message file when I will have it
> done?
You are welcome to send it to me and I will commit it (along with
updating po/LINGUAS), once CVS is up
> Is it possible to add an entry to the about box with something like
> "Translation: author <email>"? I mean, I don't need "a separate credit
> line", it just seems to me that it would be fine if users of the localized
> version knew where the bug reports should go. (I was thinking about adding
> such line into the translation of "With special thanks to:", separated by a
> newline, but is that acceptable?)
I think that's an excellent suggestion - I'd expect you to have a
credits line as well, but having the translation of this thanks line
also give a translation contact seems a very good idea. We'd want to
ensure that the translator was happy to have their names listed there
of course.
> 4. Some more untranslated items I have found:
> gtktheory.c
> 474: "Gammon values at %d-away, %d-away (%d cube)\n\n"
The parentheses are there, but no '_'. But this is in a
g_strdup_printf() call, folowed by a gtk_text_insert() call. I assume
that if the underline were present then we'd get the translation in
the g_strdup_printf, but that gtk_text_insert wouldn't attempt a
second translation?
> 484: "Player Gammon BG\n"
Is now a bit further down and does appear in gnubg.pot
> bearoff.c
> first at 1119, more occurrences follow:
> " Player Opponent\n" "Position %12d %12d\n\n"
> 1227: " "Mean\t%7.3f\t%7.3f\t\t"
> 1242: "Std dev\t%7.3f\t%7.3f\t\t"
Yes, all missing
> show.c
> 2100,2102 aszVariations is not given to gettext, although it should be.
But they are defined in eval.c such that they should be statically
installed with the translated string I think
> format.c
> 900: " cubeless equity %s\n"
This one needs fixing as well.
--
Jim Segrave address@hidden