Question #68881 “I am new to Launchpad, what should I do next to...” : Questions : Localizable Sentences encoded as Unicode Characters

Revision history for this message

verdy_p (verdy-p) said on 2009-04-27:

#1

Next, you need to create a source .pot file that will be used as the source for translations.
You can look at the format of this source file in one of the many other projects hosted here.

The format is quite basic: each message has an identifier, which generally is the english message itself (between double quotes) after the keyword "msgid" (without the quotes), but in your project it could be the name you have assigned to your characters, and that you'll use as the root pseudo-language. English resources can then be created from this, just like other translations.
These msgid lines can be prededed by comments (that will be visible on this site) appearing on lines starting by a hash sign (#) followed by a space.

If the hash sign is followed by some punctuation, it is used to indicated special formatting options, like the presence of placeholders in C/C++ printf-like formats, or in Java (the language can be parametered). For your project, I don't think you need placeholders. There's a hash-punctuation defined for checking the status of "fuzzy" translations.

In .po files that this site will help generate for you, these ids are followed by the translated message appears just after it, on the next line starting by the keyword "msgstr" (without the quotes). These .po files are generated by the site itself, but the translations in all other languages can be entered directly from the forms. To get the generated .po files, just use the "export" function.

Note that this site acts like a translation memory (meaning that messages that have the same msgid across all hosted projects can show automatically the proposed translations for similar messages in other projects). This can accelerate the production of reliable translations, or can help you verifying them.

Be careful about the intended licence for the translations: they are shared by default across all hosted projects. Read the FAQ if needed. If the policy does not match your project, just cancel it and find another hosting solution. I cannot decide it for you.

Next, you need to create a source .pot file that will be used as the source for translations.
You can look at the format of this source file in one of the many other projects hosted here.

The format is quite basic: each message has an identifier, which generally is the english message itself (between double quotes) after the keyword "msgid" (without the quotes), but in your project it could be the name you have assigned to your characters, and that you'll use as the root pseudo-language. English resources can then be created from this, just like other translations.
These msgid lines can be prededed by comments (that will be visible on this site) appearing on lines starting by a hash sign (#) followed by a space.

If the hash sign is followed by some punctuation, it is used to indicated special formatting options, like the presence of placeholders in C/C++ printf-like formats, or in Java (the language can be parametered). For your project, I don't think you need placeholders. There's a hash-punctuation defined for checking the status of "fuzzy" translations.

In .po files that this site will help generate for you, these ids are followed by the translated message appears just after it, on the next line starting by the keyword "msgstr" (without the quotes). These .po files are generated by the site itself, but the translations in all other languages can be entered directly from the forms. To get the generated .po files, just use the "export" function.

Note that this site acts like a translation memory (meaning that messages that have the same msgid across all hosted projects can show automatically the proposed translations for similar messages in other projects). This can accelerate the production of reliable translations, or can help you verifying them.

Be careful about the intended licence for the translations: they are shared by default across all hosted projects. Read the FAQ if needed. If the policy does not match your project, just cancel it and find another hosting solution. I cannot decide it for you.

Revision history for this message

verdy_p (verdy-p) said on 2009-04-27:

#2

Some other resources may help you: look for GNU "gettext" howtos, and other relted information about internationalization within GNU libraries. Because .pot translation templates and .po translated files are in the source format used by GNU gettext (which is used not only in C/C++, but also now in Java, Python, Perl, VB... and many others, including various CMS systems for translating web sites (HTML pages, JSP/ASP pages) or blogs (there's even a gettext-compatible plugin for MediaWiki), or various help file formats.

Revision history for this message

William Overington (wjgo-10009) said on 2009-04-28:

#3

Thanks verdy_p, that solved my question.

Revision history for this message

William Overington (wjgo-10009) said on 2009-04-28:

#4

The above is the automated message.

Thank you, I shall try to produce a .pot file for the sixteen localizable sentences thus far being used in the project.

William Overington

28 April 2009

Revision history for this message

verdy_p (verdy-p) said on 2009-04-28:

#5

If you already have a source file containing those source strings in one of
the languages supported by xgettext (like C, C++, Java, PHP, ...)), just
mark those strings using the _() macro around those strings in your source
(see the GNU gettext howto), and then extract the .pot file from this
source using the xgettext tool (that is part of the installable gettext
package).

Of course you can easily edit the .pot file yourself. It's not complicate to
do (but the first msgid in your source must contain an empty string id used
to give a description, there are also special comments at the start, used to
specify various localization options).

Template .pot files and translated .po files have the same format: the only
difference is that .pot files contain empty strings in the "msgstr" line
that follows each "msgid" id containing the translation source. You can
check he format of the .pot file using the gettext tools, or by trying to
import it to your Launchpad project (it will run it for you, however, if
this fails, you will have few details in the email received after the import
has been performed, so installing the gettext tools on your PC may help).

Common syntax errors in .pot/.po files include missing quotes, or duplicate
msgids, or invalid tokens in option lines (lines starting with "#,"). Make
sure that comment lines start by a "#" followed by at least one space (the
first character after "#" is significant).
Note that long quoted strings can be split on multiple lines by closing the
double quotes on one line and reopening it on the next line. You also need a
special comment at the start to indicate which ending (UTF-8) you are using
in your .pot template file.

William Overington
> Envoyé : mardi 28 avril 2009 08:51
> À : <email address hidden>
> Objet : Re: [Question #68881]: I am new to Launchpad,what
> should I do next to progress this project please?
>
>
> Question #68881 on Localizable Sentences encoded as Unicode
> Characters changed:
> https://answers.launchpad.net/locsenchar/+question/68881
>
> William Overington posted a new comment:
> The above is the automated message.
>
> Thank you, I shall try to produce a .pot file for the sixteen
> localizable sentences thus far being used in the project.
>
> William Overington
>
> 28 April 2009
>
> --
> You received this question notification because you are a
> direct subscriber of the question.
>
> Aucun virus détecté dans ce courrier reçu.
> Analyse effectuée par AVG - www.avg.fr
> Version: 8.5.287 / Base de données virale: 270.12.3/2076 -
> Date: 04/27/09 06:19:00
>

If you already have a source file containing those source strings in one of
the languages supported by xgettext (like C, C++, Java, PHP, ...)), just
mark those strings using the _() macro around those strings in your source
(see the GNU gettext howto), and then extract the .pot file from this
source using the xgettext tool (that is part of the installable gettext
package).

Of course you can easily edit the .pot file yourself. It's not complicate to
do (but the first msgid in your source must contain an empty string id used
to give a description, there are also special comments at the start, used to
specify various localization options).

Template .pot files and translated .po files have the same format: the only
difference is that .pot files contain empty strings in the "msgstr" line
that follows each "msgid" id containing the translation source. You can
check he format of the .pot file using the gettext tools, or by trying to
import it to your Launchpad project (it will run it for you, however, if
this fails, you will have few details in the email received after the import
has been performed, so installing the gettext tools on your PC may help).

Common syntax errors in .pot/.po files include missing quotes, or duplicate
msgids, or invalid tokens in option lines (lines starting with "#,"). Make
sure that comment lines start by a "#" followed by at least one space (the
first character after "#" is significant).
Note that long quoted strings can be split on multiple lines by closing the
double quotes on one line and reopening it on the next line. You also need a
special comment at the start to indicate which ending (UTF-8) you are using
in your .pot template file.

William Overington
> Envoyé : mardi 28 avril 2009 08:51
> À : verdy_p@wanadoo.fr
> Objet : Re: [Question #68881]: I am new to Launchpad,what 
> should I do next to progress this project please?
> 
> 
> Question #68881 on Localizable Sentences encoded as Unicode 
> Characters changed:
> https://answers.launchpad.net/locsenchar/+question/68881
> 
> William Overington posted a new comment:
> The above is the automated message.
> 
> Thank you, I shall try to produce a .pot file for the sixteen 
> localizable sentences thus far being used in the project.
> 
> William Overington
> 
> 28 April 2009
> 
> --
> You received this question notification because you are a 
> direct subscriber of the question.
> 
> Aucun virus détecté dans ce courrier reçu.
> Analyse effectuée par AVG - www.avg.fr
> Version: 8.5.287 / Base de données virale: 270.12.3/2076 - 
> Date: 04/27/09 06:19:00
>

Revision history for this message

William Overington (wjgo-10009) said on 2009-04-30:

#6

I have been informed as follows.

quote

What you want is the GNU gettext tools. You'll find extensive documentation here:

http://www.gnu.org/software/gettext/manual/

end quote

Localizable Sentences encoded as Unicode Characters

I am new to Launchpad, what should I do next to progress this project please?

Question information

Subscribers

Localizable Sentences encoded as Unicode Characters

I am new to Launchpad, what should I do next to progress this project please?

Question information

Related bugs

Related FAQ:

Subscribers