Just perfect. Just “sudo apt-get install kbabel” and you’re there. Also install the poxml package, it’s in Ubuntu. You’ll need it to test the translations; the “po2xml” program takes the original English XML file and a PO file that includes both the English and translated strings, and produces a corresponding translated XML file. See “man po2xml”.
I really recommend the KBabel. Poedit (“apt-get install poedit”) is an alternative, but I haven’t used it and can’t say much about it. KBabel is a KDE program but runs just as well under Gnome. You should probably not start it with “kbabel”, but run “catalogmanager”, which allows opening a project that contains multiple PO files, as Book of Vaadin does.
In Catalog Manager, go to Project->New. Give conf file name “book-of-vaadin.kbabel” or something, Polish as the language, and “other” as the project type. Set the base folder for the PO files a folder that contains the translated files. This should be “manual/translations/pl_PL” under the documentation source tree (“svn co http://dev.vaadin.com/svn/doc/trunk”). Give the “manual/translations/pot” under the doc source tree as the root folder for POT files. Click Finish. The Catalog Manger should now scan both the PO and POT folders and list the translations. There are no existing translations for Polish so it’s all untranslated files.
Below is a project for the Finnish translations (there’s only the book.po):
When you click a file, it automatically copies the POT file to a PO file (in the pl_PL folder), which will contain the translations, and opens the actual KBabel editor.
You probably want XML validation and such for the BoV files, so go through the settings and enable the XML validator, if it’s not already enabled (it should be).
Oh, and please set the translator identity in the project settings to get your credits correctly so that’s it’s at least included in the PO files (see the header section of
http://dev.vaadin.com/browser/doc/trunk/manual/translations/fi_FI/book.po
for example). We’ll have to put some sort of translator field in the generation scripts to have translator names included in the book properly.
(POT files are template files for translating, do not edit them. A POT file is in the beginning copied to PO file, which is translated and then contains both English and translated strings. As the documentation is updated and new POT files generated, it’s possible to merge (with msgmerge) the new texts to the POT files, thus reusing the old translations. Strings that do not match exactly but have a close match are marked as “fuzzy” and need to be checked. This happens, for example, if one word changes in a paragraph.)
Just ask me if you have any questions.