Looking at Localisation

Today I’ve also had a look at the options available for localisation of D&D Manager. I’ve started out by creating some of the files with lists of strings to translate. I’ve created a strings file for each of the user interface “.xib” files that are used to create the interface. This has produced a series of .strings files which I’ve uploaded to my website.

These files can be downloaded in one big zip here.

From these files you have to convert the text after the equals sign on each line to the language you are translating to. i.e. in the following example

/* Class = “NSBox”; title = “Box”; ObjectID = “14489”; */

“14489.title” = “Box”;

You are wanting to translate the word Box, so (assuming translate.google.com is holding up) if you wanted to translate that into Spanish you’d land up with:

/* Class = “NSBox”; title = “Box”; ObjectID = “14489”; */

“14489.title” = “Recuadro”;

Now that seems fairly simple, and that’s all that has to be done for each entry. Now for the bad news, there are around 2500 strings to translate so its going to be a fair bit of work. Fortunately though there are quite a few duplicates (like the box example where there are actually 170 different uses of the word Box, so with multi-file find and replace you can do them all in one go.) and there are also things like +, – etc. which probably won’t need translating.

Although there are 170 versions of the word Box appearing that is one of the more common things to appear in the user interface files – and actually I don’t know if it appears in the final UI at all so it could be skipped. I’d personally make a very, very rough estimate that there are 750 unique strings that will need translating (though there could easily be as many as 1000).

To make the situation even worse this isn’t actually a complete list. There are approximately a further 200 – 250 strings that are created programatically, but quite a few of those will overwrite some strings in the user interface – it might be difficult to be sure which is which, but if you load the program, any strings which change are created programmatically. I haven’t got them in this list as they aren’t in the form required for them to be translated into different languages.

If you are still interested in translating to your language add a comment to this post so everyone can see what has been done, also if anyone is interested in doing a decent amount of translation then I’ll convert the 200-odd strings that are created programmatically into a form which means they can be translated easily.

Its also worth noting that the localisation files have been created from SVN revision 185, in case anyone wants to re-create them later

EDIT: To save the irony of the only text in the program not being localised being the menu item which links to this post, I also need: Information on Localisation translating.

Posted in Software and tagged , , .


  1. Pingback: Matthew’s Software » Localisation part 2

Leave a Reply

Your email address will not be published. Required fields are marked *