http://www.mediafire.com/?b0e1ka46l49r0u2 (9/09/2013)
Here is the current batch of translation files. I'll edit this post as necessary to update the batch, e.g. when 3peso releases a new official patch. I'm open to suggestions for proper collaboration, although I think setting up a svn/git repo is overkill, while dropbox lacks the diffs (I think).
strdump.txt is a complete text dump. Nothing should be missing from this, but if you find something in game please transcribe it and I'll try to track it down. I've removed the filtration that was causing problems... which means it's now even more of a mess.
strdump_current.txt is the complete text dump with all of the translations, all in one place.
Translation_From_LoT1.txt is all of the perfectly matching translations from LoT1. Anything in here should be used again in LoT1 for stuff like item names and all (The main exception is "うふふ", which should be "Ufufu", because Ufufu is the best)
All of the other text files are purely translations. The names should be more or less self explanatory.
The format of these files is simple:
(8 bytes address): <text>
The colon and space are
mandatory, just in case I want to add any other fancy stuff to the format. Any line that does not follow that pattern will be ignored. You can use this to add comments if you like, but right now these will not carry over when I port the files to a new patch.
To translate stuff, find the Japanese in the string dump and copy the address. Paste the address with the translation into a new file or as an edit to an existing translation file. Please do not try to edit the string dump directly, that's just too messy. There is no need to keep the lines in order, nor to copy over anything from the stringdump you do not intend to translate.
All files must be saved as ANSI. They absolutely cannot contain any fancy characters outside of Shift-JIS. For example, word's angled quotation marks, apostrophes, and long hyphens. They will show up in game as random Japanese.
An example. Japanese is fine, but it's usually a good idea to not mix half-width roman characters with full-width Japanese characters. It can potentially break things.
The only files with special formatting are the dialogue files. Most lines will contain a @. This is a linebreak and will drop the text down to the next line. Do not put any spaces
after this character - before is fine if you want it for readability. There is no manual wordwrapping. Treat them the same as any other linebreak - if you need to drop down to a new line, do so, even if the sentence continues in the original and you aren't at the end.
Lines with stuff like "%d" or "%s" are using character substitution. If you need to add a literal %, type in "%%". Please make absolutely certain that you do not change the total number of unpaired percent signs in these lines. "Adds %d% to some stat" will not have the expected results. Instead use "Adds %d%% to some stat"
Please document any and all changes to the translation in the spreadsheet linked in the OP. This is to help maintain consistency with the translation and to make it easier to find and change them in the future if necessary. References should also be mentioned in the spreadsheet for the same reasons. Also, whenever possible, keep translations consistent with the first game. For example, "マエリベリー･ハーン" should be left as "Maribel Han", since that's what she was in the first game (and this is not changing unless ZUN himself explicitly spells out how to romanize her stupid name).
