ARM Backup/Ar tonelico 2 retranslation/EvdSynchroTools

From EXA_PICO Wiki
Jump to navigation Jump to search

From the original post:

Morgan & Antares tiredly presents, the evd Synchro-Tools!

http://winter.limitedwish.railsplayground.net/evd_synchrotools_1.0.rar

As I mentioned, the use for these is a bit annoying...

Basically, you need to use rpkextract.rb with the --postfix option to extract the evd files from the NA and JP versions. If you do an extraction *without* the postfix as well, the text from that copy will be used to prefill the "new" area. (The idea behind this is to be able to move text betweeen different programs using the evd files as the intermediary.)

evd_dualextract.rb will produce a .txt file for each __EN.evd file it finds. (0__EN.evd and 0__JP.evd produce 0.txt.) If the .txt file already exists, it will skip that file. This is to reduce the risk of accidentally overwriting a work in progress.

evd_dualrepack will look for .txt files for all the __EN.evd files, and build (not postfixed) .evd files from them.

If the "new" line is blank, the text won't be changed. You can also delete a block entirely to leave it unchanged. Line breaks and blank lines have no effect on the inserted text, so you can insert them if it's helpful for some reason. Lines beginning with // will also be stripped out (for comments).

Both programs take a while to run if applied to the full set of .evd files, so they'll print a line for each new file to let you know it's hasn't frozen up...

Still using 2-byte text.

... I think that's everything.

-Morgan, not good at writing documentation...

A warning...

Nothing can be placed before the line at the beginning of the text file saying ":RAW:", or the repacker will reject the file.

About names_2.txt:

This file is used to convert the numbers used within the evd files to names, and is pretty simple to view and edit. For example, the line:

0=Luca

means that whenever a line has a name id of 0, the extractor puts "Luca" in the file.

Unfortunately, the order of names to name ids doesn't seem to match the order of names in the image files, the executable, or anything else I can find, so only a relatively small number of names will be marked.