![]() ![]() It's really a long process (and story!), but hope you enjoy the results. For the final font, I scrolled downwards to see some 00's, scrolled upwards to see the end of file, then used "Edit > Delete from" to delete all the 00's. I repeated this whole process few times until I get all the fonts. Renamed the file to "ttf.ttf", opened the font, looked at the name, then renamed the file based on the OBB filename again. For the next fonts, I extracted the file using OBBPatcher again, fired XVI32 and opened the section file, searched for OS/2 again, scrolled upwards to find the beginning of the file (it's marked with the bytes "00 01 00 00" as mentioned above), used "Edit > Delete to" to delete any stuff before it, searched for OS/2 again, scrolled upwards to see the end of file again, then used "Edit > Delete from" again. After opened the font, I noticed that the font name is "Ashley Script MT", so I renamed it to AshleyScriptMTStd (from the OBB filename), and done. They're actually full of 00's so that's why I know it's the end of file, used "Edit > Delete from" to delete any stuff after the end of Ashley Script font, saved, renamed the file from "ction" to ttf.ttf (well, at first I didn't know what font is it). So, I searched for "OS/2", then pressed F3 (to search for the next font after Ashley Script), scrolled upwards to find the end of Ashley Script font. I was actually confused at first, since it only outputted one font (Ashley Script), but when I hex edited it, I found the problem: The fonts seem to be mixed into one file. I just realised this when I double-checked the offsets. The fonts uploaded here are the one that successfully extracted, so don't worry. When I compressed both files, they have different CRC32 checksums, so that means there are some corruptions going on with the failed one. It says "Extraction incomplete", but I found out that it still wrote the file to ction.Īctually, it failed to extract the files successfully because some bytes are not selected (my bad!) It should be The result is 331A5, then I switched to decimal mode to see the decimal of it, and it says 209317, so I typed this on the versions.cfg:Īfter that, I fired up OBBPatcher, selected the OBB, used the fontsobb preset, and extracted it. I fired up the Windows built-in calculator, switched to programmer mode, switched the mode to hex, then subtracted 2EBC1A5 by 2E89000. They're the actual files, and I found where the compressed block starts and ends, it starts at 0x2E89000 and ends at 0x2EBC1B3. T T F"), and found another one, it's at 0x2E880CA. ![]() The first offset found is at 0x2875E, but it's not the actual files, I kept pressing F3 (to find the next thing that contains ". TTF is the file extension, then the dot (.) is obviously not the first character in the filename, so no need to find the gibberish things manually.Īfter that, I typed ".TTF" again, but clicked the "Text -> Hex" button, just to convert it to hex, and add three 00 bytes on each characters: 2E 00 00 00 54 00 00 00 54 00 00 00 46 (which means ". Actually, every filenames listed in the OBB are spaced by three 00 bytes for each letters, and the first letter in the filename has some gibberish characters in it, e.g., Pico12 -> P(random characters here) I C O 1 2, so that's why I didn't get the file! However, since. I didn't give up, instead, I looked at the versions.cfg, just to see the offset of the RTON file and find a way to get the TTF filename. I've tried to directly find the TTF file by searching ".TTF" in the search box. After a while, I realised that OBBs are compressed and therefore all of the data inside the OBB are compressed. I opened the OBB, then using the find command, I typed this on the text box: "OS/2", but got no results. This is where the fun begins: I'm started playing with the XVI32 hex editor. See the OBBPatcher's readme file for more info. As mentioned in the OBBPatcher's readme, the pattern of compressed files are always like this: pgsr.Then, I learned the characteristic of OBB files: I used these characteristics to my advantage, since these applies to every single TTF fonts. If you use a hex editor to view the font's data, you'll ALWAYS find the text "OS/2" inside the font's data, which possibly means that TTF fonts are compatible with the really old IBM OS/2 from the 90's.Well, surprisingly enough, there are some unique characteristics of TTF: Well, I was bored and decided to mess around with OBBPatcher to see what I can find, but after a few moments I decided to find the PvZ2 fonts inside the OBB, since no one ever done that yet.Īt first, I learned the characteristics of TTF fonts. It's a pretty complicated process to extract all the fonts successfully (I have some problems extracting them correctly), but using the XVI32 hex editor I was be able to do it. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |