Redump Forum — PakkISO update (source?)

Re: PakkISO update (source?)

2012-07-10T10:36:11Z

Sounds impressive, iR0b0t. Agree with jamjam about renaming without decompression (impressive). Biggest gripe atm seems to be how to efficiently manage it...

Re: PakkISO update (source?)

2012-07-09T12:37:41Z

Renaming without decompression sounds impressive. Did you have to get your head stuck in the 7z format spec or was it not so bad to figure out?

Re: PakkISO update (source?)

2012-07-09T11:39:40Z

I began to work on a "pakkman" tool, which was supposed to manage updates from older datfile to the new one, with full image names instead of generic once.

>> pakkiso - for compression purposes only.
>> pakkman - file manager, the idea was/is:
1. to sort out all not matching titles.
2. after comparing the hashes renaming them without(!) a decompression, which works already fine here.
3. for titles where hashes changed or are not matching try an auto-fix, this is the most complicated part, as it has to be done in the proper order, but it is to manage
4. if the first three steps failed, delete or move those to a storage folder.
5. create a log file with all the changes!

that's it!
i hope to finish it someday

Re: PakkISO update (source?)

2012-07-09T07:55:20Z

Ah, thanks.

I think I may have an older version so I'll be sure and check it out.

Re: PakkISO update (source?)

2012-07-08T23:35:03Z

rmdtrash.exe was already abandoned in earlier versions! >> handled by -mta=off -mtm=off.
Scrub file attributes pre-compression >> handled program internally too.
check the version file.

a generic hash (.dat) file is included in the latest version, which i may have not released yet, i don't remember exactly.

will take a look later, as i don't remember which new functions were added in the latest (0.4d) release.

PakkISO update (source?)

2012-07-08T22:58:41Z

I have some suggestions for improving PakkISO and was hoping the author of the program would be willing to send me the source code to make a few changes. My ideas are based off of what I've been accomplishing via shell scripts up until now and, more recently, some of the work jamjam has been doing with CDGroup and 7z generic.

The suggestions are as follows:

- Update the compressor to 7z 9.25A
- Update the 7z cmd switches to: a -mx9 -m0=LZMA -mmt=2 -ms=on -mf=off -mhc=off -mhe=off -mtc=off -mta=off -mtm=off (the last two forego the need to use rmdtrash.exe as it instructs 7z to not store any timestamp information)
- Scrub file attributes pre-compression
- Generate and store a generic hash (.hsh) file with each pack containing bin/cue/iso name, size and sha1

Example:

Input:
Dino Crisis (Japan) (Track 1).bin
Dino Crisis (Japan) (Track 2).bin
Dino Crisis (Japan).cue

Output:
Dino Crisis (Japan) (Track 1).7z
Dino Crisis (Japan) (Track 2).ape
Dino Crisis (Japan).cue
Dino Crisis (Japan).hsh

Another option could be to output just a single file: Dino Crisis (Japan).7z which contains the previous output (using 7zCopy) but as follows:

Track 1.7z
Track 2.ape
*.cue
*.hsh

The controller would rename all binary contents pre-compression to generic track names and update cue sheet entries accordingly to allow renaming without a repack (updated archive file name in this instance). I'm just toying around with ideas for a cross-compatible update which outputs files that are easier to maintain than at present.

Thoughts and comments are welcome.