Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[rummeltje]

I understand that this is completely out of scope for normal use of clrMame,
but I would love clrMame to be able to recognize cbr cbz and cb7 as archives and treat them as such.

As a retrogaming enthusiast I have worked with clrMame for Mame Roms, and I know clrMame is a great program.
And as a comicbook collector I think it that clrMame could be the awesome tool in the comicbook department as it is in the rom department.

cbr stands for ComicBook Rar, and are images inside a rar archive, but the extention changed to .cbr
cbz stands for ComicBook Zip, and are images inside a zip archive, but the extention changed to .cbz
cb7 stands for ComicBook 7z, and are images inside a 7z archive, but the extention changed to .cb7
(cbt stands for ComicBook Tar, and are images inside a tar archive, but the extention changed to .cbt) *

*cbt is very rare and not many comicbook readers support it

I and some friend have been already using clrMame to scan the collections of comicbooks of ourselves,
but clrMame does not scan inside inside the archives. It instead treats them as roms.
We want it to scan inside and crc the images, and not just crc the cbr archive as a whole.

All the functionality is already present in clrMame,
except a way to tell clrMame that these cbr / cbz / cb7 / (cbt) comicbooks are archives with renamed extentions.
(If I rename the extention cbr back to rar, clrMame does exactly what I want and crc the images inside the archive)

So, Dear Sirs, Gods of Coding, could you please make an option to give in archive aliasses,
or have clrMame recognize cbr, cbz, cb7, (cbt) as archive natively?

[Roman]

hmmm.....if it's just a real zip or a real 7z or a real rar then there might be a chance with some kind of extension alias (tell cmpro that .zip is .cbr). That would be relatively easy I guess....

The problem might be that the used libraries might have additional tests on the filename extension (especially when it comes to split archives...but they might not play a role here). If the files are not only renamed but also used different identifiers inside the archive structure...well..then you might face another problem

I will check what's possible.....do you have some example files by the way?

[rummeltje]

Yes, they are real archives.
Standard, non-split.

example files here:

https://drive.google.com/drive/folders/1zl3c6sL2O3liAtO3lvlYFIuLbZzVnh9K?usp=sharing

[Roman]

Thanks for the files. I've did some tests and yes it works pretty fine and easy when .zip is simply replaced with .cbz alias (same for rar/7z)
What would be more interesting is when you want to support both .zip/.cbz at the same time. So for example rebuilding to .cbz but reading .zip files etc...
That'd require some additional coding since you a) somehow need to define what the prefered writing alias is and b) there are some checks which need to run through all alias then.

Since it's nearly holiday time, I put it on my to do list....but yes, in general it seems to be not too much work....

[rummeltje]

I am really happy that you are seriously looking in to this.

 

Being able to scan the pages and have others rebuild them, would be a leap forward for our comic collectors group.
We could rename all pages to a standard naming scheme, and the cbr / cbz to a standard naming scheme.


Some thing comes to mind though:

1- I vagely recollect that some comics use 1 folder or multiple folders inside the archive.
I don't know if that poses any extra challenges?

2- Input wise, it would be 99 % cbr or cbz
but it would be awesome if a zip / rar / 7z containing the images or folder containing the images would also be possible, so all bases are covered.

3- Output wise :
cbr (rar max compression)
cbz (zip max compression / torrentzipped)
cbz (store)
cb7 (7z max compression / torrent7z)

Most of the comic book scene prefers CBR RAR at max compression, because RAR was traditionaly better at compressing images then ZIP, and they are kind of stuck with that. (Even though 7z performs equally well and the differences with zip are small)

I personaly started prefering CBZ ZIP (store), because it opens much faster on my slower devices.
ZIP is accepted and integrated pretty much everywhere, windows can browse it like a normal folder these days.
and because it's store mode, there's no need for the better compression of 7z / RAR.
With HDD size and price being what they are, I no longer care that much about my collection being 1 TB or 1,1 TB


Enjoy your holidays
And thanks a million already.

[Roman]

1) folders are supported. They need to be listed in the datfile though. So instead of rom name="1346b.cpu-u25" size="2048" crc=".... you got rom name="my_sub_folder/1346b.cpu-u25" size="2048" crc="....

2) yeah, definetly the better way to simply understand zip and cbz as an alias for it and handling both.....need to think about it a bit...

3) compression settings can be setup in the rar/7z tabs of settings->Compressor. For zip you can define the compression level in cmpro.ini (Packer_Zip_CompressionLevel = 9)


since holiday is just some days away, don't expect anything anytime soon ;-)

[rummeltje]

I've been thinking about this idea for about 8 years, so no need to hurry.

Enjoy your holidays. 

[Roman]

Did I say "do not expect anything anytime soon"? Well....looks like I was wrong...

I did some restructuring of the packer code and actually cmpro is now able to handle anything which is "like 7z/rar/zip" based.
And all at the same time. So cbz, cbr, cb7 isn't a problem....and of course you can think of jar, war etc as well.... ;-)

So where is it? Well, of course I want to test it a bit more so I can't promise if I put out a nightly before my holiday....

[Roman]

https://mamedev.emulab.it/clrmamepro/binaries/cmpro20200704.rar

Beware...this is just a very early test!
You should use a new separate installation for it (going back from it would kill your set compression settings (like rebuilder pack type etc) at the moment, so better use a fresh install ;-) )

By default you of course got rar/7z/zip support.....You can add additional ones by editing cmpro.ini and add/modify:

Packer_ZIP_Alias = .cbz
Packer_RAR_Alias = .cbr
Packer_7Z_Alias = .cb7

The alias settings allow multiple values separated by space, so e.g. Packer_ZIP_Alias = .cbz .jar .war

If you find anything weird, collect and post the quirks....I will look at it after my holiday...

[rummeltje]

You are fantastic.

Downloading it now. Will test over the week and reply my findings back to you.

[rummeltje]

1st test seems to be all ok.

[rummeltje]

I've done a few tests and it all seems to be working fine.
I have made a test sample of comics in cbr format and performed dat2dir,
then took all the cbr and changed them to rar, zip, 7z, cbz, cb7 and folders containing images and performed several rebuilds.
It seems to do all I have asked for, without problems.
Thank you so much. This is a huge step forward towards organizing the comic book scene.

 

1 question: How can I set clrMame so it outputs the unneeded files in 1 seperate directory?
There are much more versions out there then we have indexed so far.
And unneeded = Unrecognized and we want people to send those to us for indexing.
As with most retro gaming collections, there's 2 lists that are important.
1 to recognize all the comic books out there and tag them as such. (like Goodtools)
1 to recognize the current best version and rebuild that set (like No-Intro)



The next steps are completely out of scope for clrmame, but I'll mention them anyway.
-Special torrentzip / torrent7z so that they can work with cbr / cbz / cb7
-Get a scanning method independent of image EXIF and other metadata, only the raw image data.

[Roman]

Good to hear it is working 😁
Regarding your question: unneeded files are moved to the backup folder when being fixed. If the fix option is not enabled you should have something like a copy to or move to context menu option for all listed sets. At least if I remember that correctly. Cant check at the moment.

Special torrent7z etc: do they really need to be different? Currently cmpro uses the settings from settings/compressor/7z. The 7z exe doesnt care about extensions. Maybe it works right away with torrent7z.

Cmpro supports headers by using xml definitions where you can specify rules how checksums are calculated. Maybe a new header definition can be easily created for exif. I will check if I find the link to the xml header documentation.

Update: here is the link https://mamedev.emulab.it/clrmamepro/docs/xmlheaders.txt

[rummeltje]

Thanks for the link to xml headers.
I'll read up on that.

I tried torrentzip and torrent7z

torrentzip does not accept cbr / cbz / cb7 at all.
so it would need work to accept them, and output as .cbz

torrentz7 does accept them, unpacks and repacks them, but the resulting archive has the extention .7z
so it only needs the output extention changed to .7z