From 10bdc7acdda9e9d27e69f0fa8097605d4d90a48c Mon Sep 17 00:00:00 2001 From: DTB Date: Wed, 20 Mar 2024 20:31:05 -0600 Subject: [PATCH] file explanation --- cbz/cbarchives.txt | 75 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 75 insertions(+) create mode 100644 cbz/cbarchives.txt diff --git a/cbz/cbarchives.txt b/cbz/cbarchives.txt new file mode 100644 index 0000000..afe317a --- /dev/null +++ b/cbz/cbarchives.txt @@ -0,0 +1,75 @@ +Comic Book Archives (CB7, CBR, CBZ) + +A Comic Book Archive is just an archive with a bunch of sequentially-named +images in it which collate ASCIIbetically to the order in which they should be +read. Each image corresponds to a page presented by a comic book reader and a +Comic Book Archive represents a digital comic book. + +No guarantees can be made regarding image format, image names (though you can +expect them to match [:digit:]*\..*; that is, a bunch of numbers followed by a +file extension), archive attributes (compression or metadata), or what is in +archives besides the images. In fact, when a comic book only has one image, the +most common type of file in a comic book archive may not be an image at all. + +The extension corresponds to what type of archive is being used. I haven't seen +an extension that hasn't fit within DOS 8.3, which makes sense as amateur +skiddies and repackers mostly use Windows. + + + + +Here's a table of Comic Book Archive types. I can't imagine this list is +comprehensive but I can't find more on-line. + ____________________ +| Ext | Archive used | +|-----|--------------| +| CB7 | 7-Zip | +| CBA | ACE | +| CBR | RAR | +| CBT | tar(1) | +| CBZ | PKZip | +'-----'--------------' + +I normalize the files I get to the following settings: + +ARCHIVE: PKzip. DEFLATE algorithm. No other configuration. +CONTENTS: ONLY images. Whatever encoding in which I found them. + Sequential naming starting from one, with leading zeroes to ensure file + names are all the same length. + + +7-Zip is free and open source. There are a number of implementations and you +can easily extract 7-Zip archives on all modern systems. + + +ACE is a proprietary archive format owned by e-merge GmbH. Nobody uses this. +There is a free software extractor written in Python available from + and free software Python implementations +available for most popular systems. + + +RAR is a proprietary archive format owned by win.rar GmbH. It and CBR are both +unfortunately pretty common because RAR is popularly considered better at +compression than PKZip. The reference implementation has a license that is a +little more permissive than a contract with the devil and support for later +versions is spotty in free software. I've found it's best to bite the bullet, +use the reference unrar(1) utility, convert my CBRs to CBZs, and hope I never +need to use it again. + + +TAR (tape archive) is an archive format released in its first incarnation in +1979. It doesn't do any compression and it's easy to extract files even by hand +with a hex editor if you can read the binary structure (which is thoroughly +documented). Tar extractors are ubiquitous, excellent, and built into every +modern operating system (this notably does not include Windows because +Microsoft sucks). Later varieties of tar (such as ustar) are standardized (in +IEEE 1003.1-2017) and files in this format will likely be readable for a very +long time. + + +PKZip is free, open source, and like tar's later varieties, standardized (in +ISO/IEC 21320-1.2015). Archivers and unarchivers are ubiquitous and available +for all modern operating systems. The format is officially called ZIP but I +call it PKZip after its original implementation (PKZIP, by Phil Katz) and to +differentiate it from other "zip" names such as bzip, gzip, and xzip (which are +all compression algorithms).