calibre-devs team mailing list archive

[Kovid Goyal <kovid@xxxxxxxxxxxxxx>]

That's great, I'll integrate it into calibre at the earliest opportunity. From 
my perspective (having to maintain build systems on 3 platforms), the best 
solution for lzxcomp is to simply add it to the current lzx module. This is 
because calibre uses the python distutils build system to compile c code with 
minimal effort on my part. Using it via ctypes would mean maintaining an extra 
set of 3 build scripts, unless you know how to get distutils to produce C 
libraries?

I suggest putting oeb.py into the calibre.ebooks.lit package for now as it 
will only be used there. In the future, if we decide to use it elsewhere, for 
example, for a mobi output plugin, we can move it into ebooks.metadata

Incidentally how do you losslessly serialize opf 2 to opf 1?

Kovid.

On Thursday 04 December 2008 11:11:41 Marshall T. Vandegrift wrote:
> Kovid etc,
>
> As mentioned on moibleread, I've got some oeb2lit code which I should
> have ready-to-go pretty soon (perhaps tonight even? although doing some
> functional testing might be a good idea).  Aside from solving/punting-on
> the "ahc"/"aht" hash algorithm issue, there are a few integration issues
> I need to sort out:
>
>   - LZX compression integration.  I found some C LZX compression code at
>     [1].  I decided to use this as an opportunity to learn the 'ctypes'
>     module, so right now I build liblzxcomp as a pure-C shared object.
>     But having done that, I'm now not sure how to bundle it in with
>     calibre, and in a way that my ctypes-using Python module can find
>     it.  I'd kind of like to avoid re-writing it as a C/Python module
>     because C types is really quite nice :-).
>
>   - Yet another OPF parser.  I wanted to be able to both (a) extract LIT
>     files then re-archive them without losing any OPF metadata, yet also
>     (b) LIT-archive e.g. EPUB books using OPF 2.0.  To achieve this I
>     ended up writing a new OPF-/Dublin Core-centric "Oeb" class which
>     can read either OPF 1 or 2 and losslessly serialize back out to
>     either OPF 1 or 2.  Should I just toss 'oeb.py' into
>     ebooks/metadata?
>
> Peace,
>
> -Marshall
>
> [1] http://www.russotto.net/chm/
>
> _______________________________________________
> Mailing list: https://launchpad.net/~calibre-devs
> Post to     : calibre-devs@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~calibre-devs
> More help   : https://help.launchpad.net/ListHelp
>
> !DSPAM:3,49382b7575721965720550!

-- 
_____________________________________

Kovid Goyal  MC 452-48
California Institute of Technology
1200 E California Blvd
Pasadena, CA 91125

cell  : +01 626 390 8699
office: +01 626 395 6595 (449 Lauritsen)
email : kovid@xxxxxxxxxxxxxxxxxx
web   : http://www.kovidgoyal.net
_____________________________________