Re: Glib::IO

["max augsburg gmx de" <max augsburg gmx de>]

Dear Jeremy,

> I don't think you can make any assumptions about the encoding of the
> underlying file, or even that is is text rather than a binary format.
> I agree with you that the byte array should be converted into a byte
> string but beyond that the user needs to decide what to do with it
> based on the file type expected by their application.

I think, you are absolutely right! I just wondered, why in Python you
don't have to convert the byte digits in a byte string or to decode the
bytestring in UTF-8 format (see for example:
https://developer.gnome.org/gnome-devel-demos/stable/filechooserdialog.py.html.en)
But probably they have modified the Gtk3::TextBuffer->set_buffer()
method (I think the Gtk3::TextBuffer works always in UTF-8).

So I would let the Encode::decode stuff in the overrides out, and just
convert the bytes to a byte string with the pack() or the unpack()
functions...

> I went ahead and forked the Glib::IO codebase here:
>
> https://github.com/jvolkening/perl-Glib-IO
>
> I realize we can't send pull requests to the "official" github repo
> since it's only a mirrored copy of the GNOME repo, but the fork provides
> a place to collaborate prior to sending the patches upstream. If you'd
> like I can give you commit permissions. My work on
> it will probably be sporadic.

Collaborating on github is a very good idea, although I am still not
very familiar with github. On the other hand for me it is a great
enrichment to discuss my succested proposals here with more experienced
programmers. But perhaps this could be annoying for people that are not
interested in the Glib::IO module ;-)

My github account is https://github.com/MaxPerl

All the best,
Max