Hi Calvin,
Quoth Calvin Morrison:
> I'm not sure why nobody has decided to use the documented way from Gtk
> webkit. I think it isn't very sloppy (unless I am just the only
> confused one here)
Probably either it wasn't got around to, or it was a mistaken
attempt to distance the browser somewhat from libsoup. I say
mistaken as, much as I dislike libsoup, surf is a browser resting
atop webkit-gtk, and needs to follow the appropriate conventions for
that in order to have a chance of being useful.
> The code required to do this is minimal, and in my code it took 2 lines [2]
>
> w.jar = soup_cookie_jar_text_new(g_build_filename(g_get_home_dir(),
> DEFAULT_COOKIE_FILE, NULL), FALSE);
> soup_session_add_feature(w.session, SOUP_SESSION_FEATURE(w.jar));
>
> When I do this I can use multiple processes to access and write to the
> cookies. Logging out in one process will logout me from the website in
> another process using the same site, once I reload the page.
That sounds good to me, sure. If you could sort this into a patch
and test it lightly against surf, that'd be awesome.
I'm still saddened by the lack of any reasonable browser. I found
hv3 the other day (tcl/tk based), but it isn't actively developed
appears not to run any more without hacking. I still hold out hope
for netsurf. I know I shouldn't care, and should get on with my
life, but I find that tough.
- application/pgp-signature attachment: stored
Received on Tue Feb 21 2012 - 20:52:32 CET