Closed GoogleCodeExporter closed 9 years ago
I'll fix it once the new version of the book store plugins is ready.
Original comment by andresgattinoni
on 8 Feb 2011 at 12:26
I started this on revision 70f7c8d1f0. I'm not sure yet how to make the UI list
all the available formats for download.
Original comment by andresgattinoni
on 9 Feb 2011 at 1:20
Basically, in parseBranch we parse an Atom feed (it's actually OPDS) and split
it into a list of books and a list of links.
Then those two are passed to a template (category.tmpl) which renders a nice
HTML page from that data.
You may want to add stuff into the objects in the book list, or pass another
data structure with the extra links and then use the template to display them
on the right place.
Original comment by roberto.alsina
on 9 Feb 2011 at 1:30
This issue was updated by revision 746106065d.
I moved forward a little with this. I'm still having random problems to connect
to ManyBooks.net, so I added a message for the user.
Now that ManyBooks provides a long list of available formats, we should
probably change the way the links are displayed.
Maybe we could use a <select> with a Download button. I need to see how to do
this on webkit.
Original comment by andresgattinoni
on 10 Feb 2011 at 3:05
Also I don't know how to avoid the plugin from caching the empty result it gets
when it can't connect to ManyBooks.net, so that I can try to reload it without
restarting Aranduka.
Original comment by andresgattinoni
on 10 Feb 2011 at 3:10
I am having problems with excessive caching, too, for example, on the epub
viewer.
I will ask around.
Original comment by roberto.alsina
on 10 Feb 2011 at 3:38
This issue was updated by revision 85255342f0.
Well, I'm not sure why this doesn't work. The javascript code for the
redirection works, if I put
"window.location.href='http://www.google.com'", it takes me to Google. If I
make an alert of
o.options[o.selectedIndex].value I get the right URL. But I'm not being
redirected nor the book is being
downloaded. Any ideas?
Original comment by andresgattinoni
on 15 Feb 2011 at 2:27
Apparently the JavaScript redirection doesn't trigger the url parser function.
You can try it by adding an html link to, say google.com, and in console you
get a debug message saying that is following the link, but if you set the
JavaScript redirect to google instead of the book url you don't get that debug
message. I'm investigating how to fix this..
Original comment by algoz...@gmail.com
on 13 Apr 2011 at 2:43
I don't have much idea of how the webkit works, but could it have something to
do with the connect/disconnect of the parseBranch slot?
I noticed that one of the first thing that parseBranch does is to call
disconnect.
Original comment by andresgattinoni
on 13 Apr 2011 at 2:48
I think that the problem is that the javascript "location.href =
'http://blabla.com'" doesn't trigger the linkClicked method of the QWebView,
because, well technically it isn't a click.. OTOH, it doesn't even trigger the
urlChanged() method! at least with the current code..
Original comment by algoz...@gmail.com
on 13 Apr 2011 at 3:05
Hmm... is it possible to access QWebView events from Javascript?
If not, what about something like this?
http://stackoverflow.com/questions/943554/how-to-get-javascript-in-a-qwebview-to
-create-new-instances-of-c-based-classes
Maybe that way we can create something from Qt side which does the redirection
and tell javascript to call a Qt method.
Original comment by andresgattinoni
on 13 Apr 2011 at 3:14
Yep, may be that's the way to do it.. I was taking a look at the
addToJavaScriptWindowObject() method of QWebFrame, that could be another
solution, it exports objects to the javascript window object. But I couldn't
make it work yet. Roberto might have a better idea of a solution, I know zero
JavaScript and Aranduka was my first time with WebKit.. :-) Going to bed now..
tomorrow I'll give it another shot.
Original comment by algoz...@gmail.com
on 13 Apr 2011 at 3:53
Good news fellows, Revision d34a7a31f7 Fixes the JavaScript problem. It
downloads epubs but I get Exceptions with other formats, the openUrl function
needs some love... :-)
Original comment by algoz...@gmail.com
on 15 Apr 2011 at 3:42
This issue was updated by revision f6121efa6202.
I've made a few changes. I think this is ready to be tested and merged to the
integrate branch.
Can any of you guys check it out?
Original comment by andresgattinoni
on 19 Jan 2012 at 3:01
Andy,
I've downloaded 2 books from Manybooks.net in several formats and found no problem! :-) I say merge!
PS: By the way.. Aranduka looks great on Gnome3 ;-)
Original comment by algoz...@gmail.com
on 19 Jan 2012 at 11:05
Cool :)
I'll merge it right now.
Original comment by andresgattinoni
on 19 Jan 2012 at 11:21
This issue was closed by revision 4f368fd4e1d3.
Original comment by andresgattinoni
on 19 Jan 2012 at 11:23
[deleted comment]
Original comment by andresgattinoni
on 29 Jan 2012 at 2:41
Original issue reported on code.google.com by
andresgattinoni
on 7 Feb 2011 at 1:18