Ah, that's good news. Happens to me sometimes as well when trying to
translate C docs to Python :) Sometimes Python's help can be your friend
"help(WebKit.DOMDocument.evaluate)"
On Tue, Jul 2, 2013 at 5:12 PM, Niranjan Rao <nhr...@gmail.com> wrote:
> Simon,
>
> Thanks for the pointer. I noticed that I made a stupid mistake and was
> sending document as a first parameter.
>
> Your example opened my eyes and now it looks like its running ok.
>
> Thanks,
>
> Niranjan
>
>
> On 07/02/2013 02:37 PM, Simon Feltman wrote:
>
>
> On Tue, Jul 2, 2013 at 10:35 AM, Niranjan Rao <nhr...@gmail.com> wrote:
>>
>> And yes, I had tried creating empty WebKit.DOMXPathResult object and
>> passing it. Did not work.
>>
>
> It would be helpful to know the details of what didn't work. Code
> examples of what you have tried, or steps to reproduce the problems you are
> seeing can be invaluable for getting help. A simple example on my system
> shows using an empty DOMXPathResult is actually working:
>
> In [1]: from gi.repository import WebKit
> In [2]: view = WebKit.WebView()
> In [3]: dom = view.get_dom_document()
> In [4]: dom.evaluate('/', dom, WebKit.DOMXPathNSResolver(), 0,
> WebKit.DOMXPathResult())
> Out[4]: <DOMXPathResult object at 0x7fc670b08eb0 (WebKitDOMXPathResult
> at 0x2ea90f0)>
>
> This is with PyGObject/GTK+ version 3.8.2 (not sure about WebKit
> version). If this does not work with older versions, supporting a fallback
> with ctypes hacks sounds reasonable. This would allow support for older
> versions while ensuring use of proper API on newer versions. If the
> __gpointer__ attribute is not available or not useful, you can still find
> the pointer to the GObject by using offsets of the pointer field on the
> PyGObject wrapper. But code examples of what you have already tried will be
> needed to further aid you with this.
>
> -Simon
>
>
>
>
_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@gnome.org
https://mail.gnome.org/mailman/listinfo/gtk-app-devel-list
