[Webkit-unassigned] [Bug 228233] [GTK] Allow to set GProxyResolver directly

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Jul 27 04:40:48 PDT 2021


https://bugs.webkit.org/show_bug.cgi?id=228233

--- Comment #9 from Michael Catanzaro <mcatanzaro at gnome.org> ---
(In reply to Milan Crha from comment #7)
> I also do not know how the g_proxy_resolver_lookup() works in the
> background. If it can return different results for different
> http://|https:// requests, then it's impossible to convert into WebKit's
> proxy settings. 

Correct, it is impossible to convert an arbitrary GProxyResolver into WebKitNetworkProxySettings, because GProxyResolver can (and will!) return different results for different requests. E.g. a common configuration would be to use a special proxy only for a specific list of domains, and connect directly to everything else. This cannot be expressed with WebKitNetworkProxySettings (it's just a wrapper around GSimpleProxyResolver) and so that limits what you can do with your application's proxy configuration dialog relative to what an app that doesn't use WebKit can do. But the more complex configuration will still be supported just fine if configured using system settings, and in practice that's probably sufficient I think, especially given that almost nobody uses proxies anymore anyway.

> The g_proxy_resolver_lookup() also returns an array of the
> proxy server URIs, but the
> webkit_network_proxy_settings_add_proxy_for_scheme() seems to allow only a
> single proxy URI to be added for the scheme.

Right.

> As you can see, many things are due to my lack of knowledge. The Evolution's
> settings are not that complicated as the GProxyResolver API allows, thus I
> might be able to make this work with the current WebKit implementation.

Hopefully!

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20210727/1f252aab/attachment.htm>


More information about the webkit-unassigned mailing list