+ Split up browser window and content related elements of html & text search.
+ Associate search contexts with contents, rather than browser windows.
+ Remove highlighting of search terms dependency on current_redraw_browser.
+ Fixes issues with search contexts being applied to the wrong content.
svn path=/trunk/netsurf/; revision=12646
handlers.
+ Disentangle all box tree manipulation from browser window
code and put it where it belongs.
+ Move other content specific and other irrelevant code from
browser window handling to appropriate places.
+ Put mouse state enum in new mouse header, since it's not just
used by browser window code, and it is used by treeview
windows on the treeview branch.
svn path=/trunk/netsurf/; revision=10561
Remaining work:
1) Clone content_html_data
2) Cloning content_css_data requires the charset of the old content
3) Calling hlcache_handle_abort() before a content has been created must clean up the retrieval context.
svn path=/trunk/netsurf/; revision=10236
It is now defined as converting the content into a state in which it is ready for use.
The user of the content is now responsible for performing an initial reformat (sic) of the content before it can be redrawn.
Purge width/height parameters from hlcache_handle_retrieve/content_convert/*_convert APIs.
Fix up content handlers affected by the above change in semantics.
Ensure that browser_window_callback performs an initial reformat of its content.
svn path=/trunk/netsurf/; revision=10207
Extend text selection, copying, saving and searching code to handle textplain contents; modified textplain code to accept other line terminators
svn path=/import/netsurf/; revision=2081
