To comment on the following update, log in, then open the issue:
http://www.openoffice.org/issues/show_bug.cgi?id=84137


User mba changed the following:

                What    |Old value                 |New value
================================================================================
                      CC|'as,aziem,kr,kso,mhu,mmeek|'as,aziem,kr,kso,mba,mhu,m
                        |s,sb'                     |meeks,sb'
--------------------------------------------------------------------------------




------- Additional comments from m...@openoffice.org Tue Jan 27 15:02:45 +0000 
2009 -------
It seems we are stuck. So it's a good time to remember the requirements. If I
got it right they are

(1) Every application that uses gtk must initialize the gvfs UCP in its "main"
thread.
(2) The implementation of the gvfc UCP must not use VCL.
(3) The current workaround in appinit.cxx can have a performance impact and
should be removed.
(4) The current workaround in appinit.cxx is ugly as we would prefer to have a
solution that does not require code outside of the UCP to run it properly even
in gtk based applications.

While (1) and (2) seem to be clear, I'm unsure if (3) or (4) is the reason to
remove the current workaround. Can somebody enlighten me? Is (4) a critical
requirement or just a "nice to have" and would a solution that removes the
performance impact but still binds the usage of the gvfs-UCP to a precondition
that the application using it must fulfill?

If (4) is not a critical requirement we could encapsulate code that forces the
initialization into a UNO service that gtk based applications must provide if
they want to use the gvfs-UCP. The UCP then can try to instantiate the service
and in case it is not present assume that it's safe to initialize gfvs on 
demand.

If (4) is a critical requirement, I wonder whether thinking about moving the
gvfs-UCP into a separate process (not based on gtk) and use it remotely might be
an option. I don't know how much this could hurt the performance of operations
based on it and how critical *this* would be.


---------------------------------------------------------------------
Please do not reply to this automatically generated notification from
Issue Tracker. Please log onto the website and enter your comments.
http://qa.openoffice.org/issue_handling/project_issues.html#notification

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@ucb.openoffice.org
For additional commands, e-mail: issues-h...@ucb.openoffice.org


---------------------------------------------------------------------
To unsubscribe, e-mail: allbugs-unsubscr...@openoffice.org
For additional commands, e-mail: allbugs-h...@openoffice.org

Reply via email to