On 17/07/2013 James Hogarth wrote:
> Apache OpenOffice 4.0, due in the last decade of July 2013 ...
This should probably be fixed to due at the end of July 2013 or
No problem. OpenOffice 4 is already on mirrors and will be announced in
a couple of days, so I've put the exact release date on the wiki page,
thus revealing a very badly kept secret...
> The OpenOffice sources have been updated to allow a clean build
> the default tools shipped with Fedora 19.
Has it been built cl cleanly in rawhide and koji yet?
Not yet, only on F19 so far. The final source package was approved 2
> The /usr/bin/soffice and /usr/bin/unopkg executables/symlinks
are still a
> problem since (in the Fedora packages) they would conflict between
> LibreOffice and Apache OpenOffice: it is recommended to fix it in
> the LibreOffice packages too.
Have the LO maintainers agreed on this yet?
I don't know more than you (or the rest of this mailing list audience)
> upstream LibreOffice packages do not rely on a
As I recall the concern ended up not being LO itself but third party
scripts relying on soffice and breaking those... In addition if
behaviour of the two 'owners' of soffice changes when it is called then
what should the accepted consequence be?
This is a valid concern. I would very much like to see two or three of
these third-party scripts, but it doesn't seem much different than
having third-party scripts relying on the possibility to invoke
"sendmail" (or awk, of whatever) with some options and getting the
expected behavior. It would probably help here to have two additional
explicit aliases, like, say, "/usr/sbin/libreoffice" and
"/usr/sbin/openoffice", that those third-party providers wishing to use
one particular program can call.
Personally I think the LO maintainers need to weigh in and if there
any concerns from that direction push back to F21 with work done on
rawhide after branch...
Pushing back when a software does not exist yet is just reasonable.
Pushing back when a software does exist is a little less reasonable,
since the conflict is probably (and I welcome the opinion of the
LibreOffice maintainers too) not big enough to justify, by itself, a