discussion re: libgoffice scheduling

[Josh Sled]

I wanted to relay the following conversation I had, earlier, with Jody
f/ Gnumeric regarding the split-out _from_ Gnumeric of what's being
called libgoffice.  It'll include stuff like toolbar
foreground/background-color selection, font-selection, &c... all sorts
of standard-office-app "stuff".  The most important part for gnucash is,
specifically, the GnomeOffice Graphing [GOG] stuff...

In summary: it looks like it'll be a while [~weeks] before the relevant
code is sufficiently split out of gnumeric to be even copied into the
gnucash source tree, and a while after that before the library is
released, let alone started to be included in distros; he's targeting a
6/25 "release" of libgoffice.


<jsled> jody: question re: libgoffice/gog.
<jsled> Any 1-line status update on it?  And what's the best way to monitor it's progress so I don't need to bug you.
<jsled> s/.$/?/
<jody> jsled: progress is being made
<jody> I've got GODocument and GODocumentControl in my tree now and have ported gnumeric to that
<jody> that gives me a place to hang to plugins and importers/exporters
<jody> target is to make the split for 1.4
<jody> jsled: it will include show/hide toolbar and fullscreen support to start with
<jsled> jody: followup question about libgoffice.  It sounds like there's a bunch of stuff in it that gnucash doesn't care about at the moment.
<jsled> Frankly, we're probably going to copy the gog sources into the gnucash tree for the initial G2 port/release, anyways.
<jody> jsled: there will be some things in there that it may not care about
<jody> jsled: I'd ask that you not do that
<jody> I'm happy to work with you to ensure we release on a schedule that is convenient
<jsled> hmm.  you can ask, but we're pretty serious about not depending on stuff that's hasn't been in distributions for 6-9 months.
<jsled> We're currently talking about making FC1 our base/ref. platform.
<jsled> Or _maybe_ making it FC2.
<jsled> regardless, how long before libgoffice is going to be in out in the field?
<jody> jsled: before june 25
<jody> aka guadec
<jsled> Sorry ... a bit of a rhetorical question.   The point being that there's no realistic alternative for our needs besides GOG, but we're hoping to get the g2 port out the door before ...
<jsled> 2004.06.25 + 6m => 2004.12.25.
<jsled> Now.  I'm not sure that we _will_ do that, but we can only hope for the best. :)
* jody scratches head
<jody> At this point the code is not seperable from gnumeric
<jody> It still uses the plugins and format/parsing engine
<jody> both need to move down
* jsled nods
<jody> It is also based on gtk-2.4
* jsled nods
* jody apologizes for that but there are not alot of choices
<jsled> Well, so are we at this point... we've moved a good chunk of 2.4 into our source tree as well. :/
<jsled> we'd already had libegg in there, and I've moved a bit more which has showed up in gtk-2.4.0...
<jody> You're going to mix 2.4 code with 2.2 libraries ?
<jsled> yes.
<jody> brave
<jsled> heh.
<jsled> What's your primary concern with copying the gog code into the gnc tree?
<jsled> general in-elegance, or...?
<jody> Limited developer resources
<jsled> eh?
<jody> bugs and api requests will definitely occur for what is basicly the first major user of the code outside of gnumeric
<jody> having those appear against an out of date version
<jsled> hmm.
<jody> or worse, get fixed there and potentially not get upstreamed ...
<jsled> yes.
<jsled> point well taken; we will take care to find a way to prevent that.
<jody> Where is your cvs hosted ?
<jsled> cvs.gnucash.org -- private box in MA.
<jody> If it's on gnome.org you could do a virtual module include
<jody> svu: I'm nost sure what the question is ?
<jsled> a solution might be for the developers / early testers to actually develop against libgoffice ... but then cut-off a version when we're nearing the end of our dev/release cycle.
<jsled> there will undoubtedly be bugs after that, but we'd be a better position to manage the patches then.
<jsled> In any case, we'll have to deal with that at some point in the future... but will keep it in mind.
<jsled> jody: permission to copy this conversation to the other primary GnuCash developers?
<jody> jsled: sure, and I'll extend an invitation to get cvs hosted on gnome.org
<jody> which would simplify importing things easily 


...jsled

-- 
http://www.asynchronous.org/ - `a=jsled; b=asynchronous.org; echo ${a}@${b}`
# A: Because it breaks the flow of normal conversation.
# Q: Why don't we put the response before the request?