OpenOffice.org 2.0 overlay archive

TheCoop · Posted: Fri Jul 16, 2004 8:41 am Post subject:

yup. added at http://thecoop.dyndns.org/openoffice-cvs-20040709.ebuild (same as last time)
_________________
95% of all computer errors occur between chair and keyboard (TM)

"One World, One web, One program" - Microsoft Promo ad.
"Ein Volk, Ein Reich, Ein Führer" - Adolf Hitler

Change the world - move a rock

geki · Posted: Fri Jul 16, 2004 11:45 am Post subject:

Building project sysui will fail

fix: strip off suse, redhat, solaris and win32 related stuff from /var/tmp/portage/openoffice-cvs-20040709/work/OpenOffice/sysui/prj/build.lst

1 hour compiling right now

robmoss · Posted: Fri Jul 16, 2004 1:42 pm Post subject:

That should remove the RPM dep, right?
_________________
Reality is for those who can't face Science Fiction.

emerge -U will kill your Gentoo
ecatmur, Lord of Portage Bash Scripts

geki · Posted: Fri Jul 16, 2004 3:19 pm Post subject:

i think so

but GCC 3.4 cvs or snapshot >20040709 needed for OOo 2.0

so 3.4.1-r1 is too old (needed patches not implemented...)

Build project wizards fails... no idea why... binary code errors

geki · Posted: Sat Jul 17, 2004 9:21 am Post subject:

patched my GCC 3.4.1-r1 version also with enum patch

OOo Devs added module i18nutil into branch cws_680_gcc340fixes01

atm compiling OOo again to see if that error belongs to the enum issue...

geki · Posted: Sat Jul 17, 2004 9:42 am Post subject:

new ebuild will be up soon!

same servers new ebuild name: openoffice-cvs-20040717.ebuild

you have to update your GCC 3.4.1-r1 with the enum patch!

have fun...

new:

no rpm dependency

enum patch for gcc necessary!
Gentoo branding !!!

no more i18nutil fetch restriction... its in branch finally

i think thats it!

I am so happy jejejejejejejeje ... have a nice WE! :twisted:

geki · Posted: Sat Jul 17, 2004 11:45 am Post subject:

apply these patches:

http://www.openoffice.org/nonav/issues/showattachment.cgi/16467/connectivity.gcc34.patch

robmoss · Posted: Sat Jul 17, 2004 3:34 pm Post subject:

Does it actually work yet, or are you just "getting somewhere"? It appears to be a little more than unwell on AMD64... :cry:

_________________
Reality is for those who can't face Science Fiction.

emerge -U will kill your Gentoo
ecatmur, Lord of Portage Bash Scripts

geki · Posted: Sat Jul 17, 2004 3:57 pm Post subject:

http://www.openoffice.org/issues/show_bug.cgi?id=31640

it compiles for 1h and 30mins...

a small bug there but no idea how to fix... ill get into it now

oh yeah it will be fine soon :twisted:

i am twisted

how far are you and Lv with new GCC 3.4.1 ? that enum patch fixed another issue here!

TheCoop · Posted: Sat Jul 17, 2004 4:24 pm Post subject:

new version: http://thecoop.dyndns.org/openoffice-cvs-20040717.ebuild

edited: tar.bz2 version: http://thecoop.dyndns.org/openoffice-cvs.tar.bz2
_________________
95% of all computer errors occur between chair and keyboard (TM)

"One World, One web, One program" - Microsoft Promo ad.
"Ein Volk, Ein Reich, Ein Führer" - Adolf Hitler

Change the world - move a rock

geki · Posted: Sat Jul 17, 2004 4:25 pm Post subject:

not newest but if u follow thread its easy to fix things

edit #1
TheCoop you got mail with OOo sources patch within and GCC patch warning :twisted:

geki · Posted: Sun Jul 18, 2004 3:42 pm Post subject:

for any brave guy on AMD64 or IA64 out there you can test

cws_src680_ooo64bit02 (may be this helps, rob)

as branch in ebuild... its work in progress... :twisted:

cws_src680_gcc340fixes01 - this branch is closed so no cvs update necessary atm. although it needs fixing

robmoss · Posted: Sun Jul 18, 2004 9:19 pm Post subject:

geki, you star, I'm going to have a go at this... tonight!

_________________
Reality is for those who can't face Science Fiction.

emerge -U will kill your Gentoo
ecatmur, Lord of Portage Bash Scripts

geki · Posted: Sun Jul 18, 2004 9:27 pm Post subject:

well, star... i don't know how to handle that bug - that undefined variable... no idea how to set or change... i give it one more try...

the function sn_display_new is in libsn/sn-common.h. i included that file, too. there where sn.h got included. if that does not help. i have to wait for OOo Devs to fix that if they do.

or i should lower the cflags like the ebuild warning says hehehe

...lets see

could be cool if any C++ Professional helped here

robmoss · Posted: Sun Jul 18, 2004 9:34 pm Post subject:

geki · Posted: Sun Jul 18, 2004 9:42 pm Post subject:

thebell fixed a bug in licq some time ago

is licq c++? hehehe

well, my next o'reilly book will be about C++

edit #1
on IZ of OOo the gcc 3.4 fixes were mainly simple like:

before: variable / function
after: this->variable / this->function

or

before: function
after: ::class::function

but mine is slightly different

geki · Posted: Mon Jul 19, 2004 8:50 am Post subject:

i think i found a proper patch for that saldisp.cxx file...

compiling atm ^^

edit #1
patch works! ill add on OOo issue...

geki · Posted: Mon Jul 19, 2004 11:52 am Post subject:

better use --disable-mozilla on configure...

i added patch for mozilla errors but i wont use it anymore to test

atm 2h 10mins compiling

let's see how far this time without mozilla

sure less time hehehe...

geki · Posted: Mon Jul 19, 2004 3:48 pm Post subject:

# genlop -c

* app-office/openoffice-cvs-20040717

current merge time: 3 hours, 26 minutes and 53 seconds.
ETA: unknown.

and 5~10 mins more and it fails... but the half is done

its not because of OOo sources!

/usr/include/pthread.h is faulty... may be a typo...

line 427: __jmp_buf -> __jmp_buf_tag ... ill check it out

ill be back in ~3h 30mins. :twisted:

edit #1
that file belongs to glibc... who is interested

anyone calls Devs? to let them check?

geki · Posted: Mon Jul 19, 2004 7:34 pm Post subject:

well, something is borked there. i will file an issue...

merged bison-1.875d (alpha version) :roll:

and i try to use two processes for multiprocessing

lets see how far it gets this time... and may be bison helps with that thread error... don't think so but who knows? anyone? no... owww

rob, whats up with amd64 and that 64bit branch - how far do you get compiling? :twisted:

robmoss · Posted: Wed Jul 21, 2004 12:42 am Post subject:

geki · Posted: Wed Jul 21, 2004 6:16 am Post subject:

last time i updated gcc34 branch it deleted some files of bridges module and now it breaks there telling bridges module needs to be rebuilt

I will talk to Devs whats up with that...

It is possible to check 2 CVS branches out in one ebuild?! How that?

edit #1
well, that OOo cvs does not do this. If you want to check out 2 branches do one manually... but to warn you... bridges module is f***ed in gcc34 branch

if you have an old gcc34 branch checkout take it and tarball the bridges dir for me cuz mine is f***ed

...and please upload it somewhere so i can get it ...

geki · Posted: Wed Jul 21, 2004 9:37 am Post subject:

got hopingly a new old bridges module of one Dev that will work

edit #1
also basic module error... got tarball of the dev

lets see where it will stuck next time...

geki · Posted: Wed Jul 21, 2004 4:21 pm Post subject:

geki · Posted: Wed Jul 21, 2004 9:07 pm Post subject:

Rob, for your case... i would get the gcc34 branch as base and then gettin the 64bit branch manually and patchin the gcc34 branch with the differences case-based