Re: should internal depencies be push into librpm dependencies ?

2007-06-09 Thread Olivier Thauvin
Le vendredi 08 juin 2007, Jeff Johnson a écrit : > On Jun 7, 2007, at 8:34 PM, Olivier Thauvin wrote: > > But then rpm package should provide it's rpmlib internal deps, and > > then this > > deps will push into index. [snip all bla bla :)] > rpmlib() dependencies were the first run-time probes. >

%clean replace ??

2007-06-09 Thread Olivier Thauvin
According comment and code in build/parseBuildInstallClean.c the %clean section is no longer use and replace by a direct "rm -fr %buildroot". I am not sure a so direct replacement is a good idea: + rm -rf /home/users/olivier/RPM/tmp//munin-root (quick translation from french) Can't chdir in foo,

Re: should internal depencies be push into librpm dependencies ?

2007-06-09 Thread Jeff Johnson
On Jun 9, 2007, at 4:43 AM, Olivier Thauvin wrote: Le vendredi 08 juin 2007, Jeff Johnson a écrit : On Jun 7, 2007, at 8:34 PM, Olivier Thauvin wrote: But then rpm package should provide it's rpmlib internal deps, and then this deps will push into index. [snip all bla bla :)] rpmlib() dep

Trouble moving rpm-4.4.7 to rpm-4.4.9 on AIX

2007-06-09 Thread Frank Fegert
Hi all, although AIX is not the primary platform for RPM, i'd like to move from rpm-4.4.7 to rpm-4.4.9. The first try went south: bash-3.2# rpm -q vim rpmdb: PANIC: fatal region error detected; run recovery error: db4 error(-30975) from dbenv->open: DB_RUNRECOVERY: Fatal error, run databas

Re: Trouble moving rpm-4.4.7 to rpm-4.4.9 on AIX

2007-06-09 Thread Jeff Johnson
On Jun 9, 2007, at 9:44 AM, Frank Fegert wrote: Hi all, although AIX is not the primary platform for RPM, i'd like to move from rpm-4.4.7 to rpm-4.4.9. The first try went south: bash-3.2# rpm -q vim rpmdb: PANIC: fatal region error detected; run recovery error: db4 error(-30975) from db

Re: [CVS] RPM: rpm-4_5: rpm/ CHANGES build.c rpm/build/ spec.c rpm/lib/ de...

2007-06-09 Thread Ralf S. Engelschall
On Sat, Jun 09, 2007, Jeff Johnson wrote: > [...] > - add D_(...) i18n marker for debug messages, with default disable. Jeff, one important question from my point of view (as someone who is interested in RPM 5.0 and not in RPM 4.5 as any backward compat is not required for me): as you are mak

Re: [CVS] RPM: rpm-4_5: rpm/ CHANGES build.c rpm/build/ spec.c rpm/lib/ de...

2007-06-09 Thread Jeff Johnson
On Jun 9, 2007, at 12:54 PM, Ralf S. Engelschall wrote: On Sat, Jun 09, 2007, Jeff Johnson wrote: [...] - add D_(...) i18n marker for debug messages, with default disable. Jeff, one important question from my point of view (as someone who is interested in RPM 5.0 and not in RPM 4.5 as

Re: Trouble moving rpm-4.4.7 to rpm-4.4.9 on AIX

2007-06-09 Thread Frank Fegert
Jeff Johnson wrote: > So rpm-4.4.7 works reliably and rpm-4.4.9 does not. At least as far as i used it. AIX is too quirky to completely rule out more surprises ;-) > There have been few changes. Berkeley DB is connected > to rpm in rpmdb/db3.c. > > I just ran >cvs diff -r rpm-4_4_7-release

Re: [CVS] RPM: rpm-4_5: rpm/ CHANGES build.c rpm/build/ spec.c rpm/lib/ de...

2007-06-09 Thread Jeff Johnson
On Jun 9, 2007, at 1:40 PM, Jeff Johnson wrote: On Jun 9, 2007, at 12:54 PM, Ralf S. Engelschall wrote: I'll clean sweep the changes from rpm-4_5 and have on HEAD by next weekend? I should supply more details and start setting some explicit guidelines. All changes on rpm-4_5 are -- in ge

Re: Trouble moving rpm-4.4.7 to rpm-4.4.9 on AIX

2007-06-09 Thread Jeff Johnson
On Jun 9, 2007, at 1:54 PM, Frank Fegert wrote: Jeff Johnson wrote: So rpm-4.4.7 works reliably and rpm-4.4.9 does not. At least as far as i used it. AIX is too quirky to completely rule out more surprises ;-) There have been few changes. Berkeley DB is connected to rpm in rpmdb/db3.c. I

Re: [CVS] RPM: rpm-4_5: rpm/ CHANGES build.c rpm/build/ spec.c rpm/lib/ de...

2007-06-09 Thread Jeff Johnson
On Jun 9, 2007, at 2:06 PM, Jeff Johnson wrote: I'm knocking out the i18n debugging, and the type-punning on the 4.5 branch, largely because I'd like to use the rpm-4_5 branch for 5.0 reference, and the changes are straightforward enough (but noisy) that I'd like to just bang them into rpm-4_5

Re: %clean replace ??

2007-06-09 Thread Michael Jennings
On Saturday, 09 June 2007, at 12:36:58 (+0200), Olivier Thauvin wrote: > According comment and code in build/parseBuildInstallClean.c the > %clean section is no longer use and replace by a direct "rm -fr > %buildroot". To be accurate, it's actually: %{?__spec_clean_body}%{!?__spec_clean_body:%{?b

Re: %clean replace ??

2007-06-09 Thread Olivier Thauvin
Le samedi 09 juin 2007, Michael Jennings a écrit : > On Saturday, 09 June 2007, at 12:36:58 (+0200), > > Olivier Thauvin wrote: > > According comment and code in build/parseBuildInstallClean.c the > > %clean section is no longer use and replace by a direct "rm -fr > > %buildroot". > > To be accurat

Re: %clean replace ??

2007-06-09 Thread Michael Jennings
On Saturday, 09 June 2007, at 20:58:48 (+0200), Olivier Thauvin wrote: > hum, code has since rpm 4.4.8 so. Check rpm-4_5 branch. :) > I know what is the problem, I added the chown myself. The chmod is one possible solution, but there are others. IMHO, the chmod could easily be moved into %inst

Re: %clean replace ??

2007-06-09 Thread Jeff Johnson
On Jun 9, 2007, at 2:58 PM, Olivier Thauvin wrote: This software is munin, and there are other application doing that. If the packaging can't be changed, then I'll resurrect everywhere compatible %clean behavior for Mandriva. The vast majority of spec files are not doing anything more t

Re: %clean replace ??

2007-06-09 Thread Olivier Thauvin
Le samedi 09 juin 2007, Michael Jennings a écrit : > On Saturday, 09 June 2007, at 20:58:48 (+0200), > > Olivier Thauvin wrote: > > hum, code has since rpm 4.4.8 so. BTW, read "code has change", friends said forgetting words is my gpg signature, others play to "nanarthon", aka finding the missing

Re: %clean replace ??

2007-06-09 Thread Olivier Thauvin
Le samedi 09 juin 2007, Jeff Johnson a écrit : > On Jun 9, 2007, at 2:58 PM, Olivier Thauvin wrote: > > This software is munin, and there are other application doing that. > > If the packaging can't be changed, then I'll resurrect everywhere > compatible > %clean behavior for Mandriva. The softwar