Minutes
RM
Mac - nothing new for tiger token issue
Windows - RM parsing output EOL issue due to difference in how unixes and windows specific eol 
which causes RM to miss build failures. Navid has a fix, first he modified the build system for RM2 to use Qt's build system, rather than SCons, as the SCons was becoming unwieldy. Navid has committed the necessary changes and can now get back to implementing the EOL fix. Why Qmake rather than SCons? Well, RM2 (as well as GoGui which also uses Qmake), is built on top of Qt. Qmake takes care of Qt's intermediate moc files automatically. Qmake is very basic, does less than make in fact. It can be set up quickly with few lines of Qmake code, but it would not work for GR or ST, since it is not configurable.
Preparing for a SCons release of ST for the upcoming reprocessing
Richard had asked about making a SCons release of ST. Of course RM2 already performs the builds automatically for LATEST, HEADs or tags. There was a recent fix to astro, which Navid believes is already in ST HEAD. There is a new problem for pylikelihood, where the swig compile fails due to the *.in and *.i files being out of synch. Eric Winter had reported the problem and Navid has issued him a fix, if it works, Navid will contact Jim Chiang to get the patch put into CVS. The remaining issue is getting a Mac build produced. Due to the outstanding token issue, Navid will perform the Mac build by hand. Navid requested that we attempt this on an upcoming ST HEAD before a tag is in place, so we can test out the Mac build and patch if necessary. Heather will ask Jim to give us a heads up before he makes a new ST tag.
Navid wants to encourage people to look at the SCons builds so that developers will become more proactive in fixing problems. The automated notifications are not really working so far. Heather asked if build failures could be noted on the CMT RM page, to perhaps attract more attention. Navid will look into the feasibility of that. Joanne suggested asking people like Jim directly what would grab their attention.
RMviewer and Tag collector - No News
check with Karen about compile output
swig stuff is working now -- need more to get env setup for
something similar for rootcint like swig
then justabou about there
need more externals for 2008
CHS
several packages with inadequate sconscript files or none at all - or not tagged
the one Joanne knows about has been taged by Joanne
Joanne could ask Bryson when he's around, and Joanne could pursue if necessary
Fred and Fox
brute force debugging, with print statements, to see where things are hanging up
two particular ruby files -- one is FredMainWindowsrb hanging about not creating
wind before parent window
also noticed winxp is one of the parameters passed Fred.rb
Other news
installed python 262 for rhel4
installer tony
Agenda
- Release Manager Status
- Backend
- Windows failing to compile
- Automatically notify package owners of build failures -- done, some suggestions from Toby
- Mac - fixed!
- u35 now used by SCons -- how is that going
- General Status
rhel3, rhel4 32 and 64 bit, rhel5 32 bit, and mac
Where is each as far as the 3 stages of builds? - Other Issues?
- Backend
- rhel4 and rhel5
- Any Babar news?
- rh9 when does FSW plan to dump it?
Whenever we're ready - Installer/ RMviewer /Tag Collector - Navid
- user releases are good to go
- developer release status
- Version 0.5 current test release
- available for testing from ftp://ftp-glast.slac.stanford.edu/glast.u05/RMViewer/
- status of current RM binary installs?
- To Do List
- Installer for Developer Releases
- Contact Tony concerning web based installer
- Tag Collecting
- Front End - Karen
- Web Front End - Karen
- http://glast-ground.slac.stanford.edu/rm2/
current test version is: http://glast-tomcat03.slac.stanford.edu:8080/rm2/ - How to get checkout and compile output like RMViewer?
- Recent Updates
- Reformat time stamps
- http://glast-ground.slac.stanford.edu/rm2/
- Web Front End - Karen
...