Commit Graph

72 Commits (4902a2752e6ee5dfd47be3d94955f88bf73bf5d8)

Author SHA1 Message Date
Aleric Inglewood c4acbaf7c4 SNOW-766: Add a 'printbuilddirs' command to develop.py 2010-08-27 17:32:41 +02:00
Aimee Linden 6baedc1370 VWR-20748 (SNOW-520) FIXED Make develop.py play nicely with express editions of Visual Studio 2010-08-18 12:30:26 +01:00
Aimee Linden 8aae5a105b Fix executable perms on develop.py which seem to have got lost somewhere in the move. 2010-08-16 13:58:33 +01:00
Oz Linden 06b0d72efa Change license from GPL to LGPL (version 2.1) 2010-08-13 07:24:57 -04:00
Nathan Wilcox d194d95391 DEV-44838 - Fix a logic flaw in the windows run() retry code which would re-execute a process if its exit status was 0. 2010-01-13 16:09:07 -08:00
Nathan Wilcox bee98d70e3 DEV-44838 - Add the string "DEV-44838" to assertion failures to future devs understand preconditions on windows run(). 2010-01-13 16:08:30 -08:00
Nathan Wilcox dd7840aa9e Add an assertion for my sanity. 2010-01-13 15:47:57 -08:00
Nathan Wilcox 2d337ca8c6 DEV-44838 - See if buildconsole is confused by having the absolute path (with spaces) in the commandline. To accomplish this we put relative paths in the command string but lookup the absolute path for the executable parameter. 2010-01-11 12:30:27 -08:00
Nathan Wilcox 33df069238 DEV-44838 - Fix a silly bug in yet another place: find_in_path returns a list, so select the first element. 2010-01-11 11:07:04 -08:00
Nathan Wilcox 11eac588c2 DEV-44838 - Fix another bug introduced by passing "executable" in WindowsSetup.run(); find_in_path() does not handle absolute paths, but name is guaranteed to be an absolute path.
This explains the bogus message in the log of the form "process $foo exited with nonzero status; process $foo is not found", described here:
https://jira.lindenlab.com/jira/browse/DEV-44838?focusedCommentId=328441&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#action_328441
2010-01-11 10:45:01 -08:00
Nathan Wilcox f37f806660 DEV-44838 - Fix a bug introduced by inconsistently changing the get_build_cmd() interface. 2010-01-11 10:38:13 -08:00
Nathan Wilcox 8246b382b7 DEV-44838 - find_in_path() returns a list of candidates. Select the first. 2010-01-08 18:00:32 -08:00
Nathan Wilcox 61bc236b35 DEV-44838 - Remove a typo line which would cause duplicate executions for every run call. 2010-01-08 17:58:56 -08:00
Nathan Wilcox 6b33d5955a DEV-44838 - See if the executable parameter to subprocess.Popen must be absolute on windows. 2010-01-08 17:42:22 -08:00
Nathan Wilcox 010bfed411 DEV-44838 - This is an attempt at the minimal possible change to fix the quoting bug on windows.
This commit is untested!  The simplest approach to testing is to push into a repo watched by parabuild.

Also, develop.py could be improved to use subprocess consistently across all platforms.  Doing so could simplify the code, but until I understand how to test it better, I'm going to leave it alone.
2010-01-08 15:50:09 -08:00
Nathan Wilcox 5d2167c66a Fix a typo bug caught by pyflakes. eys.exit(1) -> sys.exit(1) 2010-01-08 15:10:59 -08:00
CG Linden 7f335dbb8c Use "strict" invocation of buildconsole according to Xoreax 2009-12-24 11:46:25 -08:00
CG Linden 6b47e36b15 Add some debug output to develop.py 2009-11-15 11:39:00 -08:00
CG Linden 853faf5316 Invoke native python for develop.py, not CYGWIN python, revert develop.py's original usage 2009-11-15 09:41:07 -08:00
CG Linden da4f76c17a Set PATH to devenv.com to avoid problem with develop.py's use of os.system() 2009-11-15 09:22:00 -08:00
brad kittenbrink 2fe083d36a Minor develop.py tweaks to make mac logs more readable in parabuild output.
Supressing all the setenv spam.
2009-11-10 12:58:22 -05:00
brad kittenbrink bae94e4340 Merged login-api with viewer-2.0.0-3 up through r132290. 2009-09-01 12:21:51 -04:00
Christian Goetze cbad737f59 Fix syntax error introduced by your's truely into develop.py 2009-09-01 01:46:05 +00:00
Christian Goetze 01d957c3e4 Add retry logic on certain exit codes 2009-09-01 00:55:00 +00:00
brad kittenbrink d020dae7b9 Merged latest login-api with latest viewer-2.0.0-3 up through plugin-api-05 merge (r131929).
Some minor post-merge cleanups still required.
2009-08-27 16:25:02 -04:00
Monroe Williams 745845f799 svn merge -r 129841:129910 svn+ssh://svn.lindenlab.com/svn/linden/branches/moss/pluginapi_05-merge@129910
svn merge -r 129913:131718 svn+ssh://svn.lindenlab.com/svn/linden/branches/pluginapi/pluginapi_05

Some branch shenannigans in the pluginapi_05 branch caused this to become a two-part merge.
2009-08-27 19:00:18 +00:00
brad kittenbrink a7efd0d8dd Merged in viewer-2.0.0-3 changes up through r129050. Boost got clobbered so I'll need to rebuild that, plus a couple other minor clenaups. 2009-07-30 19:39:08 -07:00
Brad Kittenbrink be7e9a46db Merge in disabling of universal builds.
PPC is now oficially deprecated.

svn merge -r 127807:128098  svn+ssh://svn.lindenlab.com/svn/linden/branches/brad/killppc-1
2009-07-24 01:31:53 +00:00
brad kittenbrink ccba924dac 1st attempt at disabling ppc build on macs. 2009-07-16 11:20:16 -07:00
Bryan O'Sullivan 7aec119d84 Merge with SVN login-api-3 branch 2009-06-25 11:12:51 -07:00
Christian Goetze 7e84d9125e Fix Mac Build 2009-06-24 15:13:53 +00:00
Paul Oppenheim 351ebe9fcb result of "svn merge -r123757:124890 svn+ssh://svn.lindenlab.com/svn/linden/branches/improve-unit-testing/iface-04" for QAR-1496 2009-06-19 23:24:11 +00:00
Christian Goetze f9b9372027 Remove -parallelizeTargets from xcodebuild invocation, it seems to not work well enough 2009-06-03 23:18:22 +00:00
Christian Goetze 93cf3d89e5 Put the os.environ["DISTCC"] = ... call into main() 2009-05-21 21:25:09 +00:00
Christian Goetze 77c1e2b761 Fix develop.py putenv bug, use new etch build-linux hosts for parabuild. 2009-05-21 20:59:56 +00:00
Mark Palange 8c5ca38f1d Fix to develop.py to build on 64 bit windows.
Added March 2009 DirectX SDK to cmake config.
Reviewed by brad
2009-05-19 23:41:59 +00:00
Christian Goetze 0257214763 Set DISTCC_DIR to a location inside the source tree to avoid causing NFS trouble with the default setting. 2009-05-14 21:20:37 +00:00
Christian Goetze 65c86eadcb Max out distcc jobs to 12 in develop.py 2009-05-14 20:56:03 +00:00
Christian Goetze 32100098f5 There are 71 eniacs, not just 64 2009-05-14 20:26:00 +00:00
Christian Goetze 21a8d1f638 Make develop.py work on eniacs the same way it works on stations 2009-05-14 19:24:24 +00:00
Christian Goetze 4bcbf33422 svn merge -r118008:118374 svn+ssh://svn.lindenlab.com/svn/linden/branches/cg/viewer-etch-1
QAR-1459 etch builds enabled
2009-04-30 06:52:38 +00:00
Christian Goetze c7c34d9ad2 Working around DEV-30401 by yanking the etch 32bit build code until the viewer also builds on etch 2009-04-10 01:14:30 +00:00
Bryan O'Sullivan 1aca3ff6f5 Get 32-bit builds going on 64-bit Debian Etch.
Reviewed by Brad.
2009-04-09 18:05:56 +00:00
Bryan O'Sullivan 5476aca88e DEV-16956 - Print a more useful error message if cmake is not found
Reviewed by Poppy.
2009-01-09 22:28:17 +00:00
Aaron Brashears e3cf284388 Result of svn merge -r107256:107258 svn+ssh://svn/svn/user/phoenix/license_2009_merge into trunk. QAR-1165 2009-01-08 00:05:06 +00:00
Mark Palange fb793870fe QAR-1142 merging 1.22 RC0-RC4 changes.
svn merge -c 106471 svn+ssh://svn.lindenlab.com/svn/linden/qa/viewer_1-22-106055_merge
2008-12-23 19:39:58 +00:00
Adam Moss fd46865a50 QAR-1040 maint-viewer-11 + OpenAL combo mergeme
svn merge -c104451
svn+ssh://svn.lindenlab.com/svn/linden/branches/moss/openal-maint-viewer-11-combo-r104448
2008-12-02 20:35:40 +00:00
Mark Palange ac68eb16a2 Merging the changes in viewer_1-22-0 (1.22 RC0) back into trunk, revs. 101837 to 103519 2008-11-21 01:18:39 +00:00
James Cook 2a44e81675 QAR-1018 Lightweight Windows Setup App (windows-setup-3). svn merge -r102882:102883 svn+ssh://svn.lindenlab.com/svn/linden/branches/windows-setup/windows-setup-3-merge 2008-11-19 23:26:29 +00:00
Bryan O'Sullivan c4d32870e1 Fix heinous traceback if commands are not found 2008-10-31 18:23:00 +00:00