summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Fix bug 2389 - indicate that the install dir will depend on the python being ↵William Deegan2010-01-111-2/+2
| | | | used
* Add the package that contains epstopdf, which is now used in testing.Steven Knight2010-01-091-0/+1
|
* Python 1.5 fix (no continue within a try:-except: block within a loop).Steven Knight2010-01-091-2/+10
|
* I got the order of the arguments wrong in a Clean() call I added to help fix tehRobert Managan2010-01-071-1/+1
| | | | auxiliary file problem.
* Issue 2505: fix use of pre-compiled headers when the source .cppSteven Knight2010-01-062-2/+121
| | | | file is listed in both the env.PCH() and env.Program() calls.
* fix typos in the man pageGreg Noel2010-01-061-7/+14
|
* Fix some tests so they can still run on Windows system withoutSteven Knight2010-01-063-54/+58
| | | | .py file associations.
* Fix path name checks for case-insensitive file systems.Steven Knight2010-01-061-5/+8
|
* Fix handling the -C option by recording the absolute path toSteven Knight2010-01-051-11/+6
| | | | the original script directory.
* Change scanner to properly search for included file from theRobert Managan2010-01-045-8/+400
| | | | | | | | directory of the main file instead of the file it is included from. Also update the emitter to add the .aux file associated with \include{filename} commands. This makes sure the required directories if any are created for variantdir cases. Half of the patch from Stefan Hepp.
* Fix time-sensitivity (< 0.002 second execution) in output format.Steven Knight2010-01-021-6/+6
|
* Remove unnecessary imports to avoid javac warnings.Steven Knight2010-01-011-13/+5
| | | | Add a necessary serialVersionUID.
* Define "static final long serialVersionUID" in java classes inSteven Knight2010-01-012-0/+16
| | | | RMIC tests to silence warnings from gcj-4.4.
* Avoid GCC 4.4 warnings from the code generated by our dummy QTSteven Knight2009-12-312-79/+9
| | | | moc and uic utilities by using fputs() instead of printf().
* Adding 1.2.0.d20091224 releaseWilliam Deegan2009-12-292-3/+3
|
* Pull up the tigris.org page for SVN revisions, at least for now, inSteven Knight2009-12-271-1/+8
| | | | response to clicks on the detailed graph.
* Interface tweaks:Steven Knight2009-12-262-15/+15
| | | | | | Display 150 revs in the dashboard thumbnails, but full history in the click-through graphs. Rewrite the top-of-page title and have it display which stat is being displayed.
* Add a -p option to support calibrating against build SCons packages.Steven Knight2009-12-221-2/+7
|
* Fix the view-change detail tab. Minor wording improvement.Steven Knight2009-12-221-2/+2
|
* Don't look for copyright strings in buildbot timings files.Steven Knight2009-12-221-0/+6
|
* Issue 2482: Avoid wrapping decider unnecessarilyGreg Noel2009-12-201-1/+2
|
* Add a message to the UserError raised when trying to do a Dir lookup with ↵Gary Oberbrunner2009-12-201-1/+2
| | | | create=False. Should not normally happen, but it did to me, and it is better to have some explanatory message when it happens rather than failing silently.
* Update the number of the "base" revision of bin/time-scons.py to useSteven Knight2009-12-191-2/+2
| | | | | for historical statistics. Close the triple-quote in the doc string I added right before the last checkin.
* Speed up the time-scons.py "build" of SCons with compiled *.pyc files,Steven Knight2009-12-192-24/+63
| | | | | | | | | | | | | by just invoking setup.py directly instead of packing and unpacking the distributable .tar.gz file. Refactor to be able to capture historical statistics (of previous revisions) by preparing the "built" revision once and then running all of the requested timing scripts, with output going to a separate log file for each revision + script. Fix TestSCons.py so that it interprets the scons script name to be tested relative to the invoking directory, not always relative to src/script.
* Add config.js files to the timing configurations, and update graph.htmlSteven Knight2009-12-184-13/+18
| | | | | to use them to display configuration-specific info (right now just the title) for each graph's page.
* Move the timings-specific pieces of the buildbot infrastructure intoSteven Knight2009-12-186-0/+1368
| | | | | | | | | the trunk/timings directory. We'll map them into the buildbot directory using svn:externals. This will let us keep all the pieces of a timing configuration, including its buildbot pieces, in one place, and will let us simplify the Master initialization (since it will be able to look on-disk for the configurations for which it should set up buildbot steps, instead of querying the SVN server).
* Trace the system load average (Linux-only code) before starting theSteven Knight2009-12-181-0/+13
| | | | help / full / null builds in a timing run.
* Running PDF() on an .eps file on windows with MikTex 2.7.2767 does not work,Robert Managan2009-12-172-2/+6
| | | | | | | because it does not support '-o <outfile>'. This patch uses --outfile= instead, since this works both unter linux and windows. Patch from Stefan Hepp
* Add scanning of files included in Latex by means of \lstinputlisting{}Robert Managan2009-12-173-2/+138
| | | | Patch from Stefan Hepp.
* Avoids bailing on non-zero exit status from commands. This will letsSteven Knight2009-12-171-32/+44
| | | | | | | | | | | us collect as many stats as possible, even when a test for a revision in the middle of the list fails due to a bug. Updates the "baseline" timing revision (the revision that has the earliest "stable" TimeSCons infrastructure Removes --noqmtest, since the "baseline" revision no longer uses QMTest by default. Changes the --logfiles option to a --logsdir option. Updates the usage message.
* Ignore the exit status from the --help run on timing builds, since someSteven Knight2009-12-171-0/+4
| | | | | early revisions have bugs and we can still go on and get useful stats from the full and null builds.
* Add a separate test of the runtest.py --qmtest option.Steven Knight2009-12-171-0/+105
|
* Fix the time-scons.py --logfiles option.Steven Knight2009-12-171-9/+4
|
* Fix the tests of runtest.py now that QMTest is no longer being usedSteven Knight2009-12-1616-426/+258
| | | | | | | by default. Fix use of subprocess in Python 2.4+ and exit statuses of popen'ed scripts in earlier versions of Python. Support the ability to execute a directory's tests by naming the directory as a command- line argument.
* Better fix for #2530: dont exit at all on unimplemented options, just print ↵Gary Oberbrunner2009-12-162-3/+0
| | | | a warning and keep going.
* ...and call the correct method.Steven Knight2009-12-151-1/+1
|
* Typo fix.Steven Knight2009-12-151-1/+1
|
* Disable QMTest by default, leave it available when --qmtest is used.Steven Knight2009-12-151-13/+14
|
* Add a bin/time-scons.py script to manage invocation of runtest.py toSteven Knight2009-12-151-0/+311
| | | | | | | | collect SCons timings. The script will build SCons and time the pre-compiled *.pyc files, and also provides options for timing a range of revisions directly from SVN.
* Add a TestSCons.option_not_yet_implemented() method and use it soSteven Knight2009-12-1513-54/+40
| | | | the tests of those options expect an exit status of 1.
* Integrated patch for bug 2530: SCons should not exit with 0 (success) on ↵Gary Oberbrunner2009-12-151-1/+1
| | | | unimplemented option error.
* Provide an environment variable that can be used to overrideSteven Knight2009-12-101-1/+7
| | | | | | setting --warn=no-visual-c-missing when trying to run the new timing-script infrastructure against older versions of SCons that don't have that warning.
* Add separate memory graphs for the stats in the full and null runs.Steven Knight2009-12-101-36/+67
| | | | | Refactor stats-gathering so we collect them from the output in one step and can then trace them in multiple different ways.
* Record memory in kbytest, not bytes.Steven Knight2009-12-101-8/+17
|
* Add a script for calibrating settings for timing configurations.Steven Knight2009-12-105-22/+142
| | | | | Update the timings scripts with calibrated settings that run a full build between 9.5 and 10.0 seconds on the buildbot slave.
* Infrastructure to support calibration runs of TimeSCons tests, whichSteven Knight2009-12-104-19/+57
| | | | | only run a full build and report back the values of any variables affecting the configuration, and the elapsed time of the full build.
* Update the TimeSCons class for use of the sort= keyword instead ofSteven Knight2009-12-091-7/+23
| | | | | important=. Fix an old-version Python problem (use of True in keyword arg initialization).
* Have the TimeSCons class examine the run output for the reported memorySteven Knight2009-12-091-3/+45
| | | | | | use and timings, and translate these into TRACE: lines with keyword=value pairs that will be processed by the now more generic buildbot-side log processor to populate the timings graph data.
* Update the TimeSCons "elapsed time" message to match the log processor.Steven Knight2009-12-081-3/+3
|
* Fix typo in man page, per patch in issue #2529.Gary Oberbrunner2009-12-061-1/+1
|
generated by cgit v0.12 at 2024-08-31 07:09:09 (GMT)