summaryrefslogtreecommitdiff
path: root/testcurl.sh
Commit message (Collapse)AuthorAgeFilesLines
* This is the old script for testing curl, now use tests/testcurl.pl instead.Daniel Stenberg2004-06-211-304/+0
| | | | It is more portable.
* make this script tell that this is the obsolete version, to make it possibleDaniel Stenberg2004-05-181-0/+1
| | | | to detect
* remove ares/aclocal.m4 before cvs update to prevent bad conflictsDaniel Stenberg2004-03-091-0/+5
|
* show curl --version output as wellDaniel Stenberg2004-03-031-0/+4
|
* display src/config.h as well after configure as runDaniel Stenberg2004-03-021-0/+3
|
* Ignore the dreaded aclocal warnings on underquoted definitions that theDaniel Stenberg2004-03-011-1/+2
| | | | recent autoconf annoyingly introduced.
* When this is verified to be a CVS tree, we remove the two generated sourceDaniel Stenberg2004-02-261-0/+4
| | | | | files from the source dir to make certain they're generated in the build process.
* The build logs were created in the "wrong" dir and thus never removedDaniel Stenberg2004-02-261-3/+7
| | | | after use, this is an attempt to fix this.
* Output $CC and $CFLAGS as well. A Tor Arntsen patch.Daniel Stenberg2004-02-251-0/+2
|
* argh, use single-quotes instead of double ones to make $Revision appearDaniel Stenberg2004-02-231-1/+1
|
* use the proper source header, and set the version string to this file'sDaniel Stenberg2004-02-231-1/+23
| | | | revision number
* cd back to the root path before removing the build dir, since some systemsDaniel Stenberg2004-02-231-0/+2
| | | | refuse to remove the dir otherwise!
* Tor Arntsen's tiny fix!Daniel Stenberg2004-02-191-1/+1
|
* use a die and a log function to die and log textsDaniel Stenberg2004-02-191-54/+64
| | | | | | | | better detect test suite failures remove the buildlog at exit make a random buildlog file name, now in the same dir where the build dir is created checks if the ares build succeeded
* remove the build.log too at exit, and also use the proper $pwd prefixDaniel Stenberg2004-02-191-2/+6
| | | | | to find the files/dirs to remove so that it still works if we "die" after having done a 'cd'
* spell!Daniel Stenberg2004-02-151-1/+1
|
* Tor Arntsen made the ares build warnings etc get included as wellDaniel Stenberg2004-02-131-1/+1
|
* removed the state file renaming I accidentally left thereDaniel Stenberg2004-02-091-5/+0
|
* Remove the attempt to detect if we already tested the same source setup.Daniel Stenberg2004-02-061-16/+0
| | | | | We really don't care, and so many other things could've changed to make the new test interesting anyway.
* hide the grep resultDaniel Stenberg2004-01-301-1/+1
|
* verify that buildconf ran fine to allow it to continueDaniel Stenberg2004-01-291-1/+9
|
* Tor Arntsen fixed a spelling errorDaniel Stenberg2004-01-051-1/+1
|
* remove the build directory on exist, if there is anyDaniel Stenberg2003-12-151-0/+4
|
* only build ares if truly enabledDaniel Stenberg2003-12-121-1/+1
|
* ares-using configure should run configure in the subdir by itselfDaniel Stenberg2003-12-111-7/+1
|
* Ralph Mitchell's update that allows this script to take an existingDaniel Stenberg2003-11-071-64/+89
| | | | | | directory name and build/run/test curl in there instead of trying to update from CVS. Using this approach, this script can be used to test daily tarballs etc.
* support building ares-enabled tooDaniel Stenberg2003-10-241-0/+18
|
* the test compared numericly if though it could contain a string, and IDaniel Stenberg2003-07-231-2/+2
| | | | lowered the number of retries to 10
* made it work ;-)Daniel Stenberg2003-05-271-1/+1
|
* if cvs update fails, attempt again after 5 seconds and retry 50 timesDaniel Stenberg2003-05-271-3/+23
| | | | before giving up
* if the cvs update fails, don't continue furtherDaniel Stenberg2003-03-201-0/+5
|
* output what cvs returned, see if we can make the script bail out whenDaniel Stenberg2003-03-131-0/+3
| | | | cvs update fails
* This script clearly misses to remove the build dir at times when it exits,Daniel Stenberg2003-02-171-2/+2
| | | | so we now remove everything matching "build-*" when the script starts.
* fix the configure option queryDaniel Stenberg2003-01-231-6/+8
|
* reversed the actions on the cmp check for detecting if we're re-runningDaniel Stenberg2003-01-221-2/+2
| | | | | | a test on the same CVS setup as previous, as they seemed to be wrong. We're not actually using the result for anything at this point though.
* use LANG set to C to prevent localized dates etcDaniel Stenberg2003-01-221-0/+4
|
* pass the options to configure properlyDaniel Stenberg2003-01-221-3/+5
|
* check for empty confopts before asking for itDaniel Stenberg2003-01-221-1/+1
|
* put the configure options in the setup file was wellDaniel Stenberg2003-01-221-2/+39
| | | | | | | make -i show lib/config.h and some initial checks to prevent this running multiple times without the CVS having changed
* run 'make test-full' instead of 'make test' to get more details in case ofDaniel Stenberg2003-01-211-2/+2
| | | | errors
* use 'make test-full' instead of only 'make test' as it gives a lot of moreDaniel Stenberg2003-01-211-2/+2
| | | | info in case of failures
* use process id in build directory name to do betterDaniel Stenberg2003-01-201-4/+6
|
* first attempt at script for distributed testing on various unix hostsDaniel Stenberg2003-01-201-0/+139
View Lorry Controller Status