summaryrefslogtreecommitdiff
path: root/testsuite/driver/testlib.py
Commit message (Collapse)AuthorAgeFilesLines
...
* Don't bother testing ghostscript if we have no profiling librariesSimon Marlow2008-05-091-3/+4
|
* Add if_os, and change derefnull/divbyzero to use itIan Lynagh2008-04-201-0/+6
| | | | | I think different Linux setups generate different output for these tests, so we just ignore the output on Linux now.
* Add if_platform and remove *_if_platformIan Lynagh2008-04-201-29/+5
| | | | This removes some duplication from the testsuite driver
* Apply cmd_prefix also to the ghci wayThorkil Naur2008-02-081-1/+6
|
* Cache the results of looking to see if we have a library or notIan Lynagh2008-02-101-11/+22
| | | | | | When handling a test says reqlib('mtl') we cache the result, so we only need to run ghc-pkg once no matter how many tests say it. This makes quite a difference, especially on Windows.
* Find compiler version-specific output files automaticallySimon Marlow2008-02-061-15/+36
| | | | | | | | | | | | | | Also, clean up the way we find the output file. From the comment: # Finding the sample output. The filename is of the form # # <test>.stdout[-<compiler>][-<version>][-<wordsize>][-<platform>] # # and we pick the most specific version available. The <version> is # the major version of the compiler (e.g. 6.8.2 would be "6.8"). For # more fine-grained control use if_compiler_lt(). I'll update the wiki too.
* If the "setup" argument to test is a list, treat it as a list of functionsSimon Marlow2008-02-061-0/+5
| | | | Avoids the need to write composes()
* allow extra normalisation to be applied to the output on a per-test basisSimon Marlow2008-01-221-5/+11
|
* Simplify the testsuite driverIan Lynagh2007-12-151-78/+29
| | | | | | | Now instead of saying, e.g. namebase_if_compiler_lt('ghc','6.9', 'foo-6.8') you say if_compiler_lt('ghc','6.9', namebase('foo-6.8'))
* Be more consistent in printing about framework failuresIan Lynagh2007-11-231-4/+5
|
* remove the _with_prefix forms of compile_and_run, add cmd_prefix() config ↵Simon Marlow2007-11-221-15/+15
| | | | instead
* skip arrow for ghc < 6.9 (Control.Category doesn't exist)Ian Lynagh2007-11-101-1/+10
|
* add extra_hc_opts for ghci_scriptsimonpj2007-11-061-0/+1
|
* Add space after -ignore-dot-ghciSimon Marlow2007-10-221-1/+1
| | | | pointed out by Claus Reinke
* Add namebase_if_compiler_lt to the testsuite driverIan Lynagh2007-10-191-3/+27
| | | | | | | | Allows us to use a different name base for foo.stderr for old compilers, e.g. test('tc141', namebase_if_compiler_lt('ghc', '6.9', 'tc141-6.8'), compile_fail, [''])
* Pass -ignore-dot-ghci to GHCiSimon Marlow2007-09-261-1/+1
|
* add expect_fail_ways_if_compiler_lt()Simon Marlow2007-09-261-0/+13
| | | | somehow I forgot to add this yesterday, I'm mystified about where it went..
* Various testsuite helper functions like expect_broken_if_compiler_geIan Lynagh2007-09-211-0/+54
|
* Allow files-to-cleanup to be attached to a test using extra_clean(files)Simon Marlow2007-09-131-0/+10
| | | | This is better than a top-level clean() when using threads
* don't try to run hp2ps if the program failed with exit(>127)Simon Marlow2007-09-031-1/+2
|
* Add exit_code_if_platform functions to the testlibIan Lynagh2007-08-031-0/+7
|
* Define and use rawSystem in the testsuite driver and the driver testsIan Lynagh2007-08-021-13/+15
| | | | spawnlp doesn't exist in Windows, so this avoids using it.
* Adding compile_and_run_with_prefix and multimod_compile_and_run_with_prefix ↵andy2007-07-181-6/+18
| | | | as test-fns.
* reqlib now uses subprocess (when it's available)Ian Lynagh2007-07-091-1/+11
| | | | Fixes a problem where /dev/null doesn't exist on mingw.
* Fall back to os.spawnv is we don't have the subprocess moduleIan Lynagh2007-07-071-3/+17
|
* clean up the extra_ways machinerySimon Marlow2007-06-271-1/+4
|
* Use subprocess.call rather than os.spawnv in the testsuite driverIan Lynagh2007-07-061-5/+8
| | | | os.spawnv is broken on non-cygwin Windows (looks like it goes via a shell).
* Allow the "< /dev/null" when running tests to be disabledIan Lynagh2007-06-231-1/+8
| | | | | The user001 test in the unix package can fail when stdin comes from a file.
* add extra_ways, for adding running a test in more waysSimon Marlow2007-06-131-0/+10
|
* Add comments about runCmdsimonpj2007-06-061-0/+11
|
* sed-o in the testsuite driverIan Lynagh2007-05-051-1/+1
|
* Don't dump stderr/stdout before we've split the ghci outputIan Lynagh2007-05-051-7/+7
| | | | Fixes some framework failures.
* Fix commentIan Lynagh2007-05-041-1/+1
|
* Close 2 files that we open so we know that the writes have actually happenedIan Lynagh2007-04-161-0/+2
|
* Allow tests to be skipped based on whether or not the compiler has a tagIan Lynagh2007-04-131-0/+16
| | | | | | When working on a new foo extension, you can now put your tests in the testsuite, set ProjectTags=-foo in mk/build.mk and skip_unless_tag('foo') in the tests.
* Print stdout/stderr when we get a bad resultIan Lynagh2007-04-131-0/+12
|
* Check stderr output before stdout outputIan Lynagh2007-04-011-4/+4
|
* On failure, diff the normalised test outputsIan Lynagh2007-04-011-76/+77
|
* Use unified, rather than context, diffsIan Lynagh2007-03-301-1/+1
|
* MERGED: Add req_profiling to the driverIan Lynagh2007-03-051-0/+4
|
* Windows: the .ps file is <prog>.exe.psSimon Marlow2007-02-281-2/+5
|
* Look for .T files in packages tooSimon Marlow2007-02-211-2/+5
| | | | | | | | This means we can put package-specific tests in the repository for the package, rather than putting them in the testsuite. There should be a .T file to go with the tests, in the same way as for other tests in the testsuite (but this could be in addition to a standalone test driver that works with Cabal's 'setup test').
* Expect tests to fail if we don't have libraries that they useIan Lynagh2007-02-061-0/+9
|
* Don't put double-quotes around the command when passing it to timeoutSimon Marlow2007-02-021-3/+1
| | | | | | | | Fixes the testsuite on Cygwin/MSYS using Cygwin Python. I hope it hasn't broken it using the native Python... if it has, we'll need to revisit. I can't see a good reason for the quotes, anyway.
* Ignore output when told to in the ghci wayIan Lynagh2007-01-311-1/+2
|
* fixes to the ignore-output patchSimon Marlow2007-01-161-12/+8
|
* oops, fix bogosity in "ignore_output should be a test option"Simon Marlow2007-01-081-1/+1
|
* ignore_output should be a test optionSimon Marlow2007-01-051-41/+20
|
* fail -> brokenIan Lynagh2006-12-191-1/+1
|
* fail -> brokenIan Lynagh2006-12-191-0/+7
|
View Lorry Controller Status