GitLab

another arrows test

-farrows instead of -fgla-exts

Add a test for IO strictness

extend a test

another (successful) arrow test

more arrow tests

update arrow tests

Add a quantification test

tcfail115 is an expected failure (there's a bug open on it).

accept new output

tc167 is an expected failure at the moment (there's a bug open on it).

more arrows tests

tests for arrow notation (needs arrows-branch)

Ignore *.comp.stderr and *.run.stderr, not *.stderr.
Add *.run.stdout to ignored files.

Commit missing .stderr file.

Add test for SourceForge #756454 (method must mention class type
variable), expected failure.

Add makefile

Begin arrow test suite

Get up a bit earlier in the morning than the strictness analyser.

Add kind test

Add kind-checking test

Add a test for hasNoRedexes

Introduce a ListP for consistency with ListE. Splicing in something with
a list pattern now works too.

Added various list tests.

Update tests for renamings.

Added foldTree examples.

Added a simple gread test sample; improved explanations of several samples

Add big-tuple tests

Added an example for traversal dimension; started a TODO list here

Data.Generics received a meaninful testsuite; more work to be done.

A test for stupid contexts in data types; killed 5.02.3

This is the beginning of a useful test suite for the boilerplate Generics.
Should all work fine with gmake (stage2) in ".".
Nice! All subdirs are visited automatically.
Much more smaller test cases are needed.
To be worked on.

Intermediate commit.
Added 002 and 003 test case.
Do only compile with local-build target.

Update this file to reflect reality.

rn040 is no longer an expected failure; accept the correct output

This test is an expected failure, but its original purpose seems to
have been lost along the way somewhere.

It is indeed failing, but I susepct not for the same reasons that it
was originally failing.  After fixing a few compilation errors I
managed to get it to run, whereupon it fails with "Prelude.read: no
parse", which may or may not be the correct result.  As far as I can
tell, this test never worked.  In the previous incarnation of the test
suite, it was even disabled.

ds053 is not an expected failure any more; accept the correct output.

Reclassify these test failures as "expected failures".  We are now at
ZBB (zero bug bounce).

The Great Renaming. I hope I've kept everything in sync - and all the tests
pass. Now datatypes follow the

data Foo = <a kind of Foo>Foo
         | <another kind of Foo>Foo

convention and the smart constructors can be uniformly created by
lowercasing the first letter of the constructor.

Added support for newtypes to TH and altered a test for them.

Add test for Show on unboxed types