aboutsummaryrefslogtreecommitdiff
path: root/haddock-test/src/Test/Haddock.hs
Commit message (Collapse)AuthorAgeFilesLines
* Clean up logic for guessing `-B` and `--lib` (#1026)Alec Theriault2019-02-131-0/+1
| | | | | | | | | | Haddock built with the `in-ghc-tree` flag tries harder to find the GHC lib folder and its own resources. This should make it possible to use `in-ghc-tree`-built Haddock without having to specify the `-B` and `--lib` options (just how you can use in-tree GHC without always specifying the `-B` option). The logic to do this relies on `getExecutablePath`, so we only get this auto-detection on platforms where this function works.
* Better test output when Haddock crashes on a test (#902)Alec Theriault2018-08-211-15/+29
| | | | | In particular: we report the tests that crashed seperately from the tests that produced incorrect output. In order for tests to pass (and exit 0), they must not crash and must produce the right output.
* Refactor handling of parens in types (#874)Alec Theriault2018-07-201-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | * Fix type parenthesization in Hoogle backend Ported the logic in the HTML and LaTeX backends for adding in parens into something top-level in 'GhcUtil'. Calling that from the Hoogle backend fixes #873. * Remove parenthesizing logic from LaTeX and XHTML backends Now, the only times that parenthesis in types are added in any backend is through the explicit 'HsParTy' constructor. Precedence is also represented as its own datatype. * List out cases explicitly vs. catch-all * Fix printing of parens for QuantifiedConstraints The priority of printing 'forall' types was just one too high. Fixes #877. * Accept HTML output for quantified contexts test
* haddock-test: A bit of refactoring for debuggabilityBen Gamari2016-05-101-9/+10
|
* testsuite: Rework handling of output sanitizationBen Gamari2016-02-081-6/+34
| | | | | | | | | Previously un-cleaned artifacts were kept as reference output, making it difficult to tell what has changed and causing spurious changes in the version control history. Here we rework this, cleaning the output during acceptance. To accomplish this it was necessary to move to strict I/O to ensure the reference handle was closed before accept attempts to open the reference file.
* Fix bug with accepting to non-existing directory.Łukasz Hanuszczak2015-08-241-1/+1
|
* Fix bug with test runner failing when run on multiple test packages.Łukasz Hanuszczak2015-08-221-2/+1
|
* Re-implement test acceptance functionality.Łukasz Hanuszczak2015-08-221-1/+12
|
* Fix bug with unnecessary checking old test output.Łukasz Hanuszczak2015-08-221-1/+3
|
* Fix bug with not all test output files being checked.Łukasz Hanuszczak2015-08-221-1/+2
|
* Make it possible to specify ignored files for test output.Łukasz Hanuszczak2015-08-221-1/+3
|
* Make Haddock test package more generic.Łukasz Hanuszczak2015-08-221-46/+44
|
* Implement output accepting mechanism in test package.Łukasz Hanuszczak2015-08-221-1/+15
|
* Change extension of test files used for diffing.Łukasz Hanuszczak2015-08-221-2/+2
|
* Adjust module visibility and items they export.Łukasz Hanuszczak2015-08-221-2/+2
|
* Create convenience wrappers to simplify in test entry points.Łukasz Hanuszczak2015-08-221-0/+4
|
* Make Haddock test package more generic.Łukasz Hanuszczak2015-08-221-19/+20
|
* Move Haddock runner of HTML test suite to Haddock test package.Łukasz Hanuszczak2015-08-221-0/+117