aboutsummaryrefslogtreecommitdiff
path: root/html-test/README.markdown
diff options
context:
space:
mode:
authorSimon Hengel <sol@typeful.net>2012-10-18 08:48:02 +0200
committerSimon Hengel <sol@typeful.net>2012-10-18 08:50:26 +0200
commita20efa02eef21b70c2cc3183217b55ce470b2826 (patch)
tree2891c749ca77ef3c7a096916df53f25eda9a8dad /html-test/README.markdown
parentcbc8e789b8b6308144ebdaf866ec202fe608647d (diff)
Use markdown for html-test/README
Diffstat (limited to 'html-test/README.markdown')
-rw-r--r--html-test/README.markdown27
1 files changed, 27 insertions, 0 deletions
diff --git a/html-test/README.markdown b/html-test/README.markdown
new file mode 100644
index 00000000..8d57acab
--- /dev/null
+++ b/html-test/README.markdown
@@ -0,0 +1,27 @@
+This is a testsuite for Haddock that uses the concept of "golden files". That
+is, it compares output files against a set of reference files.
+
+To add a new test:
+
+ 1. Create a module in the `html-test/src` directory.
+
+ 2. Run `cabal test`. You should now have `html-test/out/<modulename>.html`.
+ The test passes since there is no reference file to compare with.
+
+ 3. To make a reference file from the output file, run
+
+ html-test/accept.lhs <modulename>
+
+Tips and tricks:
+
+To "accept" all output files (copy them to reference files), run
+
+ runhaskell accept.lhs
+
+You can run all tests despite failing tests, like so
+
+ cabal test --test-option=all
+
+You can pass extra options to haddock like so
+
+ cabal test --test-options='all --title="All Tests"'