tests-and-benchmarks.markdown · text

on master help sign in

text/tests-and-benchmarks.markdown

63 lines 2.3 kB history raw

1Tests and benchmarks

2====================

4Prerequisites

5-------------

7To run the tests and benchmarks, you will need the test data, which

8you can clone from one of the following locations:

10* Mercurial master repository:

11 [bitbucket.org/bos/text-test-data](https://bitbucket.org/bos/text-test-data)

13* Git mirror repository:

14 [github.com/bos/text-test-data](https://github.com/bos/text-test-data)

16You should clone that repository into the `tests` subdirectory (your

17clone must be named `text-test-data` locally), then run `make -C

18tests/text-test-data` to uncompress the test files. Many tests and

19benchmarks will fail if the test files are missing.

21Functional tests

22----------------

24The functional tests are located in the `tests` subdirectory. An overview of

25what's in that directory:

27 Makefile Has targets for common tasks

28 Tests Source files of the testing code

29 scripts Various utility scripts

30 text-tests.cabal Cabal file that compiles all benchmarks

32The `text-tests.cabal` builds:

34- A copy of the text library, sharing the source code, but exposing all internal

35 modules, for testing purposes

36- The different test suites

38To compile, run all tests, and generate a coverage report, simply use `make`.

40Benchmarks

41----------

43The benchmarks are located in the `benchmarks` subdirectory. An overview of

44what's in that directory:

46 Makefile Has targets for common tasks

47 haskell Source files of the haskell benchmarks

48 python Python implementations of some benchmarks

49 ruby Ruby implementations of some benchmarks

50 text-benchmarks.cabal Cabal file which compiles all benchmarks

52To compile the benchmarks, navigate to the `benchmarks` subdirectory and run

53`cabal configure && cabal build`. Then, you can run the benchmarks using:

55 ./dist/build/text-benchmarks/text-benchmarks

57However, since there's quite a lot of benchmarks, you usually don't want to

58run them all. Instead, use the `-l` flag to get a list of benchmarks:

60 ./dist/build/text-benchmarks/text-benchmarks

62And run the ones you want to inspect. If you want to configure the benchmarks

63further, the exact parameters can be changed in `Benchmarks.hs`.