4 This directory holds many test scripts for core GIT tools. The
5 first part of this short document describes how to run the tests
8 When fixing the tools or adding enhancements, you are strongly
9 encouraged to add tests in this directory to cover what you are
10 trying to fix or enhance. The later part of this short document
11 describes how your test scripts should be organized.
13 The mechanism that powers this testsuite is directly imported from the
14 Core GIT Tests, in directory t/ of the git repository. Files are base
15 on Core GIT version 1.3.0.rc4.g5069.
21 The easiest way to run tests is to say "make -C t". This runs all
24 *** t0000-basic.sh ***
25 * ok 1: .git/objects should be empty after git-init-db in an empty repo.
26 * ok 2: .git/objects should have 256 subdirectories.
27 * ok 3: git-update-index without --add should fail adding.
29 * ok 23: no diff after checkout and git-update-index --refresh.
30 * passed all 23 test(s)
31 *** t0100-environment-names.sh ***
32 * ok 1: using old names should issue warnings.
33 * ok 2: using old names but having new names should not issue warnings.
36 Or you can run each test individually from command line, like
39 $ sh ./t3001-ls-files-killed.sh
40 * ok 1: git-update-index --add to add various paths.
41 * ok 2: git-ls-files -k to show killed files.
42 * ok 3: validate git-ls-files -k output.
43 * passed all 3 test(s)
45 You can pass --verbose (or -v), --debug (or -d), and --immediate
46 (or -i) command line argument to the test.
49 This makes the test more verbose. Specifically, the
50 command being run and their output if any are also
54 This may help the person who is developing a new test.
55 It causes the command defined with test_debug to run.
58 This causes the test to immediately exit upon the first
65 The test files are named as:
67 tNNNN-commandname-details.sh
69 where N is a decimal digit.
71 Here is a proposal for numbering, loosely based on the Core GIT
72 numbering conventions.
74 First two digit tells the particular command we are testing:
81 Third and fourth digit (optionally) tells the particular switch or
82 group of switches we are testing.
84 If you create files under t/ directory (i.e. here) that is not
85 the top-level test script, never name the file to match the above
86 pattern. The Makefile here considers all such files as the
87 top-level test script and tries to run all of them. A care is
88 especially needed if you are creating a common test library
89 file, similar to test-lib.sh, because such a library file may
90 not be suitable for standalone execution.
96 The test script is written as a shell script. It should start
97 with the standard "#!/bin/sh" with copyright notices, and an
98 assignment to variable 'test_description', like this:
102 # Copyright (c) 2005 Junio C Hamano
105 test_description='xxx test (option --frotz)
107 This test registers the following structure in the cache
108 and tries to run git-ls-files with option --frotz.'
114 After assigning test_description, the test script should source
115 test-lib.sh like this:
119 This test harness library does the following things:
121 - If the script is invoked with command line argument --help
122 (or -h), it shows the test_description and exits.
124 - Creates an empty test directory with an empty .git/objects
125 database and chdir(2) into it. This directory is 't/trash'
126 if you must know, but I do not think you care.
128 - Defines standard test helper functions for your scripts to
129 use. These functions are designed to make all scripts behave
130 consistently when command line arguments --verbose (or -v),
131 --debug (or -d), and --immediate (or -i) is given.
137 Your script will be a sequence of tests, using helper functions
138 from the test harness library. At the end of the script, call
145 There are a handful helper functions defined in the test harness
146 library for your script to use.
148 - test_expect_success <message> <script>
150 This takes two strings as parameter, and evaluates the
151 <script>. If it yields success, test is considered
152 successful. <message> should state what it is testing.
156 test_expect_success \
157 'git-write-tree should be able to write an empty tree.' \
158 'tree=$(git-write-tree)'
160 - test_expect_failure <message> <script>
162 This is the opposite of test_expect_success. If <script>
163 yields success, test is considered a failure.
167 test_expect_failure \
168 'git-update-index without --add should fail adding.' \
169 'git-update-index should-be-empty'
171 - test_debug <script>
173 This takes a single argument, <script>, and evaluates it only
174 when the test script is started with --debug command line
175 argument. This is primarily meant for use during the
176 development of a new test script.
180 Your test script must have test_done at the end. Its purpose
181 is to summarize successes and failures in the test script and
182 exit with an appropriate error code.
185 Tips for Writing Tests
186 ----------------------
188 As with any programming projects, existing programs are the best
189 source of the information. However, do _not_ emulate
190 t0000-basic.sh when writing your tests. The test is special in
191 that it tries to validate the very core of GIT. For example, it
192 knows that there will be 256 subdirectories under .git/objects/,
193 and it knows that the object ID of an empty tree is a certain
194 40-byte string. This is deliberately done so in t0000-basic.sh
195 because the things the very basic core test tries to achieve is
196 to serve as a basis for people who are changing the GIT internal
197 drastically. For these people, after making certain changes,
198 not seeing failures from the basic test _is_ a failure. And
199 such drastic changes to the core GIT that even changes these
200 otherwise supposedly stable object IDs should be accompanied by
201 an update to t0000-basic.sh.
203 However, other tests that simply rely on basic parts of the core
204 GIT working properly should not have that level of intimate
205 knowledge of the core GIT internals. If all the test scripts
206 hardcoded the object IDs like t0000-basic.sh does, that defeats
207 the purpose of t0000-basic.sh, which is to isolate that level of
208 validation in one place. Your test also ends up needing
209 updating when such a change to the internal happens, so do _not_
210 do it and leave the low level of validation to t0000-basic.sh.