e49534a1bf48017b48eaf8a93f6b7c226dad8c98
1 Core GIT Tests
2 ==============
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
6 and read their output.
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.
14 Running Tests
15 -------------
17 The easiest way to run tests is to say "make". This runs all
18 the tests.
20 *** t0000-basic.sh ***
21 ok 1 - .git/objects should be empty after git init in an empty repo.
22 ok 2 - .git/objects should have 3 subdirectories.
23 ok 3 - success is reported like this
24 ...
25 ok 43 - very long name in the index handled sanely
26 # fixed 1 known breakage(s)
27 # still have 1 known breakage(s)
28 # passed all remaining 42 test(s)
29 1..43
30 *** t0001-init.sh ***
31 ok 1 - plain
32 ok 2 - plain with GIT_WORK_TREE
33 ok 3 - plain bare
35 Since the tests all output TAP (see http://testanything.org) they can
36 be run with any TAP harness. Here's an example of parallel testing
37 powered by a recent version of prove(1):
39 $ prove --timer --jobs 15 ./t[0-9]*.sh
40 [19:17:33] ./t0005-signals.sh ................................... ok 36 ms
41 [19:17:33] ./t0022-crlf-rename.sh ............................... ok 69 ms
42 [19:17:33] ./t0024-crlf-archive.sh .............................. ok 154 ms
43 [19:17:33] ./t0004-unwritable.sh ................................ ok 289 ms
44 [19:17:33] ./t0002-gitfile.sh ................................... ok 480 ms
45 ===( 102;0 25/? 6/? 5/? 16/? 1/? 4/? 2/? 1/? 3/? 1... )===
47 prove and other harnesses come with a lot of useful options. The
48 --state option in particular is very useful:
50 # Repeat until no more failures
51 $ prove -j 15 --state=failed,save ./t[0-9]*.sh
53 You can also run each test individually from command line, like this:
55 $ sh ./t3010-ls-files-killed-modified.sh
56 ok 1 - git update-index --add to add various paths.
57 ok 2 - git ls-files -k to show killed files.
58 ok 3 - validate git ls-files -k output.
59 ok 4 - git ls-files -m to show modified files.
60 ok 5 - validate git ls-files -m output.
61 # passed all 5 test(s)
62 1..5
64 You can pass --verbose (or -v), --debug (or -d), and --immediate
65 (or -i) command line argument to the test, or by setting GIT_TEST_OPTS
66 appropriately before running "make".
68 --verbose::
69 This makes the test more verbose. Specifically, the
70 command being run and their output if any are also
71 output.
73 --debug::
74 This may help the person who is developing a new test.
75 It causes the command defined with test_debug to run.
77 --immediate::
78 This causes the test to immediately exit upon the first
79 failed test.
81 --long-tests::
82 This causes additional long-running tests to be run (where
83 available), for more exhaustive testing.
85 --valgrind::
86 Execute all Git binaries with valgrind and exit with status
87 126 on errors (just like regular tests, this will only stop
88 the test script when running under -i). Valgrind errors
89 go to stderr, so you might want to pass the -v option, too.
91 Since it makes no sense to run the tests with --valgrind and
92 not see any output, this option implies --verbose. For
93 convenience, it also implies --tee.
95 --tee::
96 In addition to printing the test output to the terminal,
97 write it to files named 't/test-results/$TEST_NAME.out'.
98 As the names depend on the tests' file names, it is safe to
99 run the tests with this option in parallel.
101 --with-dashes::
102 By default tests are run without dashed forms of
103 commands (like git-commit) in the PATH (it only uses
104 wrappers from ../bin-wrappers). Use this option to include
105 the build directory (..) in the PATH, which contains all
106 the dashed forms of commands. This option is currently
107 implied by other options like --valgrind and
108 GIT_TEST_INSTALLED.
110 --root=<directory>::
111 Create "trash" directories used to store all temporary data during
112 testing under <directory>, instead of the t/ directory.
113 Using this option with a RAM-based filesystem (such as tmpfs)
114 can massively speed up the test suite.
116 You can also set the GIT_TEST_INSTALLED environment variable to
117 the bindir of an existing git installation to test that installation.
118 You still need to have built this git sandbox, from which various
119 test-* support programs, templates, and perl libraries are used.
120 If your installed git is incomplete, it will silently test parts of
121 your built version instead.
123 When using GIT_TEST_INSTALLED, you can also set GIT_TEST_EXEC_PATH to
124 override the location of the dashed-form subcommands (what
125 GIT_EXEC_PATH would be used for during normal operation).
126 GIT_TEST_EXEC_PATH defaults to `$GIT_TEST_INSTALLED/git --exec-path`.
129 Skipping Tests
130 --------------
132 In some environments, certain tests have no way of succeeding
133 due to platform limitation, such as lack of 'unzip' program, or
134 filesystem that do not allow arbitrary sequence of non-NUL bytes
135 as pathnames.
137 You should be able to say something like
139 $ GIT_SKIP_TESTS=t9200.8 sh ./t9200-git-cvsexport-commit.sh
141 and even:
143 $ GIT_SKIP_TESTS='t[0-4]??? t91?? t9200.8' make
145 to omit such tests. The value of the environment variable is a
146 SP separated list of patterns that tells which tests to skip,
147 and either can match the "t[0-9]{4}" part to skip the whole
148 test, or t[0-9]{4} followed by ".$number" to say which
149 particular test to skip.
151 Note that some tests in the existing test suite rely on previous
152 test item, so you cannot arbitrarily disable one and expect the
153 remainder of test to check what the test originally was intended
154 to check.
157 Naming Tests
158 ------------
160 The test files are named as:
162 tNNNN-commandname-details.sh
164 where N is a decimal digit.
166 First digit tells the family:
168 0 - the absolute basics and global stuff
169 1 - the basic commands concerning database
170 2 - the basic commands concerning the working tree
171 3 - the other basic commands (e.g. ls-files)
172 4 - the diff commands
173 5 - the pull and exporting commands
174 6 - the revision tree commands (even e.g. merge-base)
175 7 - the porcelainish commands concerning the working tree
176 8 - the porcelainish commands concerning forensics
177 9 - the git tools
179 Second digit tells the particular command we are testing.
181 Third digit (optionally) tells the particular switch or group of switches
182 we are testing.
184 If you create files under t/ directory (i.e. here) that is not
185 the top-level test script, never name the file to match the above
186 pattern. The Makefile here considers all such files as the
187 top-level test script and tries to run all of them. A care is
188 especially needed if you are creating a common test library
189 file, similar to test-lib.sh, because such a library file may
190 not be suitable for standalone execution.
193 Writing Tests
194 -------------
196 The test script is written as a shell script. It should start
197 with the standard "#!/bin/sh" with copyright notices, and an
198 assignment to variable 'test_description', like this:
200 #!/bin/sh
201 #
202 # Copyright (c) 2005 Junio C Hamano
203 #
205 test_description='xxx test (option --frotz)
207 This test registers the following structure in the cache
208 and tries to run git-ls-files with option --frotz.'
211 Source 'test-lib.sh'
212 --------------------
214 After assigning test_description, the test script should source
215 test-lib.sh like this:
217 . ./test-lib.sh
219 This test harness library does the following things:
221 - If the script is invoked with command line argument --help
222 (or -h), it shows the test_description and exits.
224 - Creates an empty test directory with an empty .git/objects database
225 and chdir(2) into it. This directory is 't/trash
226 directory.$test_name_without_dotsh', with t/ subject to change by
227 the --root option documented above.
229 - Defines standard test helper functions for your scripts to
230 use. These functions are designed to make all scripts behave
231 consistently when command line arguments --verbose (or -v),
232 --debug (or -d), and --immediate (or -i) is given.
234 Do's, don'ts & things to keep in mind
235 -------------------------------------
237 Here are a few examples of things you probably should and shouldn't do
238 when writing tests.
240 Do:
242 - Put all code inside test_expect_success and other assertions.
244 Even code that isn't a test per se, but merely some setup code
245 should be inside a test assertion.
247 - Chain your test assertions
249 Write test code like this:
251 git merge foo &&
252 git push bar &&
253 test ...
255 Instead of:
257 git merge hla
258 git push gh
259 test ...
261 That way all of the commands in your tests will succeed or fail. If
262 you must ignore the return value of something (e.g., the return
263 after unsetting a variable that was already unset is unportable) it's
264 best to indicate so explicitly with a semicolon:
266 unset HLAGH;
267 git merge hla &&
268 git push gh &&
269 test ...
271 - Check the test coverage for your tests. See the "Test coverage"
272 below.
274 Don't blindly follow test coverage metrics, they're a good way to
275 spot if you've missed something. If a new function you added
276 doesn't have any coverage you're probably doing something wrong,
277 but having 100% coverage doesn't necessarily mean that you tested
278 everything.
280 Tests that are likely to smoke out future regressions are better
281 than tests that just inflate the coverage metrics.
283 Don't:
285 - exit() within a <script> part.
287 The harness will catch this as a programming error of the test.
288 Use test_done instead if you need to stop the tests early (see
289 "Skipping tests" below).
291 - Break the TAP output
293 The raw output from your test may be interpreted by a TAP harness. TAP
294 harnesses will ignore everything they don't know about, but don't step
295 on their toes in these areas:
297 - Don't print lines like "$x..$y" where $x and $y are integers.
299 - Don't print lines that begin with "ok" or "not ok".
301 TAP harnesses expect a line that begins with either "ok" and "not
302 ok" to signal a test passed or failed (and our harness already
303 produces such lines), so your script shouldn't emit such lines to
304 their output.
306 You can glean some further possible issues from the TAP grammar
307 (see http://search.cpan.org/perldoc?TAP::Parser::Grammar#TAP_Grammar)
308 but the best indication is to just run the tests with prove(1),
309 it'll complain if anything is amiss.
311 Keep in mind:
313 - Inside <script> part, the standard output and standard error
314 streams are discarded, and the test harness only reports "ok" or
315 "not ok" to the end user running the tests. Under --verbose, they
316 are shown to help debugging the tests.
319 Skipping tests
320 --------------
322 If you need to skip all the remaining tests you should set skip_all
323 and immediately call test_done. The string you give to skip_all will
324 be used as an explanation for why the test was skipped. for instance:
326 if ! test_have_prereq PERL
327 then
328 skip_all='skipping perl interface tests, perl not available'
329 test_done
330 fi
332 End with test_done
333 ------------------
335 Your script will be a sequence of tests, using helper functions
336 from the test harness library. At the end of the script, call
337 'test_done'.
340 Test harness library
341 --------------------
343 There are a handful helper functions defined in the test harness
344 library for your script to use.
346 - test_expect_success [<prereq>] <message> <script>
348 Usually takes two strings as parameter, and evaluates the
349 <script>. If it yields success, test is considered
350 successful. <message> should state what it is testing.
352 Example:
354 test_expect_success \
355 'git-write-tree should be able to write an empty tree.' \
356 'tree=$(git-write-tree)'
358 If you supply three parameters the first will be taken to be a
359 prerequisite, see the test_set_prereq and test_have_prereq
360 documentation below:
362 test_expect_success TTY 'git --paginate rev-list uses a pager' \
363 ' ... '
365 You can also supply a comma-separated list of prerequisites, in the
366 rare case where your test depends on more than one:
368 test_expect_success PERL,PYTHON 'yo dawg' \
369 ' test $(perl -E 'print eval "1 +" . qx[python -c "print 2"]') == "4" '
371 - test_expect_failure [<prereq>] <message> <script>
373 This is NOT the opposite of test_expect_success, but is used
374 to mark a test that demonstrates a known breakage. Unlike
375 the usual test_expect_success tests, which say "ok" on
376 success and "FAIL" on failure, this will say "FIXED" on
377 success and "still broken" on failure. Failures from these
378 tests won't cause -i (immediate) to stop.
380 Like test_expect_success this function can optionally use a three
381 argument invocation with a prerequisite as the first argument.
383 - test_expect_code [<prereq>] <code> <message> <script>
385 Analogous to test_expect_success, but pass the test if it exits
386 with a given exit <code>
388 test_expect_code 1 'Merge with d/f conflicts' 'git merge "merge msg" B master'
390 - test_debug <script>
392 This takes a single argument, <script>, and evaluates it only
393 when the test script is started with --debug command line
394 argument. This is primarily meant for use during the
395 development of a new test script.
397 - test_done
399 Your test script must have test_done at the end. Its purpose
400 is to summarize successes and failures in the test script and
401 exit with an appropriate error code.
403 - test_tick
405 Make commit and tag names consistent by setting the author and
406 committer times to defined stated. Subsequent calls will
407 advance the times by a fixed amount.
409 - test_commit <message> [<filename> [<contents>]]
411 Creates a commit with the given message, committing the given
412 file with the given contents (default for both is to reuse the
413 message string), and adds a tag (again reusing the message
414 string as name). Calls test_tick to make the SHA-1s
415 reproducible.
417 - test_merge <message> <commit-or-tag>
419 Merges the given rev using the given message. Like test_commit,
420 creates a tag and calls test_tick before committing.
422 - test_set_prereq SOME_PREREQ
424 Set a test prerequisite to be used later with test_have_prereq. The
425 test-lib will set some prerequisites for you, see the
426 "Prerequisites" section below for a full list of these.
428 Others you can set yourself and use later with either
429 test_have_prereq directly, or the three argument invocation of
430 test_expect_success and test_expect_failure.
432 - test_have_prereq SOME PREREQ
434 Check if we have a prerequisite previously set with
435 test_set_prereq. The most common use of this directly is to skip
436 all the tests if we don't have some essential prerequisite:
438 if ! test_have_prereq PERL
439 then
440 skip_all='skipping perl interface tests, perl not available'
441 test_done
442 fi
444 - test_external [<prereq>] <message> <external> <script>
446 Execute a <script> with an <external> interpreter (like perl). This
447 was added for tests like t9700-perl-git.sh which do most of their
448 work in an external test script.
450 test_external \
451 'GitwebCache::*FileCache*' \
452 "$PERL_PATH" "$TEST_DIRECTORY"/t9503/test_cache_interface.pl
454 If the test is outputting its own TAP you should set the
455 test_external_has_tap variable somewhere before calling the first
456 test_external* function. See t9700-perl-git.sh for an example.
458 # The external test will outputs its own plan
459 test_external_has_tap=1
461 - test_external_without_stderr [<prereq>] <message> <external> <script>
463 Like test_external but fail if there's any output on stderr,
464 instead of checking the exit code.
466 test_external_without_stderr \
467 'Perl API' \
468 "$PERL_PATH" "$TEST_DIRECTORY"/t9700/test.pl
470 - test_must_fail <git-command>
472 Run a git command and ensure it fails in a controlled way. Use
473 this instead of "! <git-command>". When git-command dies due to a
474 segfault, test_must_fail diagnoses it as an error; "! <git-command>"
475 treats it as just another expected failure, which would let such a
476 bug go unnoticed.
478 - test_might_fail <git-command>
480 Similar to test_must_fail, but tolerate success, too. Use this
481 instead of "<git-command> || :" to catch failures due to segv.
483 - test_cmp <expected> <actual>
485 Check whether the content of the <actual> file matches the
486 <expected> file. This behaves like "cmp" but produces more
487 helpful output when the test is run with "-v" option.
489 - test_when_finished <script>
491 Prepend <script> to a list of commands to run to clean up
492 at the end of the current test. If some clean-up command
493 fails, the test will not pass.
495 Example:
497 test_expect_success 'branch pointing to non-commit' '
498 git rev-parse HEAD^{tree} >.git/refs/heads/invalid &&
499 test_when_finished "git update-ref -d refs/heads/invalid" &&
500 ...
501 '
503 Prerequisites
504 -------------
506 These are the prerequisites that the test library predefines with
507 test_have_prereq.
509 See the prereq argument to the test_* functions in the "Test harness
510 library" section above and the "test_have_prereq" function for how to
511 use these, and "test_set_prereq" for how to define your own.
513 - PERL & PYTHON
515 Git wasn't compiled with NO_PERL=YesPlease or
516 NO_PYTHON=YesPlease. Wrap any tests that need Perl or Python in
517 these.
519 - POSIXPERM
521 The filesystem supports POSIX style permission bits.
523 - BSLASHPSPEC
525 Backslashes in pathspec are not directory separators. This is not
526 set on Windows. See 6fd1106a for details.
528 - EXECKEEPSPID
530 The process retains the same pid across exec(2). See fb9a2bea for
531 details.
533 - SYMLINKS
535 The filesystem we're on supports symbolic links. E.g. a FAT
536 filesystem doesn't support these. See 704a3143 for details.
538 - SANITY
540 Test is not run by root user, and an attempt to write to an
541 unwritable file is expected to fail correctly.
543 Tips for Writing Tests
544 ----------------------
546 As with any programming projects, existing programs are the best
547 source of the information. However, do _not_ emulate
548 t0000-basic.sh when writing your tests. The test is special in
549 that it tries to validate the very core of GIT. For example, it
550 knows that there will be 256 subdirectories under .git/objects/,
551 and it knows that the object ID of an empty tree is a certain
552 40-byte string. This is deliberately done so in t0000-basic.sh
553 because the things the very basic core test tries to achieve is
554 to serve as a basis for people who are changing the GIT internal
555 drastically. For these people, after making certain changes,
556 not seeing failures from the basic test _is_ a failure. And
557 such drastic changes to the core GIT that even changes these
558 otherwise supposedly stable object IDs should be accompanied by
559 an update to t0000-basic.sh.
561 However, other tests that simply rely on basic parts of the core
562 GIT working properly should not have that level of intimate
563 knowledge of the core GIT internals. If all the test scripts
564 hardcoded the object IDs like t0000-basic.sh does, that defeats
565 the purpose of t0000-basic.sh, which is to isolate that level of
566 validation in one place. Your test also ends up needing
567 updating when such a change to the internal happens, so do _not_
568 do it and leave the low level of validation to t0000-basic.sh.
570 Test coverage
571 -------------
573 You can use the coverage tests to find code paths that are not being
574 used or properly exercised yet.
576 To do that, run the coverage target at the top-level (not in the t/
577 directory):
579 make coverage
581 That'll compile Git with GCC's coverage arguments, and generate a test
582 report with gcov after the tests finish. Running the coverage tests
583 can take a while, since running the tests in parallel is incompatible
584 with GCC's coverage mode.
586 After the tests have run you can generate a list of untested
587 functions:
589 make coverage-untested-functions
591 You can also generate a detailed per-file HTML report using the
592 Devel::Cover module. To install it do:
594 # On Debian or Ubuntu:
595 sudo aptitude install libdevel-cover-perl
597 # From the CPAN with cpanminus
598 curl -L http://cpanmin.us | perl - --sudo --self-upgrade
599 cpanm --sudo Devel::Cover
601 Then, at the top-level:
603 make cover_db_html
605 That'll generate a detailed cover report in the "cover_db_html"
606 directory, which you can then copy to a webserver, or inspect locally
607 in a browser.
609 Smoke testing
610 -------------
612 The Git test suite has support for smoke testing. Smoke testing is
613 when you submit the results of a test run to a central server for
614 analysis and aggregation.
616 Running a smoke tester is an easy and valuable way of contributing to
617 Git development, particularly if you have access to an uncommon OS on
618 obscure hardware.
620 After building Git you can generate a smoke report like this in the
621 "t" directory:
623 make clean smoke
625 You can also pass arguments via the environment. This should make it
626 faster:
628 GIT_TEST_OPTS='--root=/dev/shm' TEST_JOBS=10 make clean smoke
630 The "smoke" target will run the Git test suite with Perl's
631 "TAP::Harness" module, and package up the results in a .tar.gz archive
632 with "TAP::Harness::Archive". The former is included with Perl v5.10.1
633 or later, but you'll need to install the latter from the CPAN. See the
634 "Test coverage" section above for how you might do that.
636 Once the "smoke" target finishes you'll see a message like this:
638 TAP Archive created at <path to git>/t/test-results/git-smoke.tar.gz
640 To upload the smoke report you need to have curl(1) installed, then
641 do:
643 make smoke_report
645 To upload the report anonymously. Hopefully that'll return something
646 like "Reported #7 added.".
648 If you're going to be uploading reports frequently please request a
649 user account by E-Mailing gitsmoke@v.nix.is. Once you have a username
650 and password you'll be able to do:
652 SMOKE_USERNAME=<username> SMOKE_PASSWORD=<password> make smoke_report
654 You can also add an additional comment to attach to the report, and/or
655 a comma separated list of tags:
657 SMOKE_USERNAME=<username> SMOKE_PASSWORD=<password> \
658 SMOKE_COMMENT=<comment> SMOKE_TAGS=<tags> \
659 make smoke_report
661 Once the report is uploaded it'll be made available at
662 http://smoke.git.nix.is, here's an overview of Recent Smoke Reports
663 for Git:
665 http://smoke.git.nix.is/app/projects/smoke_reports/1
667 The reports will also be mirrored to GitHub every few hours:
669 http://github.com/gitsmoke/smoke-reports
671 The Smolder SQLite database is also mirrored and made available for
672 download:
674 http://github.com/gitsmoke/smoke-database
676 Note that the database includes hashed (with crypt()) user passwords
677 and E-Mail addresses. Don't use a valuable password for the smoke
678 service if you have an account, or an E-Mail address you don't want to
679 be publicly known. The user accounts are just meant to be convenient
680 labels, they're not meant to be secure.