]> git.ipfire.org Git - thirdparty/git.git/blob - t/README
Merge branch 'ds/reachable'
[thirdparty/git.git] / t / README
1 Core GIT Tests
2 ==============
3
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.
7
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.
12
13
14 Running Tests
15 -------------
16
17 The easiest way to run tests is to say "make". This runs all
18 the tests.
19
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
34
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):
38
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... )===
46
47 prove and other harnesses come with a lot of useful options. The
48 --state option in particular is very useful:
49
50 # Repeat until no more failures
51 $ prove -j 15 --state=failed,save ./t[0-9]*.sh
52
53 You can give DEFAULT_TEST_TARGET=prove on the make command (or define it
54 in config.mak) to cause "make test" to run tests under prove.
55 GIT_PROVE_OPTS can be used to pass additional options, e.g.
56
57 $ make DEFAULT_TEST_TARGET=prove GIT_PROVE_OPTS='--timer --jobs 16' test
58
59 You can also run each test individually from command line, like this:
60
61 $ sh ./t3010-ls-files-killed-modified.sh
62 ok 1 - git update-index --add to add various paths.
63 ok 2 - git ls-files -k to show killed files.
64 ok 3 - validate git ls-files -k output.
65 ok 4 - git ls-files -m to show modified files.
66 ok 5 - validate git ls-files -m output.
67 # passed all 5 test(s)
68 1..5
69
70 You can pass --verbose (or -v), --debug (or -d), and --immediate
71 (or -i) command line argument to the test, or by setting GIT_TEST_OPTS
72 appropriately before running "make".
73
74 -v::
75 --verbose::
76 This makes the test more verbose. Specifically, the
77 command being run and their output if any are also
78 output.
79
80 --verbose-only=<pattern>::
81 Like --verbose, but the effect is limited to tests with
82 numbers matching <pattern>. The number matched against is
83 simply the running count of the test within the file.
84
85 -x::
86 Turn on shell tracing (i.e., `set -x`) during the tests
87 themselves. Implies `--verbose`.
88 Ignored in test scripts that set the variable 'test_untraceable'
89 to a non-empty value, unless it's run with a Bash version
90 supporting BASH_XTRACEFD, i.e. v4.1 or later.
91
92 -d::
93 --debug::
94 This may help the person who is developing a new test.
95 It causes the command defined with test_debug to run.
96 The "trash" directory (used to store all temporary data
97 during testing) is not deleted even if there are no
98 failed tests so that you can inspect its contents after
99 the test finished.
100
101 -i::
102 --immediate::
103 This causes the test to immediately exit upon the first
104 failed test. Cleanup commands requested with
105 test_when_finished are not executed if the test failed,
106 in order to keep the state for inspection by the tester
107 to diagnose the bug.
108
109 -l::
110 --long-tests::
111 This causes additional long-running tests to be run (where
112 available), for more exhaustive testing.
113
114 -r::
115 --run=<test-selector>::
116 Run only the subset of tests indicated by
117 <test-selector>. See section "Skipping Tests" below for
118 <test-selector> syntax.
119
120 --valgrind=<tool>::
121 Execute all Git binaries under valgrind tool <tool> and exit
122 with status 126 on errors (just like regular tests, this will
123 only stop the test script when running under -i).
124
125 Since it makes no sense to run the tests with --valgrind and
126 not see any output, this option implies --verbose. For
127 convenience, it also implies --tee.
128
129 <tool> defaults to 'memcheck', just like valgrind itself.
130 Other particularly useful choices include 'helgrind' and
131 'drd', but you may use any tool recognized by your valgrind
132 installation.
133
134 As a special case, <tool> can be 'memcheck-fast', which uses
135 memcheck but disables --track-origins. Use this if you are
136 running tests in bulk, to see if there are _any_ memory
137 issues.
138
139 Note that memcheck is run with the option --leak-check=no,
140 as the git process is short-lived and some errors are not
141 interesting. In order to run a single command under the same
142 conditions manually, you should set GIT_VALGRIND to point to
143 the 't/valgrind/' directory and use the commands under
144 't/valgrind/bin/'.
145
146 --valgrind-only=<pattern>::
147 Like --valgrind, but the effect is limited to tests with
148 numbers matching <pattern>. The number matched against is
149 simply the running count of the test within the file.
150
151 --tee::
152 In addition to printing the test output to the terminal,
153 write it to files named 't/test-results/$TEST_NAME.out'.
154 As the names depend on the tests' file names, it is safe to
155 run the tests with this option in parallel.
156
157 --verbose-log::
158 Write verbose output to the same logfile as `--tee`, but do
159 _not_ write it to stdout. Unlike `--tee --verbose`, this option
160 is safe to use when stdout is being consumed by a TAP parser
161 like `prove`. Implies `--tee` and `--verbose`.
162
163 --with-dashes::
164 By default tests are run without dashed forms of
165 commands (like git-commit) in the PATH (it only uses
166 wrappers from ../bin-wrappers). Use this option to include
167 the build directory (..) in the PATH, which contains all
168 the dashed forms of commands. This option is currently
169 implied by other options like --valgrind and
170 GIT_TEST_INSTALLED.
171
172 --root=<directory>::
173 Create "trash" directories used to store all temporary data during
174 testing under <directory>, instead of the t/ directory.
175 Using this option with a RAM-based filesystem (such as tmpfs)
176 can massively speed up the test suite.
177
178 --chain-lint::
179 --no-chain-lint::
180 If --chain-lint is enabled, the test harness will check each
181 test to make sure that it properly "&&-chains" all commands (so
182 that a failure in the middle does not go unnoticed by the final
183 exit code of the test). This check is performed in addition to
184 running the tests themselves. You may also enable or disable
185 this feature by setting the GIT_TEST_CHAIN_LINT environment
186 variable to "1" or "0", respectively.
187
188 You can also set the GIT_TEST_INSTALLED environment variable to
189 the bindir of an existing git installation to test that installation.
190 You still need to have built this git sandbox, from which various
191 test-* support programs, templates, and perl libraries are used.
192 If your installed git is incomplete, it will silently test parts of
193 your built version instead.
194
195 When using GIT_TEST_INSTALLED, you can also set GIT_TEST_EXEC_PATH to
196 override the location of the dashed-form subcommands (what
197 GIT_EXEC_PATH would be used for during normal operation).
198 GIT_TEST_EXEC_PATH defaults to `$GIT_TEST_INSTALLED/git --exec-path`.
199
200
201 Skipping Tests
202 --------------
203
204 In some environments, certain tests have no way of succeeding
205 due to platform limitation, such as lack of 'unzip' program, or
206 filesystem that do not allow arbitrary sequence of non-NUL bytes
207 as pathnames.
208
209 You should be able to say something like
210
211 $ GIT_SKIP_TESTS=t9200.8 sh ./t9200-git-cvsexport-commit.sh
212
213 and even:
214
215 $ GIT_SKIP_TESTS='t[0-4]??? t91?? t9200.8' make
216
217 to omit such tests. The value of the environment variable is a
218 SP separated list of patterns that tells which tests to skip,
219 and either can match the "t[0-9]{4}" part to skip the whole
220 test, or t[0-9]{4} followed by ".$number" to say which
221 particular test to skip.
222
223 For an individual test suite --run could be used to specify that
224 only some tests should be run or that some tests should be
225 excluded from a run.
226
227 The argument for --run is a list of individual test numbers or
228 ranges with an optional negation prefix that define what tests in
229 a test suite to include in the run. A range is two numbers
230 separated with a dash and matches a range of tests with both ends
231 been included. You may omit the first or the second number to
232 mean "from the first test" or "up to the very last test"
233 respectively.
234
235 Optional prefix of '!' means that the test or a range of tests
236 should be excluded from the run.
237
238 If --run starts with an unprefixed number or range the initial
239 set of tests to run is empty. If the first item starts with '!'
240 all the tests are added to the initial set. After initial set is
241 determined every test number or range is added or excluded from
242 the set one by one, from left to right.
243
244 Individual numbers or ranges could be separated either by a space
245 or a comma.
246
247 For example, to run only tests up to a specific test (21), one
248 could do this:
249
250 $ sh ./t9200-git-cvsexport-commit.sh --run='1-21'
251
252 or this:
253
254 $ sh ./t9200-git-cvsexport-commit.sh --run='-21'
255
256 Common case is to run several setup tests (1, 2, 3) and then a
257 specific test (21) that relies on that setup:
258
259 $ sh ./t9200-git-cvsexport-commit.sh --run='1 2 3 21'
260
261 or:
262
263 $ sh ./t9200-git-cvsexport-commit.sh --run=1,2,3,21
264
265 or:
266
267 $ sh ./t9200-git-cvsexport-commit.sh --run='-3 21'
268
269 As noted above, the test set is built by going through the items
270 from left to right, so this:
271
272 $ sh ./t9200-git-cvsexport-commit.sh --run='1-4 !3'
273
274 will run tests 1, 2, and 4. Items that come later have higher
275 precedence. It means that this:
276
277 $ sh ./t9200-git-cvsexport-commit.sh --run='!3 1-4'
278
279 would just run tests from 1 to 4, including 3.
280
281 You may use negation with ranges. The following will run all
282 test in the test suite except from 7 up to 11:
283
284 $ sh ./t9200-git-cvsexport-commit.sh --run='!7-11'
285
286 Some tests in a test suite rely on the previous tests performing
287 certain actions, specifically some tests are designated as
288 "setup" test, so you cannot _arbitrarily_ disable one test and
289 expect the rest to function correctly.
290
291 --run is mostly useful when you want to focus on a specific test
292 and know what setup is needed for it. Or when you want to run
293 everything up to a certain test.
294
295
296 Running tests with special setups
297 ---------------------------------
298
299 The whole test suite could be run to test some special features
300 that cannot be easily covered by a few specific test cases. These
301 could be enabled by running the test suite with correct GIT_TEST_
302 environment set.
303
304 GIT_TEST_SPLIT_INDEX=<boolean> forces split-index mode on the whole
305 test suite. Accept any boolean values that are accepted by git-config.
306
307 GIT_TEST_FULL_IN_PACK_ARRAY=<boolean> exercises the uncommon
308 pack-objects code path where there are more than 1024 packs even if
309 the actual number of packs in repository is below this limit. Accept
310 any boolean values that are accepted by git-config.
311
312 GIT_TEST_OE_SIZE=<n> exercises the uncommon pack-objects code path
313 where we do not cache object size in memory and read it from existing
314 packs on demand. This normally only happens when the object size is
315 over 2GB. This variable forces the code path on any object larger than
316 <n> bytes.
317
318 GIT_TEST_OE_DELTA_SIZE=<n> exercises the uncommon pack-objects code
319 path where deltas larger than this limit require extra memory
320 allocation for bookkeeping.
321
322 GIT_TEST_VALIDATE_INDEX_CACHE_ENTRIES=<boolean> checks that cache-tree
323 records are valid when the index is written out or after a merge. This
324 is mostly to catch missing invalidation. Default is true.
325
326 GIT_TEST_COMMIT_GRAPH=<boolean>, when true, forces the commit-graph to
327 be written after every 'git commit' command, and overrides the
328 'core.commitGraph' setting to true.
329
330 GIT_TEST_FSMONITOR=$PWD/t7519/fsmonitor-all exercises the fsmonitor
331 code path for utilizing a file system monitor to speed up detecting
332 new or changed files.
333
334 GIT_TEST_INDEX_VERSION=<n> exercises the index read/write code path
335 for the index version specified. Can be set to any valid version
336 (currently 2, 3, or 4).
337
338 GIT_TEST_PRELOAD_INDEX=<boolean> exercises the preload-index code path
339 by overriding the minimum number of cache entries required per thread.
340
341 GIT_TEST_INDEX_THREADS=<n> enables exercising the multi-threaded loading
342 of the index for the whole test suite by bypassing the default number of
343 cache entries and thread minimums. Setting this to 1 will make the
344 index loading single threaded.
345
346 Naming Tests
347 ------------
348
349 The test files are named as:
350
351 tNNNN-commandname-details.sh
352
353 where N is a decimal digit.
354
355 First digit tells the family:
356
357 0 - the absolute basics and global stuff
358 1 - the basic commands concerning database
359 2 - the basic commands concerning the working tree
360 3 - the other basic commands (e.g. ls-files)
361 4 - the diff commands
362 5 - the pull and exporting commands
363 6 - the revision tree commands (even e.g. merge-base)
364 7 - the porcelainish commands concerning the working tree
365 8 - the porcelainish commands concerning forensics
366 9 - the git tools
367
368 Second digit tells the particular command we are testing.
369
370 Third digit (optionally) tells the particular switch or group of switches
371 we are testing.
372
373 If you create files under t/ directory (i.e. here) that is not
374 the top-level test script, never name the file to match the above
375 pattern. The Makefile here considers all such files as the
376 top-level test script and tries to run all of them. Care is
377 especially needed if you are creating a common test library
378 file, similar to test-lib.sh, because such a library file may
379 not be suitable for standalone execution.
380
381
382 Writing Tests
383 -------------
384
385 The test script is written as a shell script. It should start
386 with the standard "#!/bin/sh", and an
387 assignment to variable 'test_description', like this:
388
389 #!/bin/sh
390
391 test_description='xxx test (option --frotz)
392
393 This test registers the following structure in the cache
394 and tries to run git-ls-files with option --frotz.'
395
396
397 Source 'test-lib.sh'
398 --------------------
399
400 After assigning test_description, the test script should source
401 test-lib.sh like this:
402
403 . ./test-lib.sh
404
405 This test harness library does the following things:
406
407 - If the script is invoked with command line argument --help
408 (or -h), it shows the test_description and exits.
409
410 - Creates an empty test directory with an empty .git/objects database
411 and chdir(2) into it. This directory is 't/trash
412 directory.$test_name_without_dotsh', with t/ subject to change by
413 the --root option documented above.
414
415 - Defines standard test helper functions for your scripts to
416 use. These functions are designed to make all scripts behave
417 consistently when command line arguments --verbose (or -v),
418 --debug (or -d), and --immediate (or -i) is given.
419
420 Do's & don'ts
421 -------------
422
423 Here are a few examples of things you probably should and shouldn't do
424 when writing tests.
425
426 Here are the "do's:"
427
428 - Put all code inside test_expect_success and other assertions.
429
430 Even code that isn't a test per se, but merely some setup code
431 should be inside a test assertion.
432
433 - Chain your test assertions
434
435 Write test code like this:
436
437 git merge foo &&
438 git push bar &&
439 test ...
440
441 Instead of:
442
443 git merge hla
444 git push gh
445 test ...
446
447 That way all of the commands in your tests will succeed or fail. If
448 you must ignore the return value of something, consider using a
449 helper function (e.g. use sane_unset instead of unset, in order
450 to avoid unportable return value for unsetting a variable that was
451 already unset), or prepending the command with test_might_fail or
452 test_must_fail.
453
454 - Check the test coverage for your tests. See the "Test coverage"
455 below.
456
457 Don't blindly follow test coverage metrics; if a new function you added
458 doesn't have any coverage, then you're probably doing something wrong,
459 but having 100% coverage doesn't necessarily mean that you tested
460 everything.
461
462 Tests that are likely to smoke out future regressions are better
463 than tests that just inflate the coverage metrics.
464
465 - When a test checks for an absolute path that a git command generated,
466 construct the expected value using $(pwd) rather than $PWD,
467 $TEST_DIRECTORY, or $TRASH_DIRECTORY. It makes a difference on
468 Windows, where the shell (MSYS bash) mangles absolute path names.
469 For details, see the commit message of 4114156ae9.
470
471 - Remember that inside the <script> part, the standard output and
472 standard error streams are discarded, and the test harness only
473 reports "ok" or "not ok" to the end user running the tests. Under
474 --verbose, they are shown to help debug the tests.
475
476 And here are the "don'ts:"
477
478 - Don't exit() within a <script> part.
479
480 The harness will catch this as a programming error of the test.
481 Use test_done instead if you need to stop the tests early (see
482 "Skipping tests" below).
483
484 - Don't use '! git cmd' when you want to make sure the git command
485 exits with failure in a controlled way by calling "die()". Instead,
486 use 'test_must_fail git cmd'. This will signal a failure if git
487 dies in an unexpected way (e.g. segfault).
488
489 On the other hand, don't use test_must_fail for running regular
490 platform commands; just use '! cmd'. We are not in the business
491 of verifying that the world given to us sanely works.
492
493 - Don't feed the output of a git command to a pipe, as in:
494
495 git -C repo ls-files |
496 xargs -n 1 basename |
497 grep foo
498
499 which will discard git's exit code and may mask a crash. In the
500 above example, all exit codes are ignored except grep's.
501
502 Instead, write the output of that command to a temporary
503 file with ">" or assign it to a variable with "x=$(git ...)" rather
504 than pipe it.
505
506 - Don't use command substitution in a way that discards git's exit
507 code. When assigning to a variable, the exit code is not discarded,
508 e.g.:
509
510 x=$(git cat-file -p $sha) &&
511 ...
512
513 is OK because a crash in "git cat-file" will cause the "&&" chain
514 to fail, but:
515
516 test "refs/heads/foo" = "$(git symbolic-ref HEAD)"
517
518 is not OK and a crash in git could go undetected.
519
520 - Don't use perl without spelling it as "$PERL_PATH". This is to help
521 our friends on Windows where the platform Perl often adds CR before
522 the end of line, and they bundle Git with a version of Perl that
523 does not do so, whose path is specified with $PERL_PATH. Note that we
524 provide a "perl" function which uses $PERL_PATH under the hood, so
525 you do not need to worry when simply running perl in the test scripts
526 (but you do, for example, on a shebang line or in a sub script
527 created via "write_script").
528
529 - Don't use sh without spelling it as "$SHELL_PATH", when the script
530 can be misinterpreted by broken platform shell (e.g. Solaris).
531
532 - Don't chdir around in tests. It is not sufficient to chdir to
533 somewhere and then chdir back to the original location later in
534 the test, as any intermediate step can fail and abort the test,
535 causing the next test to start in an unexpected directory. Do so
536 inside a subshell if necessary.
537
538 - Don't save and verify the standard error of compound commands, i.e.
539 group commands, subshells, and shell functions (except test helper
540 functions like 'test_must_fail') like this:
541
542 ( cd dir && git cmd ) 2>error &&
543 test_cmp expect error
544
545 When running the test with '-x' tracing, then the trace of commands
546 executed in the compound command will be included in standard error
547 as well, quite possibly throwing off the subsequent checks examining
548 the output. Instead, save only the relevant git command's standard
549 error:
550
551 ( cd dir && git cmd 2>../error ) &&
552 test_cmp expect error
553
554 - Don't break the TAP output
555
556 The raw output from your test may be interpreted by a TAP harness. TAP
557 harnesses will ignore everything they don't know about, but don't step
558 on their toes in these areas:
559
560 - Don't print lines like "$x..$y" where $x and $y are integers.
561
562 - Don't print lines that begin with "ok" or "not ok".
563
564 TAP harnesses expect a line that begins with either "ok" and "not
565 ok" to signal a test passed or failed (and our harness already
566 produces such lines), so your script shouldn't emit such lines to
567 their output.
568
569 You can glean some further possible issues from the TAP grammar
570 (see https://metacpan.org/pod/TAP::Parser::Grammar#TAP-GRAMMAR)
571 but the best indication is to just run the tests with prove(1),
572 it'll complain if anything is amiss.
573
574
575 Skipping tests
576 --------------
577
578 If you need to skip tests you should do so by using the three-arg form
579 of the test_* functions (see the "Test harness library" section
580 below), e.g.:
581
582 test_expect_success PERL 'I need Perl' '
583 perl -e "hlagh() if unf_unf()"
584 '
585
586 The advantage of skipping tests like this is that platforms that don't
587 have the PERL and other optional dependencies get an indication of how
588 many tests they're missing.
589
590 If the test code is too hairy for that (i.e. does a lot of setup work
591 outside test assertions) you can also skip all remaining tests by
592 setting skip_all and immediately call test_done:
593
594 if ! test_have_prereq PERL
595 then
596 skip_all='skipping perl interface tests, perl not available'
597 test_done
598 fi
599
600 The string you give to skip_all will be used as an explanation for why
601 the test was skipped.
602
603 End with test_done
604 ------------------
605
606 Your script will be a sequence of tests, using helper functions
607 from the test harness library. At the end of the script, call
608 'test_done'.
609
610
611 Test harness library
612 --------------------
613
614 There are a handful helper functions defined in the test harness
615 library for your script to use.
616
617 - test_expect_success [<prereq>] <message> <script>
618
619 Usually takes two strings as parameters, and evaluates the
620 <script>. If it yields success, test is considered
621 successful. <message> should state what it is testing.
622
623 Example:
624
625 test_expect_success \
626 'git-write-tree should be able to write an empty tree.' \
627 'tree=$(git-write-tree)'
628
629 If you supply three parameters the first will be taken to be a
630 prerequisite; see the test_set_prereq and test_have_prereq
631 documentation below:
632
633 test_expect_success TTY 'git --paginate rev-list uses a pager' \
634 ' ... '
635
636 You can also supply a comma-separated list of prerequisites, in the
637 rare case where your test depends on more than one:
638
639 test_expect_success PERL,PYTHON 'yo dawg' \
640 ' test $(perl -E 'print eval "1 +" . qx[python -c "print 2"]') == "4" '
641
642 - test_expect_failure [<prereq>] <message> <script>
643
644 This is NOT the opposite of test_expect_success, but is used
645 to mark a test that demonstrates a known breakage. Unlike
646 the usual test_expect_success tests, which say "ok" on
647 success and "FAIL" on failure, this will say "FIXED" on
648 success and "still broken" on failure. Failures from these
649 tests won't cause -i (immediate) to stop.
650
651 Like test_expect_success this function can optionally use a three
652 argument invocation with a prerequisite as the first argument.
653
654 - test_debug <script>
655
656 This takes a single argument, <script>, and evaluates it only
657 when the test script is started with --debug command line
658 argument. This is primarily meant for use during the
659 development of a new test script.
660
661 - debug <git-command>
662
663 Run a git command inside a debugger. This is primarily meant for
664 use when debugging a failing test script.
665
666 - test_done
667
668 Your test script must have test_done at the end. Its purpose
669 is to summarize successes and failures in the test script and
670 exit with an appropriate error code.
671
672 - test_tick
673
674 Make commit and tag names consistent by setting the author and
675 committer times to defined state. Subsequent calls will
676 advance the times by a fixed amount.
677
678 - test_commit <message> [<filename> [<contents>]]
679
680 Creates a commit with the given message, committing the given
681 file with the given contents (default for both is to reuse the
682 message string), and adds a tag (again reusing the message
683 string as name). Calls test_tick to make the SHA-1s
684 reproducible.
685
686 - test_merge <message> <commit-or-tag>
687
688 Merges the given rev using the given message. Like test_commit,
689 creates a tag and calls test_tick before committing.
690
691 - test_set_prereq <prereq>
692
693 Set a test prerequisite to be used later with test_have_prereq. The
694 test-lib will set some prerequisites for you, see the
695 "Prerequisites" section below for a full list of these.
696
697 Others you can set yourself and use later with either
698 test_have_prereq directly, or the three argument invocation of
699 test_expect_success and test_expect_failure.
700
701 - test_have_prereq <prereq>
702
703 Check if we have a prerequisite previously set with test_set_prereq.
704 The most common way to use this explicitly (as opposed to the
705 implicit use when an argument is passed to test_expect_*) is to skip
706 all the tests at the start of the test script if we don't have some
707 essential prerequisite:
708
709 if ! test_have_prereq PERL
710 then
711 skip_all='skipping perl interface tests, perl not available'
712 test_done
713 fi
714
715 - test_external [<prereq>] <message> <external> <script>
716
717 Execute a <script> with an <external> interpreter (like perl). This
718 was added for tests like t9700-perl-git.sh which do most of their
719 work in an external test script.
720
721 test_external \
722 'GitwebCache::*FileCache*' \
723 perl "$TEST_DIRECTORY"/t9503/test_cache_interface.pl
724
725 If the test is outputting its own TAP you should set the
726 test_external_has_tap variable somewhere before calling the first
727 test_external* function. See t9700-perl-git.sh for an example.
728
729 # The external test will outputs its own plan
730 test_external_has_tap=1
731
732 - test_external_without_stderr [<prereq>] <message> <external> <script>
733
734 Like test_external but fail if there's any output on stderr,
735 instead of checking the exit code.
736
737 test_external_without_stderr \
738 'Perl API' \
739 perl "$TEST_DIRECTORY"/t9700/test.pl
740
741 - test_expect_code <exit-code> <command>
742
743 Run a command and ensure that it exits with the given exit code.
744 For example:
745
746 test_expect_success 'Merge with d/f conflicts' '
747 test_expect_code 1 git merge "merge msg" B master
748 '
749
750 - test_must_fail [<options>] <git-command>
751
752 Run a git command and ensure it fails in a controlled way. Use
753 this instead of "! <git-command>". When git-command dies due to a
754 segfault, test_must_fail diagnoses it as an error; "! <git-command>"
755 treats it as just another expected failure, which would let such a
756 bug go unnoticed.
757
758 Accepts the following options:
759
760 ok=<signal-name>[,<...>]:
761 Don't treat an exit caused by the given signal as error.
762 Multiple signals can be specified as a comma separated list.
763 Currently recognized signal names are: sigpipe, success.
764 (Don't use 'success', use 'test_might_fail' instead.)
765
766 - test_might_fail [<options>] <git-command>
767
768 Similar to test_must_fail, but tolerate success, too. Use this
769 instead of "<git-command> || :" to catch failures due to segv.
770
771 Accepts the same options as test_must_fail.
772
773 - test_cmp <expected> <actual>
774
775 Check whether the content of the <actual> file matches the
776 <expected> file. This behaves like "cmp" but produces more
777 helpful output when the test is run with "-v" option.
778
779 - test_cmp_rev <expected> <actual>
780
781 Check whether the <expected> rev points to the same commit as the
782 <actual> rev.
783
784 - test_line_count (= | -lt | -ge | ...) <length> <file>
785
786 Check whether a file has the length it is expected to.
787
788 - test_path_is_file <path> [<diagnosis>]
789 test_path_is_dir <path> [<diagnosis>]
790 test_path_is_missing <path> [<diagnosis>]
791
792 Check if the named path is a file, if the named path is a
793 directory, or if the named path does not exist, respectively,
794 and fail otherwise, showing the <diagnosis> text.
795
796 - test_when_finished <script>
797
798 Prepend <script> to a list of commands to run to clean up
799 at the end of the current test. If some clean-up command
800 fails, the test will not pass.
801
802 Example:
803
804 test_expect_success 'branch pointing to non-commit' '
805 git rev-parse HEAD^{tree} >.git/refs/heads/invalid &&
806 test_when_finished "git update-ref -d refs/heads/invalid" &&
807 ...
808 '
809
810 - test_write_lines <lines>
811
812 Write <lines> on standard output, one line per argument.
813 Useful to prepare multi-line files in a compact form.
814
815 Example:
816
817 test_write_lines a b c d e f g >foo
818
819 Is a more compact equivalent of:
820 cat >foo <<-EOF
821 a
822 b
823 c
824 d
825 e
826 f
827 g
828 EOF
829
830
831 - test_pause
832
833 This command is useful for writing and debugging tests and must be
834 removed before submitting. It halts the execution of the test and
835 spawns a shell in the trash directory. Exit the shell to continue
836 the test. Example:
837
838 test_expect_success 'test' '
839 git do-something >actual &&
840 test_pause &&
841 test_cmp expected actual
842 '
843
844 - test_ln_s_add <path1> <path2>
845
846 This function helps systems whose filesystem does not support symbolic
847 links. Use it to add a symbolic link entry to the index when it is not
848 important that the file system entry is a symbolic link, i.e., instead
849 of the sequence
850
851 ln -s foo bar &&
852 git add bar
853
854 Sometimes it is possible to split a test in a part that does not need
855 the symbolic link in the file system and a part that does; then only
856 the latter part need be protected by a SYMLINKS prerequisite (see below).
857
858 - test_oid_init
859
860 This function loads facts and useful object IDs related to the hash
861 algorithm(s) in use from the files in t/oid-info.
862
863 - test_oid_cache
864
865 This function reads per-hash algorithm information from standard
866 input (usually a heredoc) in the format described in
867 t/oid-info/README. This is useful for test-specific values, such as
868 object IDs, which must vary based on the hash algorithm.
869
870 Certain fixed values, such as hash sizes and common placeholder
871 object IDs, can be loaded with test_oid_init (described above).
872
873 - test_oid <key>
874
875 This function looks up a value for the hash algorithm in use, based
876 on the key given. The value must have been loaded using
877 test_oid_init or test_oid_cache. Providing an unknown key is an
878 error.
879
880 Prerequisites
881 -------------
882
883 These are the prerequisites that the test library predefines with
884 test_have_prereq.
885
886 See the prereq argument to the test_* functions in the "Test harness
887 library" section above and the "test_have_prereq" function for how to
888 use these, and "test_set_prereq" for how to define your own.
889
890 - PYTHON
891
892 Git wasn't compiled with NO_PYTHON=YesPlease. Wrap any tests that
893 need Python with this.
894
895 - PERL
896
897 Git wasn't compiled with NO_PERL=YesPlease.
898
899 Even without the PERL prerequisite, tests can assume there is a
900 usable perl interpreter at $PERL_PATH, though it need not be
901 particularly modern.
902
903 - POSIXPERM
904
905 The filesystem supports POSIX style permission bits.
906
907 - BSLASHPSPEC
908
909 Backslashes in pathspec are not directory separators. This is not
910 set on Windows. See 6fd1106a for details.
911
912 - EXECKEEPSPID
913
914 The process retains the same pid across exec(2). See fb9a2bea for
915 details.
916
917 - PIPE
918
919 The filesystem we're on supports creation of FIFOs (named pipes)
920 via mkfifo(1).
921
922 - SYMLINKS
923
924 The filesystem we're on supports symbolic links. E.g. a FAT
925 filesystem doesn't support these. See 704a3143 for details.
926
927 - SANITY
928
929 Test is not run by root user, and an attempt to write to an
930 unwritable file is expected to fail correctly.
931
932 - PCRE
933
934 Git was compiled with support for PCRE. Wrap any tests
935 that use git-grep --perl-regexp or git-grep -P in these.
936
937 - LIBPCRE1
938
939 Git was compiled with PCRE v1 support via
940 USE_LIBPCRE1=YesPlease. Wrap any PCRE using tests that for some
941 reason need v1 of the PCRE library instead of v2 in these.
942
943 - LIBPCRE2
944
945 Git was compiled with PCRE v2 support via
946 USE_LIBPCRE2=YesPlease. Wrap any PCRE using tests that for some
947 reason need v2 of the PCRE library instead of v1 in these.
948
949 - CASE_INSENSITIVE_FS
950
951 Test is run on a case insensitive file system.
952
953 - UTF8_NFD_TO_NFC
954
955 Test is run on a filesystem which converts decomposed utf-8 (nfd)
956 to precomposed utf-8 (nfc).
957
958 - PTHREADS
959
960 Git wasn't compiled with NO_PTHREADS=YesPlease.
961
962 Tips for Writing Tests
963 ----------------------
964
965 As with any programming projects, existing programs are the best
966 source of the information. However, do _not_ emulate
967 t0000-basic.sh when writing your tests. The test is special in
968 that it tries to validate the very core of GIT. For example, it
969 knows that there will be 256 subdirectories under .git/objects/,
970 and it knows that the object ID of an empty tree is a certain
971 40-byte string. This is deliberately done so in t0000-basic.sh
972 because the things the very basic core test tries to achieve is
973 to serve as a basis for people who are changing the GIT internal
974 drastically. For these people, after making certain changes,
975 not seeing failures from the basic test _is_ a failure. And
976 such drastic changes to the core GIT that even changes these
977 otherwise supposedly stable object IDs should be accompanied by
978 an update to t0000-basic.sh.
979
980 However, other tests that simply rely on basic parts of the core
981 GIT working properly should not have that level of intimate
982 knowledge of the core GIT internals. If all the test scripts
983 hardcoded the object IDs like t0000-basic.sh does, that defeats
984 the purpose of t0000-basic.sh, which is to isolate that level of
985 validation in one place. Your test also ends up needing
986 updating when such a change to the internal happens, so do _not_
987 do it and leave the low level of validation to t0000-basic.sh.
988
989 Test coverage
990 -------------
991
992 You can use the coverage tests to find code paths that are not being
993 used or properly exercised yet.
994
995 To do that, run the coverage target at the top-level (not in the t/
996 directory):
997
998 make coverage
999
1000 That'll compile Git with GCC's coverage arguments, and generate a test
1001 report with gcov after the tests finish. Running the coverage tests
1002 can take a while, since running the tests in parallel is incompatible
1003 with GCC's coverage mode.
1004
1005 After the tests have run you can generate a list of untested
1006 functions:
1007
1008 make coverage-untested-functions
1009
1010 You can also generate a detailed per-file HTML report using the
1011 Devel::Cover module. To install it do:
1012
1013 # On Debian or Ubuntu:
1014 sudo aptitude install libdevel-cover-perl
1015
1016 # From the CPAN with cpanminus
1017 curl -L http://cpanmin.us | perl - --sudo --self-upgrade
1018 cpanm --sudo Devel::Cover
1019
1020 Then, at the top-level:
1021
1022 make cover_db_html
1023
1024 That'll generate a detailed cover report in the "cover_db_html"
1025 directory, which you can then copy to a webserver, or inspect locally
1026 in a browser.