Merge branch 'dl/subtree-limit-to-one-rev'
[git/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 -V::
158 --verbose-log::
159 Write verbose output to the same logfile as `--tee`, but do
160 _not_ write it to stdout. Unlike `--tee --verbose`, this option
161 is safe to use when stdout is being consumed by a TAP parser
162 like `prove`. Implies `--tee` and `--verbose`.
163
164 --with-dashes::
165 By default tests are run without dashed forms of
166 commands (like git-commit) in the PATH (it only uses
167 wrappers from ../bin-wrappers). Use this option to include
168 the build directory (..) in the PATH, which contains all
169 the dashed forms of commands. This option is currently
170 implied by other options like --valgrind and
171 GIT_TEST_INSTALLED.
172
173 --no-bin-wrappers::
174 By default, the test suite uses the wrappers in
175 `../bin-wrappers/` to execute `git` and friends. With this option,
176 `../git` and friends are run directly. This is not recommended
177 in general, as the wrappers contain safeguards to ensure that no
178 files from an installed Git are used, but can speed up test runs
179 especially on platforms where running shell scripts is expensive
180 (most notably, Windows).
181
182 --root=<directory>::
183 Create "trash" directories used to store all temporary data during
184 testing under <directory>, instead of the t/ directory.
185 Using this option with a RAM-based filesystem (such as tmpfs)
186 can massively speed up the test suite.
187
188 --chain-lint::
189 --no-chain-lint::
190 If --chain-lint is enabled, the test harness will check each
191 test to make sure that it properly "&&-chains" all commands (so
192 that a failure in the middle does not go unnoticed by the final
193 exit code of the test). This check is performed in addition to
194 running the tests themselves. You may also enable or disable
195 this feature by setting the GIT_TEST_CHAIN_LINT environment
196 variable to "1" or "0", respectively.
197
198 --stress::
199 Run the test script repeatedly in multiple parallel jobs until
200 one of them fails. Useful for reproducing rare failures in
201 flaky tests. The number of parallel jobs is, in order of
202 precedence: the value of the GIT_TEST_STRESS_LOAD
203 environment variable, or twice the number of available
204 processors (as shown by the 'getconf' utility), or 8.
205 Implies `--verbose -x --immediate` to get the most information
206 about the failure. Note that the verbose output of each test
207 job is saved to 't/test-results/$TEST_NAME.stress-<nr>.out',
208 and only the output of the failed test job is shown on the
209 terminal. The names of the trash directories get a
210 '.stress-<nr>' suffix, and the trash directory of the failed
211 test job is renamed to end with a '.stress-failed' suffix.
212
213 --stress-jobs=<N>::
214 Override the number of parallel jobs. Implies `--stress`.
215
216 --stress-limit=<N>::
217 When combined with --stress run the test script repeatedly
218 this many times in each of the parallel jobs or until one of
219 them fails, whichever comes first. Implies `--stress`.
220
221 You can also set the GIT_TEST_INSTALLED environment variable to
222 the bindir of an existing git installation to test that installation.
223 You still need to have built this git sandbox, from which various
224 test-* support programs, templates, and perl libraries are used.
225 If your installed git is incomplete, it will silently test parts of
226 your built version instead.
227
228 When using GIT_TEST_INSTALLED, you can also set GIT_TEST_EXEC_PATH to
229 override the location of the dashed-form subcommands (what
230 GIT_EXEC_PATH would be used for during normal operation).
231 GIT_TEST_EXEC_PATH defaults to `$GIT_TEST_INSTALLED/git --exec-path`.
232
233
234 Skipping Tests
235 --------------
236
237 In some environments, certain tests have no way of succeeding
238 due to platform limitation, such as lack of 'unzip' program, or
239 filesystem that do not allow arbitrary sequence of non-NUL bytes
240 as pathnames.
241
242 You should be able to say something like
243
244 $ GIT_SKIP_TESTS=t9200.8 sh ./t9200-git-cvsexport-commit.sh
245
246 and even:
247
248 $ GIT_SKIP_TESTS='t[0-4]??? t91?? t9200.8' make
249
250 to omit such tests. The value of the environment variable is a
251 SP separated list of patterns that tells which tests to skip,
252 and either can match the "t[0-9]{4}" part to skip the whole
253 test, or t[0-9]{4} followed by ".$number" to say which
254 particular test to skip.
255
256 For an individual test suite --run could be used to specify that
257 only some tests should be run or that some tests should be
258 excluded from a run.
259
260 The argument for --run is a list of individual test numbers or
261 ranges with an optional negation prefix that define what tests in
262 a test suite to include in the run. A range is two numbers
263 separated with a dash and matches a range of tests with both ends
264 been included. You may omit the first or the second number to
265 mean "from the first test" or "up to the very last test"
266 respectively.
267
268 Optional prefix of '!' means that the test or a range of tests
269 should be excluded from the run.
270
271 If --run starts with an unprefixed number or range the initial
272 set of tests to run is empty. If the first item starts with '!'
273 all the tests are added to the initial set. After initial set is
274 determined every test number or range is added or excluded from
275 the set one by one, from left to right.
276
277 Individual numbers or ranges could be separated either by a space
278 or a comma.
279
280 For example, to run only tests up to a specific test (21), one
281 could do this:
282
283 $ sh ./t9200-git-cvsexport-commit.sh --run='1-21'
284
285 or this:
286
287 $ sh ./t9200-git-cvsexport-commit.sh --run='-21'
288
289 Common case is to run several setup tests (1, 2, 3) and then a
290 specific test (21) that relies on that setup:
291
292 $ sh ./t9200-git-cvsexport-commit.sh --run='1 2 3 21'
293
294 or:
295
296 $ sh ./t9200-git-cvsexport-commit.sh --run=1,2,3,21
297
298 or:
299
300 $ sh ./t9200-git-cvsexport-commit.sh --run='-3 21'
301
302 As noted above, the test set is built by going through the items
303 from left to right, so this:
304
305 $ sh ./t9200-git-cvsexport-commit.sh --run='1-4 !3'
306
307 will run tests 1, 2, and 4. Items that come later have higher
308 precedence. It means that this:
309
310 $ sh ./t9200-git-cvsexport-commit.sh --run='!3 1-4'
311
312 would just run tests from 1 to 4, including 3.
313
314 You may use negation with ranges. The following will run all
315 test in the test suite except from 7 up to 11:
316
317 $ sh ./t9200-git-cvsexport-commit.sh --run='!7-11'
318
319 Some tests in a test suite rely on the previous tests performing
320 certain actions, specifically some tests are designated as
321 "setup" test, so you cannot _arbitrarily_ disable one test and
322 expect the rest to function correctly.
323
324 --run is mostly useful when you want to focus on a specific test
325 and know what setup is needed for it. Or when you want to run
326 everything up to a certain test.
327
328
329 Running tests with special setups
330 ---------------------------------
331
332 The whole test suite could be run to test some special features
333 that cannot be easily covered by a few specific test cases. These
334 could be enabled by running the test suite with correct GIT_TEST_
335 environment set.
336
337 GIT_TEST_GETTEXT_POISON=<non-empty?> turns all strings marked for
338 translation into gibberish if non-empty (think "test -n"). Used for
339 spotting those tests that need to be marked with a C_LOCALE_OUTPUT
340 prerequisite when adding more strings for translation. See "Testing
341 marked strings" in po/README for details.
342
343 GIT_TEST_SPLIT_INDEX=<boolean> forces split-index mode on the whole
344 test suite. Accept any boolean values that are accepted by git-config.
345
346 GIT_TEST_PROTOCOL_VERSION=<n>, when set, overrides the
347 'protocol.version' setting to n if it is less than n.
348
349 GIT_TEST_FULL_IN_PACK_ARRAY=<boolean> exercises the uncommon
350 pack-objects code path where there are more than 1024 packs even if
351 the actual number of packs in repository is below this limit. Accept
352 any boolean values that are accepted by git-config.
353
354 GIT_TEST_OE_SIZE=<n> exercises the uncommon pack-objects code path
355 where we do not cache object size in memory and read it from existing
356 packs on demand. This normally only happens when the object size is
357 over 2GB. This variable forces the code path on any object larger than
358 <n> bytes.
359
360 GIT_TEST_OE_DELTA_SIZE=<n> exercises the uncommon pack-objects code
361 path where deltas larger than this limit require extra memory
362 allocation for bookkeeping.
363
364 GIT_TEST_VALIDATE_INDEX_CACHE_ENTRIES=<boolean> checks that cache-tree
365 records are valid when the index is written out or after a merge. This
366 is mostly to catch missing invalidation. Default is true.
367
368 GIT_TEST_COMMIT_GRAPH=<boolean>, when true, forces the commit-graph to
369 be written after every 'git commit' command, and overrides the
370 'core.commitGraph' setting to true.
371
372 GIT_TEST_FSMONITOR=$PWD/t7519/fsmonitor-all exercises the fsmonitor
373 code path for utilizing a file system monitor to speed up detecting
374 new or changed files.
375
376 GIT_TEST_INDEX_VERSION=<n> exercises the index read/write code path
377 for the index version specified. Can be set to any valid version
378 (currently 2, 3, or 4).
379
380 GIT_TEST_PACK_SPARSE=<boolean> if enabled will default the pack-objects
381 builtin to use the sparse object walk. This can still be overridden by
382 the --no-sparse command-line argument.
383
384 GIT_TEST_PRELOAD_INDEX=<boolean> exercises the preload-index code path
385 by overriding the minimum number of cache entries required per thread.
386
387 GIT_TEST_REBASE_USE_BUILTIN=<boolean>, when false, disables the
388 builtin version of git-rebase. See 'rebase.useBuiltin' in
389 git-config(1).
390
391 GIT_TEST_INDEX_THREADS=<n> enables exercising the multi-threaded loading
392 of the index for the whole test suite by bypassing the default number of
393 cache entries and thread minimums. Setting this to 1 will make the
394 index loading single threaded.
395
396 GIT_TEST_MULTI_PACK_INDEX=<boolean>, when true, forces the multi-pack-
397 index to be written after every 'git repack' command, and overrides the
398 'core.multiPackIndex' setting to true.
399
400 GIT_TEST_SIDEBAND_ALL=<boolean>, when true, overrides the
401 'uploadpack.allowSidebandAll' setting to true, and when false, forces
402 fetch-pack to not request sideband-all (even if the server advertises
403 sideband-all).
404
405 Naming Tests
406 ------------
407
408 The test files are named as:
409
410 tNNNN-commandname-details.sh
411
412 where N is a decimal digit.
413
414 First digit tells the family:
415
416 0 - the absolute basics and global stuff
417 1 - the basic commands concerning database
418 2 - the basic commands concerning the working tree
419 3 - the other basic commands (e.g. ls-files)
420 4 - the diff commands
421 5 - the pull and exporting commands
422 6 - the revision tree commands (even e.g. merge-base)
423 7 - the porcelainish commands concerning the working tree
424 8 - the porcelainish commands concerning forensics
425 9 - the git tools
426
427 Second digit tells the particular command we are testing.
428
429 Third digit (optionally) tells the particular switch or group of switches
430 we are testing.
431
432 If you create files under t/ directory (i.e. here) that is not
433 the top-level test script, never name the file to match the above
434 pattern. The Makefile here considers all such files as the
435 top-level test script and tries to run all of them. Care is
436 especially needed if you are creating a common test library
437 file, similar to test-lib.sh, because such a library file may
438 not be suitable for standalone execution.
439
440
441 Writing Tests
442 -------------
443
444 The test script is written as a shell script. It should start
445 with the standard "#!/bin/sh", and an
446 assignment to variable 'test_description', like this:
447
448 #!/bin/sh
449
450 test_description='xxx test (option --frotz)
451
452 This test registers the following structure in the cache
453 and tries to run git-ls-files with option --frotz.'
454
455
456 Source 'test-lib.sh'
457 --------------------
458
459 After assigning test_description, the test script should source
460 test-lib.sh like this:
461
462 . ./test-lib.sh
463
464 This test harness library does the following things:
465
466 - If the script is invoked with command line argument --help
467 (or -h), it shows the test_description and exits.
468
469 - Creates an empty test directory with an empty .git/objects database
470 and chdir(2) into it. This directory is 't/trash
471 directory.$test_name_without_dotsh', with t/ subject to change by
472 the --root option documented above, and a '.stress-<N>' suffix
473 appended by the --stress option.
474
475 - Defines standard test helper functions for your scripts to
476 use. These functions are designed to make all scripts behave
477 consistently when command line arguments --verbose (or -v),
478 --debug (or -d), and --immediate (or -i) is given.
479
480 Do's & don'ts
481 -------------
482
483 Here are a few examples of things you probably should and shouldn't do
484 when writing tests.
485
486 Here are the "do's:"
487
488 - Put all code inside test_expect_success and other assertions.
489
490 Even code that isn't a test per se, but merely some setup code
491 should be inside a test assertion.
492
493 - Chain your test assertions
494
495 Write test code like this:
496
497 git merge foo &&
498 git push bar &&
499 test ...
500
501 Instead of:
502
503 git merge hla
504 git push gh
505 test ...
506
507 That way all of the commands in your tests will succeed or fail. If
508 you must ignore the return value of something, consider using a
509 helper function (e.g. use sane_unset instead of unset, in order
510 to avoid unportable return value for unsetting a variable that was
511 already unset), or prepending the command with test_might_fail or
512 test_must_fail.
513
514 - Check the test coverage for your tests. See the "Test coverage"
515 below.
516
517 Don't blindly follow test coverage metrics; if a new function you added
518 doesn't have any coverage, then you're probably doing something wrong,
519 but having 100% coverage doesn't necessarily mean that you tested
520 everything.
521
522 Tests that are likely to smoke out future regressions are better
523 than tests that just inflate the coverage metrics.
524
525 - When a test checks for an absolute path that a git command generated,
526 construct the expected value using $(pwd) rather than $PWD,
527 $TEST_DIRECTORY, or $TRASH_DIRECTORY. It makes a difference on
528 Windows, where the shell (MSYS bash) mangles absolute path names.
529 For details, see the commit message of 4114156ae9.
530
531 - Remember that inside the <script> part, the standard output and
532 standard error streams are discarded, and the test harness only
533 reports "ok" or "not ok" to the end user running the tests. Under
534 --verbose, they are shown to help debug the tests.
535
536 And here are the "don'ts:"
537
538 - Don't exit() within a <script> part.
539
540 The harness will catch this as a programming error of the test.
541 Use test_done instead if you need to stop the tests early (see
542 "Skipping tests" below).
543
544 - Don't use '! git cmd' when you want to make sure the git command
545 exits with failure in a controlled way by calling "die()". Instead,
546 use 'test_must_fail git cmd'. This will signal a failure if git
547 dies in an unexpected way (e.g. segfault).
548
549 On the other hand, don't use test_must_fail for running regular
550 platform commands; just use '! cmd'. We are not in the business
551 of verifying that the world given to us sanely works.
552
553 - Don't feed the output of a git command to a pipe, as in:
554
555 git -C repo ls-files |
556 xargs -n 1 basename |
557 grep foo
558
559 which will discard git's exit code and may mask a crash. In the
560 above example, all exit codes are ignored except grep's.
561
562 Instead, write the output of that command to a temporary
563 file with ">" or assign it to a variable with "x=$(git ...)" rather
564 than pipe it.
565
566 - Don't use command substitution in a way that discards git's exit
567 code. When assigning to a variable, the exit code is not discarded,
568 e.g.:
569
570 x=$(git cat-file -p $sha) &&
571 ...
572
573 is OK because a crash in "git cat-file" will cause the "&&" chain
574 to fail, but:
575
576 test "refs/heads/foo" = "$(git symbolic-ref HEAD)"
577
578 is not OK and a crash in git could go undetected.
579
580 - Don't use perl without spelling it as "$PERL_PATH". This is to help
581 our friends on Windows where the platform Perl often adds CR before
582 the end of line, and they bundle Git with a version of Perl that
583 does not do so, whose path is specified with $PERL_PATH. Note that we
584 provide a "perl" function which uses $PERL_PATH under the hood, so
585 you do not need to worry when simply running perl in the test scripts
586 (but you do, for example, on a shebang line or in a sub script
587 created via "write_script").
588
589 - Don't use sh without spelling it as "$SHELL_PATH", when the script
590 can be misinterpreted by broken platform shell (e.g. Solaris).
591
592 - Don't chdir around in tests. It is not sufficient to chdir to
593 somewhere and then chdir back to the original location later in
594 the test, as any intermediate step can fail and abort the test,
595 causing the next test to start in an unexpected directory. Do so
596 inside a subshell if necessary.
597
598 - Don't save and verify the standard error of compound commands, i.e.
599 group commands, subshells, and shell functions (except test helper
600 functions like 'test_must_fail') like this:
601
602 ( cd dir && git cmd ) 2>error &&
603 test_cmp expect error
604
605 When running the test with '-x' tracing, then the trace of commands
606 executed in the compound command will be included in standard error
607 as well, quite possibly throwing off the subsequent checks examining
608 the output. Instead, save only the relevant git command's standard
609 error:
610
611 ( cd dir && git cmd 2>../error ) &&
612 test_cmp expect error
613
614 - Don't break the TAP output
615
616 The raw output from your test may be interpreted by a TAP harness. TAP
617 harnesses will ignore everything they don't know about, but don't step
618 on their toes in these areas:
619
620 - Don't print lines like "$x..$y" where $x and $y are integers.
621
622 - Don't print lines that begin with "ok" or "not ok".
623
624 TAP harnesses expect a line that begins with either "ok" and "not
625 ok" to signal a test passed or failed (and our harness already
626 produces such lines), so your script shouldn't emit such lines to
627 their output.
628
629 You can glean some further possible issues from the TAP grammar
630 (see https://metacpan.org/pod/TAP::Parser::Grammar#TAP-GRAMMAR)
631 but the best indication is to just run the tests with prove(1),
632 it'll complain if anything is amiss.
633
634
635 Skipping tests
636 --------------
637
638 If you need to skip tests you should do so by using the three-arg form
639 of the test_* functions (see the "Test harness library" section
640 below), e.g.:
641
642 test_expect_success PERL 'I need Perl' '
643 perl -e "hlagh() if unf_unf()"
644 '
645
646 The advantage of skipping tests like this is that platforms that don't
647 have the PERL and other optional dependencies get an indication of how
648 many tests they're missing.
649
650 If the test code is too hairy for that (i.e. does a lot of setup work
651 outside test assertions) you can also skip all remaining tests by
652 setting skip_all and immediately call test_done:
653
654 if ! test_have_prereq PERL
655 then
656 skip_all='skipping perl interface tests, perl not available'
657 test_done
658 fi
659
660 The string you give to skip_all will be used as an explanation for why
661 the test was skipped.
662
663 End with test_done
664 ------------------
665
666 Your script will be a sequence of tests, using helper functions
667 from the test harness library. At the end of the script, call
668 'test_done'.
669
670
671 Test harness library
672 --------------------
673
674 There are a handful helper functions defined in the test harness
675 library for your script to use.
676
677 - test_expect_success [<prereq>] <message> <script>
678
679 Usually takes two strings as parameters, and evaluates the
680 <script>. If it yields success, test is considered
681 successful. <message> should state what it is testing.
682
683 Example:
684
685 test_expect_success \
686 'git-write-tree should be able to write an empty tree.' \
687 'tree=$(git-write-tree)'
688
689 If you supply three parameters the first will be taken to be a
690 prerequisite; see the test_set_prereq and test_have_prereq
691 documentation below:
692
693 test_expect_success TTY 'git --paginate rev-list uses a pager' \
694 ' ... '
695
696 You can also supply a comma-separated list of prerequisites, in the
697 rare case where your test depends on more than one:
698
699 test_expect_success PERL,PYTHON 'yo dawg' \
700 ' test $(perl -E 'print eval "1 +" . qx[python -c "print 2"]') == "4" '
701
702 - test_expect_failure [<prereq>] <message> <script>
703
704 This is NOT the opposite of test_expect_success, but is used
705 to mark a test that demonstrates a known breakage. Unlike
706 the usual test_expect_success tests, which say "ok" on
707 success and "FAIL" on failure, this will say "FIXED" on
708 success and "still broken" on failure. Failures from these
709 tests won't cause -i (immediate) to stop.
710
711 Like test_expect_success this function can optionally use a three
712 argument invocation with a prerequisite as the first argument.
713
714 - test_debug <script>
715
716 This takes a single argument, <script>, and evaluates it only
717 when the test script is started with --debug command line
718 argument. This is primarily meant for use during the
719 development of a new test script.
720
721 - debug <git-command>
722
723 Run a git command inside a debugger. This is primarily meant for
724 use when debugging a failing test script.
725
726 - test_done
727
728 Your test script must have test_done at the end. Its purpose
729 is to summarize successes and failures in the test script and
730 exit with an appropriate error code.
731
732 - test_tick
733
734 Make commit and tag names consistent by setting the author and
735 committer times to defined state. Subsequent calls will
736 advance the times by a fixed amount.
737
738 - test_commit <message> [<filename> [<contents>]]
739
740 Creates a commit with the given message, committing the given
741 file with the given contents (default for both is to reuse the
742 message string), and adds a tag (again reusing the message
743 string as name). Calls test_tick to make the SHA-1s
744 reproducible.
745
746 - test_merge <message> <commit-or-tag>
747
748 Merges the given rev using the given message. Like test_commit,
749 creates a tag and calls test_tick before committing.
750
751 - test_set_prereq <prereq>
752
753 Set a test prerequisite to be used later with test_have_prereq. The
754 test-lib will set some prerequisites for you, see the
755 "Prerequisites" section below for a full list of these.
756
757 Others you can set yourself and use later with either
758 test_have_prereq directly, or the three argument invocation of
759 test_expect_success and test_expect_failure.
760
761 - test_have_prereq <prereq>
762
763 Check if we have a prerequisite previously set with test_set_prereq.
764 The most common way to use this explicitly (as opposed to the
765 implicit use when an argument is passed to test_expect_*) is to skip
766 all the tests at the start of the test script if we don't have some
767 essential prerequisite:
768
769 if ! test_have_prereq PERL
770 then
771 skip_all='skipping perl interface tests, perl not available'
772 test_done
773 fi
774
775 - test_external [<prereq>] <message> <external> <script>
776
777 Execute a <script> with an <external> interpreter (like perl). This
778 was added for tests like t9700-perl-git.sh which do most of their
779 work in an external test script.
780
781 test_external \
782 'GitwebCache::*FileCache*' \
783 perl "$TEST_DIRECTORY"/t9503/test_cache_interface.pl
784
785 If the test is outputting its own TAP you should set the
786 test_external_has_tap variable somewhere before calling the first
787 test_external* function. See t9700-perl-git.sh for an example.
788
789 # The external test will outputs its own plan
790 test_external_has_tap=1
791
792 - test_external_without_stderr [<prereq>] <message> <external> <script>
793
794 Like test_external but fail if there's any output on stderr,
795 instead of checking the exit code.
796
797 test_external_without_stderr \
798 'Perl API' \
799 perl "$TEST_DIRECTORY"/t9700/test.pl
800
801 - test_expect_code <exit-code> <command>
802
803 Run a command and ensure that it exits with the given exit code.
804 For example:
805
806 test_expect_success 'Merge with d/f conflicts' '
807 test_expect_code 1 git merge "merge msg" B master
808 '
809
810 - test_must_fail [<options>] <git-command>
811
812 Run a git command and ensure it fails in a controlled way. Use
813 this instead of "! <git-command>". When git-command dies due to a
814 segfault, test_must_fail diagnoses it as an error; "! <git-command>"
815 treats it as just another expected failure, which would let such a
816 bug go unnoticed.
817
818 Accepts the following options:
819
820 ok=<signal-name>[,<...>]:
821 Don't treat an exit caused by the given signal as error.
822 Multiple signals can be specified as a comma separated list.
823 Currently recognized signal names are: sigpipe, success.
824 (Don't use 'success', use 'test_might_fail' instead.)
825
826 - test_might_fail [<options>] <git-command>
827
828 Similar to test_must_fail, but tolerate success, too. Use this
829 instead of "<git-command> || :" to catch failures due to segv.
830
831 Accepts the same options as test_must_fail.
832
833 - test_cmp <expected> <actual>
834
835 Check whether the content of the <actual> file matches the
836 <expected> file. This behaves like "cmp" but produces more
837 helpful output when the test is run with "-v" option.
838
839 - test_cmp_rev <expected> <actual>
840
841 Check whether the <expected> rev points to the same commit as the
842 <actual> rev.
843
844 - test_line_count (= | -lt | -ge | ...) <length> <file>
845
846 Check whether a file has the length it is expected to.
847
848 - test_path_is_file <path> [<diagnosis>]
849 test_path_is_dir <path> [<diagnosis>]
850 test_path_is_missing <path> [<diagnosis>]
851
852 Check if the named path is a file, if the named path is a
853 directory, or if the named path does not exist, respectively,
854 and fail otherwise, showing the <diagnosis> text.
855
856 - test_when_finished <script>
857
858 Prepend <script> to a list of commands to run to clean up
859 at the end of the current test. If some clean-up command
860 fails, the test will not pass.
861
862 Example:
863
864 test_expect_success 'branch pointing to non-commit' '
865 git rev-parse HEAD^{tree} >.git/refs/heads/invalid &&
866 test_when_finished "git update-ref -d refs/heads/invalid" &&
867 ...
868 '
869
870 - test_write_lines <lines>
871
872 Write <lines> on standard output, one line per argument.
873 Useful to prepare multi-line files in a compact form.
874
875 Example:
876
877 test_write_lines a b c d e f g >foo
878
879 Is a more compact equivalent of:
880 cat >foo <<-EOF
881 a
882 b
883 c
884 d
885 e
886 f
887 g
888 EOF
889
890
891 - test_pause
892
893 This command is useful for writing and debugging tests and must be
894 removed before submitting. It halts the execution of the test and
895 spawns a shell in the trash directory. Exit the shell to continue
896 the test. Example:
897
898 test_expect_success 'test' '
899 git do-something >actual &&
900 test_pause &&
901 test_cmp expected actual
902 '
903
904 - test_ln_s_add <path1> <path2>
905
906 This function helps systems whose filesystem does not support symbolic
907 links. Use it to add a symbolic link entry to the index when it is not
908 important that the file system entry is a symbolic link, i.e., instead
909 of the sequence
910
911 ln -s foo bar &&
912 git add bar
913
914 Sometimes it is possible to split a test in a part that does not need
915 the symbolic link in the file system and a part that does; then only
916 the latter part need be protected by a SYMLINKS prerequisite (see below).
917
918 - test_oid_init
919
920 This function loads facts and useful object IDs related to the hash
921 algorithm(s) in use from the files in t/oid-info.
922
923 - test_oid_cache
924
925 This function reads per-hash algorithm information from standard
926 input (usually a heredoc) in the format described in
927 t/oid-info/README. This is useful for test-specific values, such as
928 object IDs, which must vary based on the hash algorithm.
929
930 Certain fixed values, such as hash sizes and common placeholder
931 object IDs, can be loaded with test_oid_init (described above).
932
933 - test_oid <key>
934
935 This function looks up a value for the hash algorithm in use, based
936 on the key given. The value must have been loaded using
937 test_oid_init or test_oid_cache. Providing an unknown key is an
938 error.
939
940 - yes [<string>]
941
942 This is often seen in modern UNIX but some platforms lack it, so
943 the test harness overrides the platform implementation with a
944 more limited one. Use this only when feeding a handful lines of
945 output to the downstream---unlike the real version, it generates
946 only up to 99 lines.
947
948
949 Prerequisites
950 -------------
951
952 These are the prerequisites that the test library predefines with
953 test_have_prereq.
954
955 See the prereq argument to the test_* functions in the "Test harness
956 library" section above and the "test_have_prereq" function for how to
957 use these, and "test_set_prereq" for how to define your own.
958
959 - PYTHON
960
961 Git wasn't compiled with NO_PYTHON=YesPlease. Wrap any tests that
962 need Python with this.
963
964 - PERL
965
966 Git wasn't compiled with NO_PERL=YesPlease.
967
968 Even without the PERL prerequisite, tests can assume there is a
969 usable perl interpreter at $PERL_PATH, though it need not be
970 particularly modern.
971
972 - POSIXPERM
973
974 The filesystem supports POSIX style permission bits.
975
976 - BSLASHPSPEC
977
978 Backslashes in pathspec are not directory separators. This is not
979 set on Windows. See 6fd1106a for details.
980
981 - EXECKEEPSPID
982
983 The process retains the same pid across exec(2). See fb9a2bea for
984 details.
985
986 - PIPE
987
988 The filesystem we're on supports creation of FIFOs (named pipes)
989 via mkfifo(1).
990
991 - SYMLINKS
992
993 The filesystem we're on supports symbolic links. E.g. a FAT
994 filesystem doesn't support these. See 704a3143 for details.
995
996 - SANITY
997
998 Test is not run by root user, and an attempt to write to an
999 unwritable file is expected to fail correctly.
1000
1001 - PCRE
1002
1003 Git was compiled with support for PCRE. Wrap any tests
1004 that use git-grep --perl-regexp or git-grep -P in these.
1005
1006 - LIBPCRE1
1007
1008 Git was compiled with PCRE v1 support via
1009 USE_LIBPCRE1=YesPlease. Wrap any PCRE using tests that for some
1010 reason need v1 of the PCRE library instead of v2 in these.
1011
1012 - LIBPCRE2
1013
1014 Git was compiled with PCRE v2 support via
1015 USE_LIBPCRE2=YesPlease. Wrap any PCRE using tests that for some
1016 reason need v2 of the PCRE library instead of v1 in these.
1017
1018 - CASE_INSENSITIVE_FS
1019
1020 Test is run on a case insensitive file system.
1021
1022 - UTF8_NFD_TO_NFC
1023
1024 Test is run on a filesystem which converts decomposed utf-8 (nfd)
1025 to precomposed utf-8 (nfc).
1026
1027 - PTHREADS
1028
1029 Git wasn't compiled with NO_PTHREADS=YesPlease.
1030
1031 Tips for Writing Tests
1032 ----------------------
1033
1034 As with any programming projects, existing programs are the best
1035 source of the information. However, do _not_ emulate
1036 t0000-basic.sh when writing your tests. The test is special in
1037 that it tries to validate the very core of GIT. For example, it
1038 knows that there will be 256 subdirectories under .git/objects/,
1039 and it knows that the object ID of an empty tree is a certain
1040 40-byte string. This is deliberately done so in t0000-basic.sh
1041 because the things the very basic core test tries to achieve is
1042 to serve as a basis for people who are changing the GIT internal
1043 drastically. For these people, after making certain changes,
1044 not seeing failures from the basic test _is_ a failure. And
1045 such drastic changes to the core GIT that even changes these
1046 otherwise supposedly stable object IDs should be accompanied by
1047 an update to t0000-basic.sh.
1048
1049 However, other tests that simply rely on basic parts of the core
1050 GIT working properly should not have that level of intimate
1051 knowledge of the core GIT internals. If all the test scripts
1052 hardcoded the object IDs like t0000-basic.sh does, that defeats
1053 the purpose of t0000-basic.sh, which is to isolate that level of
1054 validation in one place. Your test also ends up needing
1055 updating when such a change to the internal happens, so do _not_
1056 do it and leave the low level of validation to t0000-basic.sh.
1057
1058 Test coverage
1059 -------------
1060
1061 You can use the coverage tests to find code paths that are not being
1062 used or properly exercised yet.
1063
1064 To do that, run the coverage target at the top-level (not in the t/
1065 directory):
1066
1067 make coverage
1068
1069 That'll compile Git with GCC's coverage arguments, and generate a test
1070 report with gcov after the tests finish. Running the coverage tests
1071 can take a while, since running the tests in parallel is incompatible
1072 with GCC's coverage mode.
1073
1074 After the tests have run you can generate a list of untested
1075 functions:
1076
1077 make coverage-untested-functions
1078
1079 You can also generate a detailed per-file HTML report using the
1080 Devel::Cover module. To install it do:
1081
1082 # On Debian or Ubuntu:
1083 sudo aptitude install libdevel-cover-perl
1084
1085 # From the CPAN with cpanminus
1086 curl -L http://cpanmin.us | perl - --sudo --self-upgrade
1087 cpanm --sudo Devel::Cover
1088
1089 Then, at the top-level:
1090
1091 make cover_db_html
1092
1093 That'll generate a detailed cover report in the "cover_db_html"
1094 directory, which you can then copy to a webserver, or inspect locally
1095 in a browser.