test-lib: verbose mode for only tests matching a pattern
[git/git.git] / t / README
CommitLineData
986aa7f1
JH
1Core GIT Tests
2==============
3
4This directory holds many test scripts for core GIT tools. The
5first part of this short document describes how to run the tests
6and read their output.
7
8When fixing the tools or adding enhancements, you are strongly
9encouraged to add tests in this directory to cover what you are
10trying to fix or enhance. The later part of this short document
11describes how your test scripts should be organized.
12
13
14Running Tests
15-------------
16
17The easiest way to run tests is to say "make". This runs all
18the tests.
19
20 *** t0000-basic.sh ***
5099b99d
ÆAB
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
986aa7f1 24 ...
5099b99d
ÆAB
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
35Since the tests all output TAP (see http://testanything.org) they can
85b0b34e 36be run with any TAP harness. Here's an example of parallel testing
5099b99d
ÆAB
37powered 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
47prove 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
986aa7f1 52
28d836c8
MG
53You can give DEFAULT_TEST_TARGET=prove on the make command (or define it
54in config.mak) to cause "make test" to run tests under prove.
55GIT_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
5099b99d 59You can also run each test individually from command line, like this:
986aa7f1 60
5099b99d
ÆAB
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
986aa7f1
JH
69
70You can pass --verbose (or -v), --debug (or -d), and --immediate
4e1be63c
JS
71(or -i) command line argument to the test, or by setting GIT_TEST_OPTS
72appropriately before running "make".
986aa7f1
JH
73
74--verbose::
75 This makes the test more verbose. Specifically, the
76 command being run and their output if any are also
77 output.
78
ff09af3f
TR
79--verbose-only=<pattern>::
80 Like --verbose, but the effect is limited to tests with
81 numbers matching <pattern>. The number matched against is
82 simply the running count of the test within the file.
83
986aa7f1
JH
84--debug::
85 This may help the person who is developing a new test.
86 It causes the command defined with test_debug to run.
0986de94
PK
87 The "trash" directory (used to store all temporary data
88 during testing) is not deleted even if there are no
89 failed tests so that you can inspect its contents after
90 the test finished.
986aa7f1
JH
91
92--immediate::
93 This causes the test to immediately exit upon the first
13cb3bb7
SR
94 failed test. Cleanup commands requested with
95 test_when_finished are not executed if the test failed,
96 in order to keep the state for inspection by the tester
97 to diagnose the bug.
986aa7f1 98
5e2c08c6
LW
99--long-tests::
100 This causes additional long-running tests to be run (where
101 available), for more exhaustive testing.
102
952af351
TR
103--valgrind=<tool>::
104 Execute all Git binaries under valgrind tool <tool> and exit
105 with status 126 on errors (just like regular tests, this will
106 only stop the test script when running under -i).
986aa7f1 107
3da93652
JS
108 Since it makes no sense to run the tests with --valgrind and
109 not see any output, this option implies --verbose. For
110 convenience, it also implies --tee.
111
952af351
TR
112 <tool> defaults to 'memcheck', just like valgrind itself.
113 Other particularly useful choices include 'helgrind' and
114 'drd', but you may use any tool recognized by your valgrind
115 installation.
116
95d9d5ec
TR
117 As a special case, <tool> can be 'memcheck-fast', which uses
118 memcheck but disables --track-origins. Use this if you are
119 running tests in bulk, to see if there are _any_ memory
120 issues.
121
952af351 122 Note that memcheck is run with the option --leak-check=no,
9aec68d3
CMN
123 as the git process is short-lived and some errors are not
124 interesting. In order to run a single command under the same
125 conditions manually, you should set GIT_VALGRIND to point to
126 the 't/valgrind/' directory and use the commands under
127 't/valgrind/bin/'.
128
44138559
JS
129--tee::
130 In addition to printing the test output to the terminal,
131 write it to files named 't/test-results/$TEST_NAME.out'.
132 As the names depend on the tests' file names, it is safe to
133 run the tests with this option in parallel.
134
e4597aae
MO
135--with-dashes::
136 By default tests are run without dashed forms of
137 commands (like git-commit) in the PATH (it only uses
138 wrappers from ../bin-wrappers). Use this option to include
139 the build directory (..) in the PATH, which contains all
140 the dashed forms of commands. This option is currently
141 implied by other options like --valgrind and
142 GIT_TEST_INSTALLED.
143
0d4dbcd3
TR
144--root=<directory>::
145 Create "trash" directories used to store all temporary data during
146 testing under <directory>, instead of the t/ directory.
147 Using this option with a RAM-based filesystem (such as tmpfs)
148 can massively speed up the test suite.
149
e160da7f
MO
150You can also set the GIT_TEST_INSTALLED environment variable to
151the bindir of an existing git installation to test that installation.
152You still need to have built this git sandbox, from which various
153test-* support programs, templates, and perl libraries are used.
154If your installed git is incomplete, it will silently test parts of
155your built version instead.
156
157When using GIT_TEST_INSTALLED, you can also set GIT_TEST_EXEC_PATH to
158override the location of the dashed-form subcommands (what
159GIT_EXEC_PATH would be used for during normal operation).
160GIT_TEST_EXEC_PATH defaults to `$GIT_TEST_INSTALLED/git --exec-path`.
161
162
fbd458a3
JN
163Skipping Tests
164--------------
165
166In some environments, certain tests have no way of succeeding
167due to platform limitation, such as lack of 'unzip' program, or
168filesystem that do not allow arbitrary sequence of non-NUL bytes
169as pathnames.
170
171You should be able to say something like
172
173 $ GIT_SKIP_TESTS=t9200.8 sh ./t9200-git-cvsexport-commit.sh
174
175and even:
176
177 $ GIT_SKIP_TESTS='t[0-4]??? t91?? t9200.8' make
178
179to omit such tests. The value of the environment variable is a
180SP separated list of patterns that tells which tests to skip,
181and either can match the "t[0-9]{4}" part to skip the whole
182test, or t[0-9]{4} followed by ".$number" to say which
183particular test to skip.
184
185Note that some tests in the existing test suite rely on previous
186test item, so you cannot arbitrarily disable one and expect the
187remainder of test to check what the test originally was intended
188to check.
189
190
f50c9f76
PB
191Naming Tests
192------------
193
194The test files are named as:
195
196 tNNNN-commandname-details.sh
197
198where N is a decimal digit.
199
200First digit tells the family:
201
202 0 - the absolute basics and global stuff
203 1 - the basic commands concerning database
204 2 - the basic commands concerning the working tree
205 3 - the other basic commands (e.g. ls-files)
206 4 - the diff commands
207 5 - the pull and exporting commands
208 6 - the revision tree commands (even e.g. merge-base)
8f4a9b62 209 7 - the porcelainish commands concerning the working tree
8757749e
JN
210 8 - the porcelainish commands concerning forensics
211 9 - the git tools
f50c9f76
PB
212
213Second digit tells the particular command we are testing.
214
215Third digit (optionally) tells the particular switch or group of switches
216we are testing.
217
77656600
JH
218If you create files under t/ directory (i.e. here) that is not
219the top-level test script, never name the file to match the above
220pattern. The Makefile here considers all such files as the
63d32945 221top-level test script and tries to run all of them. Care is
77656600
JH
222especially needed if you are creating a common test library
223file, similar to test-lib.sh, because such a library file may
224not be suitable for standalone execution.
225
f50c9f76 226
986aa7f1
JH
227Writing Tests
228-------------
229
230The test script is written as a shell script. It should start
231with the standard "#!/bin/sh" with copyright notices, and an
232assignment to variable 'test_description', like this:
233
234 #!/bin/sh
235 #
236 # Copyright (c) 2005 Junio C Hamano
237 #
238
14cd1ff3 239 test_description='xxx test (option --frotz)
986aa7f1
JH
240
241 This test registers the following structure in the cache
242 and tries to run git-ls-files with option --frotz.'
243
f50c9f76 244
986aa7f1
JH
245Source 'test-lib.sh'
246--------------------
247
248After assigning test_description, the test script should source
249test-lib.sh like this:
250
251 . ./test-lib.sh
252
253This test harness library does the following things:
254
255 - If the script is invoked with command line argument --help
256 (or -h), it shows the test_description and exits.
257
e1ca1c9d
ÆAB
258 - Creates an empty test directory with an empty .git/objects database
259 and chdir(2) into it. This directory is 't/trash
260 directory.$test_name_without_dotsh', with t/ subject to change by
261 the --root option documented above.
986aa7f1
JH
262
263 - Defines standard test helper functions for your scripts to
264 use. These functions are designed to make all scripts behave
265 consistently when command line arguments --verbose (or -v),
266 --debug (or -d), and --immediate (or -i) is given.
267
20873f45
ÆAB
268Do's, don'ts & things to keep in mind
269-------------------------------------
270
6fd45295 271Here are a few examples of things you probably should and shouldn't do
20873f45
ÆAB
272when writing tests.
273
274Do:
275
6fd45295 276 - Put all code inside test_expect_success and other assertions.
20873f45
ÆAB
277
278 Even code that isn't a test per se, but merely some setup code
6fd45295 279 should be inside a test assertion.
20873f45
ÆAB
280
281 - Chain your test assertions
282
283 Write test code like this:
284
285 git merge foo &&
286 git push bar &&
287 test ...
288
289 Instead of:
290
291 git merge hla
292 git push gh
293 test ...
294
295 That way all of the commands in your tests will succeed or fail. If
00648ba0
EN
296 you must ignore the return value of something, consider using a
297 helper function (e.g. use sane_unset instead of unset, in order
298 to avoid unportable return value for unsetting a variable that was
299 already unset), or prepending the command with test_might_fail or
300 test_must_fail.
20873f45 301
0c357544
ÆAB
302 - Check the test coverage for your tests. See the "Test coverage"
303 below.
304
63d32945
MW
305 Don't blindly follow test coverage metrics; if a new function you added
306 doesn't have any coverage, then you're probably doing something wrong,
e8b55f5c
ÆAB
307 but having 100% coverage doesn't necessarily mean that you tested
308 everything.
309
310 Tests that are likely to smoke out future regressions are better
311 than tests that just inflate the coverage metrics.
312
95b104c8
JS
313 - When a test checks for an absolute path that a git command generated,
314 construct the expected value using $(pwd) rather than $PWD,
315 $TEST_DIRECTORY, or $TRASH_DIRECTORY. It makes a difference on
316 Windows, where the shell (MSYS bash) mangles absolute path names.
317 For details, see the commit message of 4114156ae9.
318
20873f45
ÆAB
319Don't:
320
321 - exit() within a <script> part.
322
323 The harness will catch this as a programming error of the test.
324 Use test_done instead if you need to stop the tests early (see
325 "Skipping tests" below).
326
ad78585e
JH
327 - use '! git cmd' when you want to make sure the git command exits
328 with failure in a controlled way by calling "die()". Instead,
329 use 'test_must_fail git cmd'. This will signal a failure if git
330 dies in an unexpected way (e.g. segfault).
331
332 - use perl without spelling it as "$PERL_PATH". This is to help our
333 friends on Windows where the platform Perl often adds CR before
334 the end of line, and they bundle Git with a version of Perl that
335 does not do so, whose path is specified with $PERL_PATH.
336
337 - use sh without spelling it as "$SHELL_PATH", when the script can
338 be misinterpreted by broken platform shell (e.g. Solaris).
339
340 - chdir around in tests. It is not sufficient to chdir to
341 somewhere and then chdir back to the original location later in
342 the test, as any intermediate step can fail and abort the test,
343 causing the next test to start in an unexpected directory. Do so
344 inside a subshell if necessary.
345
20873f45
ÆAB
346 - Break the TAP output
347
6fd45295
JH
348 The raw output from your test may be interpreted by a TAP harness. TAP
349 harnesses will ignore everything they don't know about, but don't step
350 on their toes in these areas:
20873f45
ÆAB
351
352 - Don't print lines like "$x..$y" where $x and $y are integers.
353
354 - Don't print lines that begin with "ok" or "not ok".
355
6fd45295 356 TAP harnesses expect a line that begins with either "ok" and "not
20873f45
ÆAB
357 ok" to signal a test passed or failed (and our harness already
358 produces such lines), so your script shouldn't emit such lines to
359 their output.
360
361 You can glean some further possible issues from the TAP grammar
362 (see http://search.cpan.org/perldoc?TAP::Parser::Grammar#TAP_Grammar)
363 but the best indication is to just run the tests with prove(1),
364 it'll complain if anything is amiss.
365
366Keep in mind:
367
6fd45295 368 - Inside <script> part, the standard output and standard error
20873f45
ÆAB
369 streams are discarded, and the test harness only reports "ok" or
370 "not ok" to the end user running the tests. Under --verbose, they
371 are shown to help debugging the tests.
372
373
b5500d16
ÆAB
374Skipping tests
375--------------
376
681186ae 377If you need to skip tests you should do so by using the three-arg form
99d9050d
ÆAB
378of the test_* functions (see the "Test harness library" section
379below), e.g.:
380
ad78585e
JH
381 test_expect_success PERL 'I need Perl' '
382 "$PERL_PATH" -e "hlagh() if unf_unf()"
383 '
99d9050d
ÆAB
384
385The advantage of skipping tests like this is that platforms that don't
386have the PERL and other optional dependencies get an indication of how
387many tests they're missing.
388
389If the test code is too hairy for that (i.e. does a lot of setup work
390outside test assertions) you can also skip all remaining tests by
391setting skip_all and immediately call test_done:
b5500d16
ÆAB
392
393 if ! test_have_prereq PERL
394 then
395 skip_all='skipping perl interface tests, perl not available'
396 test_done
397 fi
14cd1ff3 398
99d9050d
ÆAB
399The string you give to skip_all will be used as an explanation for why
400the test was skipped.
401
986aa7f1
JH
402End with test_done
403------------------
404
405Your script will be a sequence of tests, using helper functions
406from the test harness library. At the end of the script, call
407'test_done'.
408
409
410Test harness library
411--------------------
412
413There are a handful helper functions defined in the test harness
414library for your script to use.
415
9a897893 416 - test_expect_success [<prereq>] <message> <script>
986aa7f1 417
72942a61 418 Usually takes two strings as parameters, and evaluates the
986aa7f1
JH
419 <script>. If it yields success, test is considered
420 successful. <message> should state what it is testing.
421
422 Example:
423
424 test_expect_success \
425 'git-write-tree should be able to write an empty tree.' \
426 'tree=$(git-write-tree)'
427
9a897893 428 If you supply three parameters the first will be taken to be a
72942a61 429 prerequisite; see the test_set_prereq and test_have_prereq
9a897893
ÆAB
430 documentation below:
431
432 test_expect_success TTY 'git --paginate rev-list uses a pager' \
433 ' ... '
434
93a57246
ÆAB
435 You can also supply a comma-separated list of prerequisites, in the
436 rare case where your test depends on more than one:
437
438 test_expect_success PERL,PYTHON 'yo dawg' \
439 ' test $(perl -E 'print eval "1 +" . qx[python -c "print 2"]') == "4" '
440
9a897893 441 - test_expect_failure [<prereq>] <message> <script>
986aa7f1 442
41ac414e
JH
443 This is NOT the opposite of test_expect_success, but is used
444 to mark a test that demonstrates a known breakage. Unlike
445 the usual test_expect_success tests, which say "ok" on
446 success and "FAIL" on failure, this will say "FIXED" on
447 success and "still broken" on failure. Failures from these
448 tests won't cause -i (immediate) to stop.
986aa7f1 449
9a897893
ÆAB
450 Like test_expect_success this function can optionally use a three
451 argument invocation with a prerequisite as the first argument.
452
986aa7f1
JH
453 - test_debug <script>
454
455 This takes a single argument, <script>, and evaluates it only
456 when the test script is started with --debug command line
457 argument. This is primarily meant for use during the
458 development of a new test script.
459
460 - test_done
461
462 Your test script must have test_done at the end. Its purpose
463 is to summarize successes and failures in the test script and
464 exit with an appropriate error code.
465
00884968
JS
466 - test_tick
467
468 Make commit and tag names consistent by setting the author and
63d32945 469 committer times to defined state. Subsequent calls will
00884968
JS
470 advance the times by a fixed amount.
471
472 - test_commit <message> [<filename> [<contents>]]
473
474 Creates a commit with the given message, committing the given
475 file with the given contents (default for both is to reuse the
476 message string), and adds a tag (again reusing the message
477 string as name). Calls test_tick to make the SHA-1s
478 reproducible.
479
480 - test_merge <message> <commit-or-tag>
481
482 Merges the given rev using the given message. Like test_commit,
483 creates a tag and calls test_tick before committing.
986aa7f1 484
72942a61 485 - test_set_prereq <prereq>
9a897893
ÆAB
486
487 Set a test prerequisite to be used later with test_have_prereq. The
be53deef
ÆAB
488 test-lib will set some prerequisites for you, see the
489 "Prerequisites" section below for a full list of these.
490
491 Others you can set yourself and use later with either
492 test_have_prereq directly, or the three argument invocation of
493 test_expect_success and test_expect_failure.
9a897893 494
72942a61 495 - test_have_prereq <prereq>
9a897893
ÆAB
496
497 Check if we have a prerequisite previously set with
498 test_set_prereq. The most common use of this directly is to skip
499 all the tests if we don't have some essential prerequisite:
500
501 if ! test_have_prereq PERL
502 then
503 skip_all='skipping perl interface tests, perl not available'
504 test_done
505 fi
506
2fac6a4b
ÆAB
507 - test_external [<prereq>] <message> <external> <script>
508
509 Execute a <script> with an <external> interpreter (like perl). This
510 was added for tests like t9700-perl-git.sh which do most of their
511 work in an external test script.
512
513 test_external \
514 'GitwebCache::*FileCache*' \
515 "$PERL_PATH" "$TEST_DIRECTORY"/t9503/test_cache_interface.pl
516
517 If the test is outputting its own TAP you should set the
518 test_external_has_tap variable somewhere before calling the first
519 test_external* function. See t9700-perl-git.sh for an example.
520
521 # The external test will outputs its own plan
522 test_external_has_tap=1
523
524 - test_external_without_stderr [<prereq>] <message> <external> <script>
525
526 Like test_external but fail if there's any output on stderr,
527 instead of checking the exit code.
528
529 test_external_without_stderr \
530 'Perl API' \
531 "$PERL_PATH" "$TEST_DIRECTORY"/t9700/test.pl
532
892e6f7e
ÆAB
533 - test_expect_code <exit-code> <command>
534
535 Run a command and ensure that it exits with the given exit code.
536 For example:
537
538 test_expect_success 'Merge with d/f conflicts' '
539 test_expect_code 1 git merge "merge msg" B master
540 '
541
c9667456
JN
542 - test_must_fail <git-command>
543
544 Run a git command and ensure it fails in a controlled way. Use
971ecbd1
BC
545 this instead of "! <git-command>". When git-command dies due to a
546 segfault, test_must_fail diagnoses it as an error; "! <git-command>"
547 treats it as just another expected failure, which would let such a
548 bug go unnoticed.
c9667456
JN
549
550 - test_might_fail <git-command>
551
552 Similar to test_must_fail, but tolerate success, too. Use this
553 instead of "<git-command> || :" to catch failures due to segv.
554
555 - test_cmp <expected> <actual>
556
557 Check whether the content of the <actual> file matches the
558 <expected> file. This behaves like "cmp" but produces more
559 helpful output when the test is run with "-v" option.
560
fb3340a6
JN
561 - test_line_count (= | -lt | -ge | ...) <length> <file>
562
563 Check whether a file has the length it is expected to.
564
72942a61
ML
565 - test_path_is_file <path> [<diagnosis>]
566 test_path_is_dir <path> [<diagnosis>]
2caf20c5
MM
567 test_path_is_missing <path> [<diagnosis>]
568
72942a61
ML
569 Check if the named path is a file, if the named path is a
570 directory, or if the named path does not exist, respectively,
571 and fail otherwise, showing the <diagnosis> text.
2caf20c5 572
c9667456
JN
573 - test_when_finished <script>
574
575 Prepend <script> to a list of commands to run to clean up
576 at the end of the current test. If some clean-up command
577 fails, the test will not pass.
578
579 Example:
580
581 test_expect_success 'branch pointing to non-commit' '
582 git rev-parse HEAD^{tree} >.git/refs/heads/invalid &&
583 test_when_finished "git update-ref -d refs/heads/invalid" &&
584 ...
585 '
586
c4d2539a
JL
587 - test_pause
588
589 This command is useful for writing and debugging tests and must be
590 removed before submitting. It halts the execution of the test and
591 spawns a shell in the trash directory. Exit the shell to continue
592 the test. Example:
593
594 test_expect_success 'test' '
595 git do-something >actual &&
596 test_pause &&
597 test_cmp expected actual
598 '
599
be53deef
ÆAB
600Prerequisites
601-------------
602
603These are the prerequisites that the test library predefines with
604test_have_prereq.
605
606See the prereq argument to the test_* functions in the "Test harness
607library" section above and the "test_have_prereq" function for how to
608use these, and "test_set_prereq" for how to define your own.
609
610 - PERL & PYTHON
611
612 Git wasn't compiled with NO_PERL=YesPlease or
613 NO_PYTHON=YesPlease. Wrap any tests that need Perl or Python in
614 these.
615
616 - POSIXPERM
617
618 The filesystem supports POSIX style permission bits.
619
620 - BSLASHPSPEC
621
622 Backslashes in pathspec are not directory separators. This is not
623 set on Windows. See 6fd1106a for details.
624
625 - EXECKEEPSPID
626
627 The process retains the same pid across exec(2). See fb9a2bea for
628 details.
629
20073274
AS
630 - PIPE
631
632 The filesystem we're on supports creation of FIFOs (named pipes)
633 via mkfifo(1).
634
be53deef
ÆAB
635 - SYMLINKS
636
637 The filesystem we're on supports symbolic links. E.g. a FAT
638 filesystem doesn't support these. See 704a3143 for details.
2fac6a4b 639
c91cfd19
ÆAB
640 - SANITY
641
642 Test is not run by root user, and an attempt to write to an
643 unwritable file is expected to fail correctly.
2fac6a4b 644
8f852ce6
MK
645 - LIBPCRE
646
647 Git was compiled with USE_LIBPCRE=YesPlease. Wrap any tests
648 that use git-grep --perl-regexp or git-grep -P in these.
649
ac39aa61
MG
650 - CASE_INSENSITIVE_FS
651
652 Test is run on a case insensitive file system.
653
5b0b5dd4
MG
654 - UTF8_NFD_TO_NFC
655
656 Test is run on a filesystem which converts decomposed utf-8 (nfd)
657 to precomposed utf-8 (nfc).
658
986aa7f1
JH
659Tips for Writing Tests
660----------------------
661
662As with any programming projects, existing programs are the best
663source of the information. However, do _not_ emulate
664t0000-basic.sh when writing your tests. The test is special in
665that it tries to validate the very core of GIT. For example, it
666knows that there will be 256 subdirectories under .git/objects/,
667and it knows that the object ID of an empty tree is a certain
66840-byte string. This is deliberately done so in t0000-basic.sh
669because the things the very basic core test tries to achieve is
670to serve as a basis for people who are changing the GIT internal
671drastically. For these people, after making certain changes,
672not seeing failures from the basic test _is_ a failure. And
673such drastic changes to the core GIT that even changes these
674otherwise supposedly stable object IDs should be accompanied by
675an update to t0000-basic.sh.
676
677However, other tests that simply rely on basic parts of the core
678GIT working properly should not have that level of intimate
679knowledge of the core GIT internals. If all the test scripts
680hardcoded the object IDs like t0000-basic.sh does, that defeats
681the purpose of t0000-basic.sh, which is to isolate that level of
682validation in one place. Your test also ends up needing
683updating when such a change to the internal happens, so do _not_
684do it and leave the low level of validation to t0000-basic.sh.
d15e9ebc 685
0c357544
ÆAB
686Test coverage
687-------------
688
689You can use the coverage tests to find code paths that are not being
690used or properly exercised yet.
691
692To do that, run the coverage target at the top-level (not in the t/
693directory):
694
695 make coverage
696
697That'll compile Git with GCC's coverage arguments, and generate a test
698report with gcov after the tests finish. Running the coverage tests
699can take a while, since running the tests in parallel is incompatible
700with GCC's coverage mode.
701
702After the tests have run you can generate a list of untested
703functions:
704
705 make coverage-untested-functions
706
707You can also generate a detailed per-file HTML report using the
708Devel::Cover module. To install it do:
709
710 # On Debian or Ubuntu:
711 sudo aptitude install libdevel-cover-perl
712
713 # From the CPAN with cpanminus
714 curl -L http://cpanmin.us | perl - --sudo --self-upgrade
715 cpanm --sudo Devel::Cover
716
717Then, at the top-level:
718
719 make cover_db_html
720
721That'll generate a detailed cover report in the "cover_db_html"
722directory, which you can then copy to a webserver, or inspect locally
723in a browser.