1 2002-09-10 Paul D. Smith <psmith@gnu.org>
3 * scripts/features/export: Created. Add tests for export/unexport
4 capabilities, including exporting/unexporting expanded variables.
6 * scripts/features/conditionals: Add a test for expanded variables
9 2002-09-04 Paul D. Smith <psmith@gnu.org>
11 * scripts/features/reinvoke: Change touch/sleep combos to utouch
13 * scripts/features/vpathgpath: Ditto.
14 * scripts/features/vpathplus: Ditto.
15 * scripts/options/dash-n: Ditto.
16 * scripts/targets/INTERMEDIATE: Ditto.
17 * scripts/targets/SECONDARY: Ditto.
19 * scripts/options/dash-t: Added a test for the -t bug fixed by
20 Henning Makholm. This test was also contributed by Henning.
22 * scripts/misc/general4: Add a test suite for obscure algorithmic
23 features of make. First test: make sure creation subdirectories
24 as prerequisites of targets works properly.
26 * scripts/misc/version: Remove this bogus test.
28 2002-08-07 Paul D. Smith <psmith@gnu.org>
30 * scripts/misc/general3: Add a test for makefiles that don't end
33 * scripts/variables/special: Create tests for the special
34 variables (.VARIABLES and .TARGETS). Comment out .TARGETS test
35 for now as it's not yet supported.
37 2002-08-01 Paul D. Smith <psmith@gnu.org>
39 * scripts/options/dash-B: Add a test for the new -B option.
41 2002-07-11 Paul D. Smith <psmith@gnu.org>
43 * run_make_tests.pl (valid_option): Add support for Valgrind
44 <http://developer.kde.org/~sewardj/>. Use -valgrind option to the
46 (set_more_defaults): Set up the file descriptor to capture
47 Valgrind output. We have to unset its close-on-exec flag; we
48 hardcode the value for F_SETFD (2) rather than load it; hopefully
49 this will help us avoid breaking the Windows/DOS test suite.
51 2002-07-10 Paul D. Smith <psmith@gnu.org>
53 * scripts/variables/automatic: Add some tests for $$@, $$(@D), and
56 * test_driver.pl (utouch): Create a new function that creates a
57 file with a specific timestamp offset. Use of this function will
58 let us avoid lots of annoying sleep() invocations in the tests
59 just to get proper timestamping, which will make the tests run a
60 lot faster. So far it's only used in the automatic test suite.
62 2002-07-09 Paul D. Smith <psmith@gnu.org>
64 * scripts/variables/automatic: Create a test for automatic variables.
66 2002-07-08 Paul D. Smith <psmith@gnu.org>
68 * scripts/features/order_only: Test new order-only prerequisites.
70 2002-07-07 Paul D. Smith <psmith@gnu.org>
72 * scripts/functions/eval: Test new function.
73 * scripts/functions/value: Test new function.
74 * scripts/variables/MAKEFILE_LIST: Test new variable.
76 2002-04-28 Paul D. Smith <psmith@gnu.org>
78 * scripts/functions/call: New test: transitive closure
79 implementation using $(call ...) to test variable recursion.
81 2002-04-21 Paul D. Smith <psmith@gnu.org>
83 * test_driver.pl (compare_dir_tree): Ignore CVS and RCS
84 directories in the script directories.
86 2001-05-02 Paul D. Smith <psmith@gnu.org>
88 * scripts/variables/flavors: Test define/endef scripts where only
89 one of the command lines is quiet.
91 2000-06-22 Paul D. Smith <psmith@gnu.org>
93 * scripts/options/dash-q: New file; test the -q option. Includes
96 2000-06-21 Paul D. Smith <psmith@gnu.org>
98 * scripts/features/targetvars: Added a test for PR/1709: allowing
99 semicolons in target-specific variable values.
101 2000-06-19 Paul D. Smith <psmith@gnu.org>
103 * scripts/functions/addsuffix: Test for an empty final argument.
104 Actually this bug might have happened for any function, but this
107 2000-06-17 Eli Zaretskii <eliz@is.elta.co.il>
109 * scripts/options/general: If parallel jobs are not supported,
110 expect a warning message from Make.
112 2000-06-15 Eli Zaretskii <eliz@is.elta.co.il>
114 * scripts/options/general: Don't try -jN with N != 1 if parallel
115 jobs are not supported.
117 2000-05-24 Paul D. Smith <psmith@gnu.org>
119 * scripts/options/general: Test general option processing (PR/1716).
121 2000-04-11 Paul D. Smith <psmith@gnu.org>
123 * scripts/functions/strip: Test empty value to strip (PR/1689).
125 2000-04-08 Eli Zaretskii <eliz@is.elta.co.il>
127 * scripts/features/reinvoke: Sleep before updating the target
128 files in the first test, to ensure its time stamp really gets
129 newer; otherwise Make might re-exec more than once.
131 2000-04-07 Eli Zaretskii <eliz@is.elta.co.il>
133 * scripts/features/double_colon: Don't run the parallel tests if
134 parallel jobs aren't supported.
136 2000-04-04 Paul D. Smith <psmith@gnu.org>
138 * scripts/functions/word: wordlist doesn't swap arguments anymore.
140 2000-03-27 Paul D. Smith <psmith@gnu.org>
142 * scripts/features/statipattrules: Test that static pattern rules
143 whose prerequisite patterns resolve to empty strings throw an
144 error (instead of dumping core). Fixes PR/1670.
146 * scripts/features/reinvoke: Make more robust by touching "b"
147 first, to ensure it's not newer than "a".
148 Reported by Marco Franzen <Marco.Franzen@Thyron.com>.
149 * scripts/options/dash-n: Ditto.
151 * scripts/functions/call: Whoops. The fix to PR/1527 caused
152 recursive invocations of $(call ...) to break. I can't come up
153 with any way to get both working at the same time, so I backed out
154 the fix to 1527 and added a test case for recursive calls. This
155 also tests the fix for PR/1610.
157 * scripts/features/double_colon: Test that circular dependencies
158 in double-colon rule sets are detected correctly (PR/1671).
160 2000-03-26 Paul D. Smith <psmith@gnu.org>
162 * scripts/targets/INTERMEDIATE: Test that make doesn't remove
163 .INTERMEDIATE files when given on the command line (PR/1669).
165 2000-03-08 Paul D. Smith <psmith@gnu.org>
167 * scripts/options/dash-k: Add a test for error detection by
168 multiple targets depending on the same prerequisite with -k.
171 2000-02-07 Paul D. Smith <psmith@gnu.org>
173 * scripts/features/escape: Add a test for backslash-escaped spaces
174 in a target name (PR/1586).
176 2000-02-04 Paul D. Smith <psmith@gnu.org>
178 * scripts/features/patspecific_vars: Add a test for pattern-specific
179 target variables inherited from the parent target (PR/1407).
181 2000-02-02 Paul D. Smith <psmith@gnu.org>
183 * run_make_tests.pl (set_more_defaults): Hard-code the LANG to C
184 to make sure sorting order, etc. is predictable.
185 Reported by Andreas Jaeger <aj@suse.de>.
187 * run_make_tests.pl (set_more_defaults): Set the $wtime variable
188 depending on the OS. Eli Zaretskii <eliz@is.elta.co.il> reports
189 this seems to need to be *4* on DOS/Windows, not just 2. Keep it
191 * scripts/features/vpathplus (touchfiles): Use the $wtime value
192 instead of hardcoding 2.
193 * scripts/targets/SECONDARY: Ditto.
194 * scripts/targets/INTERMEDIATE: Ditto.
196 2000-01-27 Paul D. Smith <psmith@gnu.org>
198 * test_driver.pl (toplevel): Don't try to run test scripts which
199 are really directories.
201 2000-01-23 Paul D. Smith <psmith@gnu.org>
203 * scripts/features/include: Remove a check; the fix caused more
204 problems than the error, so I removed it and removed the test for
207 2000-01-11 Paul D. Smith <psmith@gnu.org>
209 * scripts/functions/call: Add a test for PR/1517 and PR/1527: make
210 sure $(call ...) doesn't eval its arguments and that you can
211 invoke foreach from it without looping forever.
213 1999-12-15 Paul D. Smith <psmith@gnu.org>
215 * scripts/targets/INTERMEDIATE: Add a test for PR/1423: make sure
216 .INTERMEDIATE settings on files don't disable them as implicit
217 intermediate possibilities.
219 1999-12-01 Paul D. Smith <psmith@gnu.org>
221 * scripts/features/double_colon: Add a test for PR/1476: Try
222 double-colon rules as non-goal targets and during parallel builds
223 to make sure they're handled serially.
225 1999-11-17 Paul D. Smith <psmith@gnu.org>
227 * scripts/functions/if: Add a test for PR/1429: put some text
228 after an if-statement to make sure it works.
230 * scripts/features/targetvars: Add a test for PR/1380: handling +=
231 in target-specific variable definitions correctly.
233 1999-10-15 Paul D. Smith <psmith@gnu.org>
235 * scripts/variables/MAKEFILES: This was really broken: it didn't
236 test anything at all, really. Rewrote it, plus added a test for
239 1999-10-13 Paul D. Smith <psmith@gnu.org>
241 * scripts/options/dash-n: Add a test for PR/1379: "-n doesn't
242 behave properly when used with recursive targets".
244 1999-10-08 Paul D. Smith <psmith@gnu.org>
246 * scripts/features/targetvars: Add a check for PR/1378:
247 "Target-specific vars don't inherit correctly"
249 1999-09-29 Paul D. Smith <psmith@gnu.org>
251 * test_driver.pl (get_osname): Change $fancy_file_names to
252 $short_filenames and reverse the logic.
253 (run_each_test): Change test of non-existent $port_host to use
254 $short_filenames--problem reported by Eli Zaretskii.
256 1999-09-23 Paul D. Smith <psmith@gnu.org>
258 * scripts/features/parallelism: Add a check to ensure that the
259 jobserver works when we re-invoke. Also cleaned up the tests a
260 little, reducing the number of rules we use so the test won't need
261 as many "sleep" commands.
263 1999-09-16 Paul D. Smith <psmith@gnu.org>
265 * scripts/features/reinvoke: Remove invocations of "touch" in
266 makefiles. See the comments on the touch function rewrite below.
267 Note that UNIX touch behaves the same way if the file already
268 exists: it sets the time to the _local_ time. We don't want
269 this. This is probably a good tip for makefile writers in
270 general, actually... where practical.
271 * scripts/options/dash-l: Ditto.
272 * scripts/options/dash-n: Ditto.
274 * test_driver.pl (run_each_test): In retrospect, I don't like the
275 .lN/.bN/.dN postfix required by DOS. So, for non-DOS systems I
276 changed it back to use .log, .base, and .diff.
278 * run_make_tests.pl (set_more_defaults): Move the check for the
279 make pathname to here from set_defaults (that's too early since it
280 happens before the command line processing).
281 Create a new variable $port_type, calculated from $osname, to
282 specify what kind of system we're running on. We should integrate
283 the VOS stuff here, too.
284 (valid_option): Comment out the workdir/-work stuff so people
285 won't be fooled into thinking it works... someone needs to fix
288 * scripts/functions/origin: Use $port_type instead of $osname.
289 * scripts/functions/foreach: Ditto.
290 * scripts/features/default_names: Ditto.
292 1999-09-15 Paul D. Smith <psmith@gnu.org>
294 * test_driver.pl (touch): Rewrite this function. Previously it
295 used to use utime() to hard-set the time based on the current
296 local clock, or, if the file didn't exist, it merely created it.
297 This mirrors exactly what real UNIX touch does, but it fails badly
298 on networked filesystems where the FS server clock is skewed from
299 the local clock: normally modifying a file causes it to get a mod
300 time based on the _server's_ clock. Hard-setting it based on the
301 _local_ clock causes gratuitous errors and makes the tests
302 unreliable except on local filesystems. The new function will
303 simply modify the file, allowing the filesystem to set the mod
306 * scripts/features/parallelism: The second test output could
307 change depending on how fast some scripts completed; use "sleep"
308 to force the order we want.
310 * test_driver.pl (toplevel): A bug in Perl 5.000 to Perl 5.004
311 means that "%ENV = ();" doesn't do the right thing. This worked
312 in Perl 4 and was fixed in Perl 5.004_01, but use a loop to delete
313 the environment rather than require specific versions.
315 * run_make_tests.pl (set_more_defaults): Don't use Perl 5 s///
316 modifier "s", so the tests will run with Perl 4.
317 (set_more_defaults): Set $pure_log to empty if there's no -logfile
318 option in PURIFYOPTIONS.
319 (setup_for_test): Don't remove any logs unless $pure_log is set.
321 1999-09-15 Eli Zaretskii <eliz@is.elta.co.il>
323 * scripts/features/reinvoke: Put the SHELL definition in the right
326 1999-09-15 Paul D. Smith <psmith@gnu.org>
328 ChangeLog file for the test suite created.