Fixes:
```
$ ./runtests.pl -g 1940
./libtest/libtests lib1940: No such file or directory.
Argument list to give program being debugged when it is started is "http://127.0.0.1:44547/1940".
```
Reported-by: Daniel Stenberg
Fixes#16893Closes#16898
To run curl, tests and servers via `wine`:
```shell
export CURL_TEST_EXE_RUNNER=wine
```
runtests prefixes commands with the specified runner. For systems where
this isn't automatic or supported, e.g. macOS.
Closes#16785
In the cases observed throughout the last year, `handle64` run once per
test run, but with no action (match or task kill). It did not help with
flakiness and seems redundant.
runtests launched it (if present) in Cygwin/MSYS jobs too, where it
probably shouldn't have, because we have seen no flakiness there. In CI
the tool was present and launched in MSYS2 jobs, but not in Cygwin.
After this patch the "clearlocks" warning remain in the log. They are
consistently appearing once in every MSVC CI log, early in the tests:
```
test 3207 SKIPPED: curl lacks OpenSSL support
[...START-OF-TESTS...]
test 0003...[HTTP POST with auth and contents but with content-length set to 0]
--pd---e--- OK (3 out of 1596, remaining: 17:50, took 1.423s, duration: 00:02)
test 0007...[HTTP with cookie parser and header recording]
--pd--oe--- OK (7 out of 1596, remaining: 07:51, took 1.485s, duration: 00:02)
test 0006...[HTTP with simple cookie send]
--pd---e--- OK (6 out of 1596, remaining: 09:11, took 1.488s, duration: 00:02)
test 0005...[HTTP over proxy]
--pd---e--- OK (5 out of 1596, remaining: 11:03, took 1.491s, duration: 00:02)
CUSTOMBUILD : error : 169: cleardir(log/8/lock) failed [D:\a\curl\curl\bld\tests\test-ci.vcxproj]
test 0001...[HTTP GET]
--pd---e--- OK (1 out of 1596, remaining: 55:34, took 1.466s, duration: 00:02)
test 0004...[Replaced internal and added custom HTTP headers]
```
Ref: https://github.com/curl/curl/actions/runs/13546192228/job/37858323380?pr=16484#step:14:167
Ref: e53523fef0#14859
Ref: 311c31ec8e#6179
Follow-up to 3a8920e5ed#16600Closes#16484
If libtests, units and servers binaries are all present, auto-enable
bundle mode.
Drop manual runtests option.
Note: Make sure to "make clean" before changing the test bundle build
setting.
Also fix to append executable extension to all libtest and unit test
executables when launching them. This should make it a tiny bit faster
on Windows.
Follow-up to f4f25505df#15000
Follow-up to 71cf0d1fca#14772Closes#16750
Support multi-target cmake builds via `CURL_DIRSUFFIX` env. For example:
`export CURL_DIRSUFFIX=Debug/`.
Multi-target generators place their output to `src/<subdir>/`,
`lib/<subdir>/`, `tests/server/<subdir>`, `tests/libtest/<subdir>` and
`tests/unit/<subdir>/` by default. Before this patch, `runtests.pl`
couldn't run on such builds because it expected the binaries under the
their `<subdir>`-less directories. This patch allows to set such subdir
and make `runtests.pl` find the binaries. In CI we use multi-target
builds with tests for MSVC. It also helps Xcode-generator builds, though
in CI we don't have such job running tests.
There may be better solutions to configure this, but passing a custom
value to `runtests.pl` including its subprocesses is somewhat tricky.
The reason the configuration value expects the slash at the end is
because MSYS is automagically expanding the env to a (wrong) absolute
path if the slash is in the front.
Also:
- drop the `-DCMAKE_RUNTIME_OUTPUT_DIRECTORY_*=` workaround from CI.
- replace `resolve` references in tests with a new `%RESOLVE` variable.
It didn't use a filename extension before. After this patch it uses
`exe_ext('TOOL')`. I'm not sure if this is the correct choice vs.
`exe_ext('SRV')`.
- fix `-c` option format in manual.
- fix some whitespace.
Note, in CI we still tweak `CMAKE_RUNTIME_OUTPUT_DIRECTORY_*` in jobs
which share steps between `./configure` and cmake. It's easier that way.
Ref: #15000
Cherry-picked from #16394Closes#16452
In certain Windows configurations, Perl resides under `C:/Program Files`
causing tests to fail when executing Perl. Fix by quoting the command.
Seen in `dl-mingw` jobs when switching to the default `bash` shell
offered by the `windows-latest` runner on GHA.
Also:
- apply the same fix for `valgrind` for consistency.
- make more use of `shell_quote()` when passing the `srcdir` directory
over the command-line. This doesn't come up in CI, but seems like
good practice. There are lots more unquoted arguments and possibly
also commands.
```
-------e--- OK (940 out of 1537, remaining: 00:32, took 0.217s, duration: 00:50)
test 1167...[Verify curl prefix of public symbols in header files]
/C/Program Files/Git/usr/bin/perl -I. -ID:/a/curl/curl/tests returned 127, when expecting 0
1167: exit FAILED
[...]
=== Start of file stderr1167
sh: line 1: /C/Program: No such file or directory
```
Ref: https://github.com/curl/curl/actions/runs/13181757313/job/36794072190?pr=16217#step:13:2107Closes#16220
lib : remove all hyper code
configure: stop detecting hyper
docs: no more mention of hyper
tests: mo more special-handling of hyper builds
CI: no jobs using hyper
Closes#15120
Using this option (only available in debug builds) makes curl always
call curl_easy_duphandle() on the handle before using it.
To help us catch curl_easy_duphandle() mistakes better.
Add a CI job using this.
Bonus: the previous runtests option -e is now also supported as
--test-event
Closes#15504
handle/handle64 requires a literal match with the filenames it's
listing.
Also:
- make handle64 log messages more unique to help text searches.
- update a comment with Windows Perl info.
Cherry-picked from #14949Closes#15436
Also introduce 'notexists' for verifying that directory entries do not
exist after a test. Now an explicit supported feature instead of using
"funny" perl in postcheck.
Closes#15046
Collect all ready runners from select() and process in a loop. This
assures fairness in processing among all runners.
Formerly, only the first ready runner in the list of all was processed,
leading to later runners being delayed in processing and reporting
overly long test durations.
Also, reduce the backend idle timeout for the h2/h3 test servers so that
process shutdowns take less time.
Closes#14967
- when server are killed by a test case, do not wait for the server lock
file to go away. These tests are mostly about client timeouts and the
server will hang until killed.
- when killing a server successfully, check for a remaining lock file,
log its existence and remove it.
- lower the delay timings on SLOWDOWN by half
- add SLOWDOWNDATA server command to only slow down the FTP data bytes,
not the control ones.
- lower some timeout values
Closes#14835
- Install stunnel.
- Regenerate certificates (as SecureTransport requires a validity period
less than 398 days).
- Restart server if it is unresponsive.
- Do not hardcode the SHA-256 base64 public pinned key.
- Ignore test 313 as SecureTransport does not support crl file.
- Ignore tests 1631 and 1632 as SecureTransport is not yet able to shut
down FTP over HTTPS gracefully.
- Add a CMake target for generating certificates.
Closes#14486
Some feature names used in tests had minor differences compared to
the well-known ones from `curl -V`. This patch syncs them to make test
results easier to grok.
Closes#14183
To reduce the risk that the user running the tests has a .curlrc present
that messes things up.
Support 'option="no-q"' for the <command> tag to switch it off on demand.
Use this new feature in test 433 and 436.
Ref: #13284Closes#13387
- test450: remove --config from the keywords
- test2080: change return code
- test428: add --config as a keyword
- test428: disable on Windows due to CI problems
Follow-up to 63b5748
Invokes the test case via lldb instead of gdb. Since using gdb is such a
pain on mac, using lldb is sometimes less quirky.
Closes#12547
This timeout is used to wait for the server to finish writing its logs
before checking them against the expected values. An overloaded machine
could take more than the two seconds previously allocated, so increase
the timeout to 5 seconds.
Ref: #11328Closes#11834
At least msys2 perl v5.32.1 doesn't seem to define this signal. Since
this signal is only used for debugging, just ignore if setting it fails.
Reported-by: Marcel Raad
Fixes#11350Closes#11366
After a test failure log a consistent log message to make it easier to
parse the log file. Also, log a consistent message with "ignored" for
failures that cause the test to be not considered at all. These should
perhaps be counted in the skipped category, but this commit does not
change that behaviour.
Errors reading and writing to the pipes are now better detected and
propagated up to the main test loop so it can be cleanly shut down. Such
errors are usually due to a runner dying so it doesn't make much sense
to try to continue the test run.
If the controller dies unexpectedly, have the runner stop its servers
and exit cleanly. Otherwise, the orphaned servers will stay running in
the background.
Give more information about test harness error conditions to help figure
out what might be wrong. Print some internal test state when SIGUSR1 is
sent to runtests.pl.
Ref: #11328
Logs are written by several servers and all of them must be finished
writing before the test results can be determined. This means each
server must have its own lock file rather than sharing a single one,
which is how it was done up to now. Previously, the first server to
complete a test would clear the lock before the other server was done,
which caused flaky tests.
Lock files are now all found in their own directory, so counting locks
equals counting the files in that directory. The result is that the
proxy logs are now reliably written which actually changes the expected
output for two tests.
Fixes#11231Closes#11259
Test files in subdirectories were not created after parallel test log
directories were moved down a level due to a now-bad comparison.
Follow-up to 92d7dd39
Ref #11264Closes#11267
Each runner needs a unique random seed to reduce the chance of port
number collisions. The new scheme uses a consistent per-runner source of
randomness which results in deterministic behaviour, as it did before.
Ref: #10818
These can be interrupted by signals, especially SIGINT to shut down, and
must be restarted so the IPC call arrives correctly. If the read just
returns an error instead, the IPC calling state will go out of sync and
a proper shutdown won't happen.
Ref: #10818
The curl binary argument wasn't being quoted properly. This seems to
have broken at some point after quoting was added in commit 606b29fe.
Reported-by: Daniel Stenberg
Ref: #11073Fixes#11074Closes#11076
When the -j option is given, a new process is spawned in which the test
programs are run and from which test servers are started. Only one
process can be started at once, but this is sufficient to test that the
infrastructure can isolate those functions in a new task. There should
be no visible difference between the two modes at the moment.
Ref: #10818Closes#11064
