nihilist/tor
1
0
Fork 0
mirror of https://gitlab.torproject.org/tpo/core/tor.git synced 2024-11-11 13:43:47 +01:00
Code Issues Packages Projects Releases Wiki Activity
b2e7c356db
tor/src/test/tinytest.c

381 lines
9.6 KiB
C
Raw Normal View History

Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
/* tinytest.c -- Copyright 2009-2010 Nick Mathewson
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#ifdef WIN32
#include <windows.h>
#else
#include <sys/types.h>
#include <sys/wait.h>
#include <unistd.h>
#endif
#ifndef __GNUC__
#define __attribute__(x)
#endif
Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
#ifdef TINYTEST_LOCAL
#include "tinytest_local.h"
#endif
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
#include "tinytest.h"
#include "tinytest_macros.h"
#define LONGEST_TEST_NAME 16384
static int in_tinytest_main = 0; /**< true if we're in tinytest_main().*/
static int n_ok = 0; /**< Number of tests that have passed */
static int n_bad = 0; /**< Number of tests that have failed. */
static int n_skipped = 0; /**< Number of tests that have been skipped. */
static int opt_forked = 0; /**< True iff we're called from inside a win32 fork*/
static int opt_nofork = 0; /**< Suppress calls to fork() for debugging. */
static int opt_verbosity = 1; /**< -==quiet,0==terse,1==normal,2==verbose */
const char *verbosity_flag = "";
enum outcome { SKIP=2, OK=1, FAIL=0 };
static enum outcome cur_test_outcome = 0;
const char *cur_test_prefix = NULL; /**< prefix of the current test group */
Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
/** Name of the current test, if we haven't logged is yet. Used for --quiet */
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
const char *cur_test_name = NULL;
#ifdef WIN32
/** Pointer to argv[0] for win32. */
static const char *commandname = NULL;
#endif
static void usage(struct testgroup_t *groups, int list_groups)
__attribute__((noreturn));
static enum outcome
_testcase_run_bare(const struct testcase_t *testcase)
{
void *env = NULL;
int outcome;
if (testcase->setup) {
env = testcase->setup->setup_fn(testcase);
Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
if (!env)
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
return FAIL;
else if (env == (void*)TT_SKIP)
return SKIP;
}
cur_test_outcome = OK;
testcase->fn(env);
outcome = cur_test_outcome;
if (testcase->setup) {
if (testcase->setup->cleanup_fn(testcase, env) == 0)
outcome = FAIL;
}
return outcome;
}
#define MAGIC_EXITCODE 42
static enum outcome
_testcase_run_forked(const struct testgroup_t *group,
const struct testcase_t *testcase)
{
#ifdef WIN32
/* Fork? On Win32? How primitive! We'll do what the smart kids do:
we'll invoke our own exe (whose name we recall from the command
line) with a command line that tells it to run just the test we
want, and this time without forking.
(No, threads aren't an option. The whole point of forking is to
share no state between tests.)
*/
int ok;
char buffer[LONGEST_TEST_NAME+256];
Make the windows build succeed with or without -DUNICODE enabled. This should keep WinCE working (unicode always-on) and get Win98 working again (unicode never-on). There are two places where we explicitly use ASCII-only APIs, still: in ntmain.c and in the unit tests. This patch also fixes a bug in windoes tor_listdir that would cause the first file to be listed an arbitrary number of times that was also introduced with WinCE support. Should fix bug 1797.
2010-08-20 18:30:25 +02:00
STARTUPINFOA si;
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
PROCESS_INFORMATION info;
DWORD exitcode;
if (!in_tinytest_main) {
printf("\nERROR. On Windows, _testcase_run_forked must be"
" called from within tinytest_main.\n");
abort();
}
if (opt_verbosity>0)
printf("[forking] ");
snprintf(buffer, sizeof(buffer), "%s --RUNNING-FORKED %s %s%s",
commandname, verbosity_flag, group->prefix, testcase->name);
memset(&si, 0, sizeof(si));
memset(&info, 0, sizeof(info));
si.cb = sizeof(si);
Make the windows build succeed with or without -DUNICODE enabled. This should keep WinCE working (unicode always-on) and get Win98 working again (unicode never-on). There are two places where we explicitly use ASCII-only APIs, still: in ntmain.c and in the unit tests. This patch also fixes a bug in windoes tor_listdir that would cause the first file to be listed an arbitrary number of times that was also introduced with WinCE support. Should fix bug 1797.
2010-08-20 18:30:25 +02:00
ok = CreateProcessA(commandname, buffer, NULL, NULL, 0,
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
0, NULL, NULL, &si, &info);
if (!ok) {
printf("CreateProcess failed!\n");
return 0;
}
WaitForSingleObject(info.hProcess, INFINITE);
GetExitCodeProcess(info.hProcess, &exitcode);
CloseHandle(info.hProcess);
CloseHandle(info.hThread);
if (exitcode == 0)
return OK;
else if (exitcode == MAGIC_EXITCODE)
return SKIP;
else
return FAIL;
#else
int outcome_pipe[2];
pid_t pid;
Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
(void)group;
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
if (pipe(outcome_pipe))
perror("opening pipe");
if (opt_verbosity>0)
printf("[forking] ");
pid = fork();
if (!pid) {
/* child. */
int test_r, write_r;
char b[1];
close(outcome_pipe[0]);
test_r = _testcase_run_bare(testcase);
assert(0<=(int)test_r && (int)test_r<=2);
b[0] = "NYS"[test_r];
Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
write_r = (int)write(outcome_pipe[1], b, 1);
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
if (write_r != 1) {
perror("write outcome to pipe");
exit(1);
}
exit(0);
} else {
/* parent */
int status, r;
char b[1];
/* Close this now, so that if the other side closes it,
* our read fails. */
close(outcome_pipe[1]);
r = (int)read(outcome_pipe[0], b, 1);
if (r == 0) {
printf("[Lost connection!] ");
return 0;
} else if (r != 1) {
perror("read outcome from pipe");
}
waitpid(pid, &status, 0);
close(outcome_pipe[0]);
return b[0]=='Y' ? OK : (b[0]=='S' ? SKIP : FAIL);
}
#endif
}
int
testcase_run_one(const struct testgroup_t *group,
const struct testcase_t *testcase)
{
enum outcome outcome;
if (testcase->flags & TT_SKIP) {
if (opt_verbosity>0)
printf("%s%s: SKIPPED\n",
group->prefix, testcase->name);
++n_skipped;
return SKIP;
}
if (opt_verbosity>0 && !opt_forked) {
printf("%s%s: ", group->prefix, testcase->name);
} else {
if (opt_verbosity==0) printf(".");
cur_test_prefix = group->prefix;
cur_test_name = testcase->name;
}
if ((testcase->flags & TT_FORK) && !(opt_forked||opt_nofork)) {
outcome = _testcase_run_forked(group, testcase);
} else {
Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
outcome = _testcase_run_bare(testcase);
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
}
if (outcome == OK) {
++n_ok;
if (opt_verbosity>0 && !opt_forked)
puts(opt_verbosity==1?"OK":"");
} else if (outcome == SKIP) {
++n_skipped;
if (opt_verbosity>0 && !opt_forked)
puts("SKIPPED");
} else {
++n_bad;
if (!opt_forked)
printf("\n [%s FAILED]\n", testcase->name);
}
if (opt_forked) {
exit(outcome==OK ? 0 : (outcome==SKIP?MAGIC_EXITCODE : 1));
} else {
return (int)outcome;
}
}
int
_tinytest_set_flag(struct testgroup_t *groups, const char *arg, unsigned long flag)
{
int i, j;
size_t length = LONGEST_TEST_NAME;
char fullname[LONGEST_TEST_NAME];
int found=0;
if (strstr(arg, ".."))
length = strstr(arg,"..")-arg;
for (i=0; groups[i].prefix; ++i) {
for (j=0; groups[i].cases[j].name; ++j) {
snprintf(fullname, sizeof(fullname), "%s%s",
groups[i].prefix, groups[i].cases[j].name);
if (!flag) /* Hack! */
printf(" %s\n", fullname);
if (!strncmp(fullname, arg, length)) {
groups[i].cases[j].flags |= flag;
++found;
}
}
}
return found;
}
static void
usage(struct testgroup_t *groups, int list_groups)
{
puts("Options are: [--verbose|--quiet|--terse] [--no-fork]");
puts(" Specify tests by name, or using a prefix ending with '..'");
Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
puts(" To skip a test, list give its name prefixed with a colon.");
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
puts(" Use --list-tests for a list of tests.");
if (list_groups) {
puts("Known tests are:");
_tinytest_set_flag(groups, "..", 0);
}
exit(0);
}
int
tinytest_main(int c, const char **v, struct testgroup_t *groups)
{
int i, j, n=0;
#ifdef WIN32
commandname = v[0];
#endif
for (i=1; i<c; ++i) {
if (v[i][0] == '-') {
if (!strcmp(v[i], "--RUNNING-FORKED")) {
opt_forked = 1;
} else if (!strcmp(v[i], "--no-fork")) {
opt_nofork = 1;
} else if (!strcmp(v[i], "--quiet")) {
opt_verbosity = -1;
verbosity_flag = "--quiet";
} else if (!strcmp(v[i], "--verbose")) {
opt_verbosity = 2;
verbosity_flag = "--verbose";
} else if (!strcmp(v[i], "--terse")) {
opt_verbosity = 0;
verbosity_flag = "--terse";
} else if (!strcmp(v[i], "--help")) {
usage(groups, 0);
} else if (!strcmp(v[i], "--list-tests")) {
usage(groups, 1);
} else {
printf("Unknown option %s. Try --help\n",v[i]);
return -1;
}
} else {
Update to the latest tinytest version This cleans up some whitespace consistency issues and, more importantly, gives you the ability to skip tests from the command line.
2010-09-09 20:34:10 +02:00
const char *test = v[i];
int flag = _TT_ENABLED;
if (test[0] == ':') {
++test;
flag = TT_SKIP;
} else {
++n;
}
if (!_tinytest_set_flag(groups, test, flag)) {
Refactor unit tests to use the tinytest framework. "Tinytest" is a minimalist C unit testing framework I wrote for Libevent. It supports some generally useful features, like being able to run separate unit tests in their own processes. I tried to do the refactoring to change test.c as little as possible. Thus, we mostly don't call the tinytest macros directly. Instead, the test.h header is now a wrapper on tinytest.h to make our existing test_foo() macros work. The next step(s) here will be: - To break test.c into separate files, each with its own test group. - To look into which things we can test - To refactor the more fiddly tests to use the tinytest macros directly and/or run forked. - To see about writing unit tests for things we couldn't previously test without forking.
2009-09-21 20:23:13 +02:00
printf("No such test as %s!\n", v[i]);
return -1;
}
}
}
if (!n)
_tinytest_set_flag(groups, "..", _TT_ENABLED);
setvbuf(stdout, NULL, _IONBF, 0);
++in_tinytest_main;
for (i=0; groups[i].prefix; ++i)
for (j=0; groups[i].cases[j].name; ++j)
if (groups[i].cases[j].flags & _TT_ENABLED)
testcase_run_one(&groups[i],
&groups[i].cases[j]);
--in_tinytest_main;
if (opt_verbosity==0)
puts("");
if (n_bad)
printf("%d/%d TESTS FAILED. (%d skipped)\n", n_bad,
n_bad+n_ok,n_skipped);
else if (opt_verbosity >= 1)
printf("%d tests ok. (%d skipped)\n", n_ok, n_skipped);
return (n_bad == 0) ? 0 : 1;
}
int
_tinytest_get_verbosity(void)
{
return opt_verbosity;
}
void
_tinytest_set_test_failed(void)
{
if (opt_verbosity <= 0 && cur_test_name) {
if (opt_verbosity==0) puts("");
printf("%s%s: ", cur_test_prefix, cur_test_name);
cur_test_name = NULL;
}
cur_test_outcome = 0;
}
void
_tinytest_set_test_skipped(void)
{
if (cur_test_outcome==OK)
cur_test_outcome = SKIP;
}
Reference in New Issue Copy Permalink