diff options
Diffstat (limited to 'src/test/test_util.c')
| -rw-r--r-- | src/test/test_util.c | 73 |
1 files changed, 66 insertions, 7 deletions
diff --git a/src/test/test_util.c b/src/test/test_util.c index 0921bae109..1a71da2794 100644 --- a/src/test/test_util.c +++ b/src/test/test_util.c @@ -38,6 +38,7 @@ #include "lib/meminfo/meminfo.h" #include "lib/time/tvdiff.h" #include "lib/encoding/confline.h" +#include "lib/net/socketpair.h" #ifdef HAVE_PWD_H #include <pwd.h> @@ -4012,6 +4013,53 @@ test_util_string_is_C_identifier(void *ptr) } static void +test_util_string_is_utf8(void *ptr) +{ + (void)ptr; + + tt_int_op(1, OP_EQ, string_is_utf8(NULL, 0)); + tt_int_op(1, OP_EQ, string_is_utf8("", 1)); + tt_int_op(1, OP_EQ, string_is_utf8("\uFEFF", 3)); + tt_int_op(1, OP_EQ, string_is_utf8("\uFFFE", 3)); + tt_int_op(1, OP_EQ, string_is_utf8("ascii\x7f\n", 7)); + tt_int_op(1, OP_EQ, string_is_utf8("Risqu\u00e9=1", 9)); + + // Validate exactly 'len' bytes. + tt_int_op(0, OP_EQ, string_is_utf8("\0\x80", 2)); + tt_int_op(0, OP_EQ, string_is_utf8("Risqu\u00e9=1", 6)); + + // Reject sequences with missing bytes. + tt_int_op(0, OP_EQ, string_is_utf8("\x80", 1)); + tt_int_op(0, OP_EQ, string_is_utf8("\xc2", 1)); + tt_int_op(0, OP_EQ, string_is_utf8("\xc2 ", 2)); + tt_int_op(0, OP_EQ, string_is_utf8("\xe1\x80", 2)); + tt_int_op(0, OP_EQ, string_is_utf8("\xe1\x80 ", 3)); + tt_int_op(0, OP_EQ, string_is_utf8("\xf1\x80\x80", 3)); + tt_int_op(0, OP_EQ, string_is_utf8("\xf1\x80\x80 ", 4)); + + // Reject encodings that are overly long. + tt_int_op(0, OP_EQ, string_is_utf8("\xc1\xbf", 2)); + tt_int_op(1, OP_EQ, string_is_utf8("\xc2\x80", 2)); + tt_int_op(0, OP_EQ, string_is_utf8("\xe0\x9f\xbf", 3)); + tt_int_op(1, OP_EQ, string_is_utf8("\xe0\xa0\x80", 3)); + tt_int_op(0, OP_EQ, string_is_utf8("\xf0\x8f\xbf\xbf", 4)); + tt_int_op(1, OP_EQ, string_is_utf8("\xf0\x90\x80\x80", 4)); + + // Reject UTF-16 surrogate halves. + tt_int_op(1, OP_EQ, string_is_utf8("\xed\x9f\xbf", 3)); + tt_int_op(0, OP_EQ, string_is_utf8("\xed\xa0\x80", 3)); + tt_int_op(0, OP_EQ, string_is_utf8("\xed\xbf\xbf", 3)); + tt_int_op(1, OP_EQ, string_is_utf8("\xee\x80\x80", 3)); + + // The maximum legal codepoint, 10FFFF. + tt_int_op(1, OP_EQ, string_is_utf8("\xf4\x8f\xbf\xbf", 4)); + tt_int_op(0, OP_EQ, string_is_utf8("\xf4\x90\x80\x80", 4)); + + done: + ; +} + +static void test_util_asprintf(void *ptr) { #define LOREMIPSUM \ @@ -5612,10 +5660,13 @@ test_util_socketpair(void *arg) tt_assert(SOCKET_OK(fds[0])); tt_assert(SOCKET_OK(fds[1])); - tt_int_op(get_n_open_sockets(), OP_EQ, n + 2); + if (ersatz) + tt_int_op(get_n_open_sockets(), OP_EQ, n); + else + tt_int_op(get_n_open_sockets(), OP_EQ, n + 2); #ifdef CAN_CHECK_CLOEXEC - tt_int_op(fd_is_cloexec(fds[0]), OP_EQ, 1); - tt_int_op(fd_is_cloexec(fds[1]), OP_EQ, 1); + tt_int_op(fd_is_cloexec(fds[0]), OP_EQ, !ersatz); + tt_int_op(fd_is_cloexec(fds[1]), OP_EQ, !ersatz); #endif #ifdef CAN_CHECK_NONBLOCK tt_int_op(fd_is_nonblocking(fds[0]), OP_EQ, 0); @@ -5623,10 +5674,17 @@ test_util_socketpair(void *arg) #endif done: - if (SOCKET_OK(fds[0])) - tor_close_socket(fds[0]); - if (SOCKET_OK(fds[1])) - tor_close_socket(fds[1]); + if (ersatz) { + if (SOCKET_OK(fds[0])) + tor_close_socket_simple(fds[0]); + if (SOCKET_OK(fds[1])) + tor_close_socket_simple(fds[1]); + } else { + if (SOCKET_OK(fds[0])) + tor_close_socket(fds[0]); + if (SOCKET_OK(fds[1])) + tor_close_socket(fds[1]); + } } #undef SOCKET_EPROTO @@ -6410,6 +6468,7 @@ struct testcase_t util_tests[] = { UTIL_TEST(clamp_double_to_int64, 0), UTIL_TEST(find_str_at_start_of_line, 0), UTIL_TEST(string_is_C_identifier, 0), + UTIL_TEST(string_is_utf8, 0), UTIL_TEST(asprintf, 0), UTIL_TEST(listdir, 0), UTIL_TEST(parent_dir, 0), |