diff options
author | Daan De Meyer <daan.j.demeyer@gmail.com> | 2022-07-22 14:49:42 +0200 |
---|---|---|
committer | Daan De Meyer <daan.j.demeyer@gmail.com> | 2022-07-25 14:16:17 +0200 |
commit | 75db32dcd810cb6f4bc0148c3e280c5c04203491 (patch) | |
tree | d200b732eeae976392e8c64ac87782ecaa0f022e /src/shared/pcre2-util.c | |
parent | 05abe850330fdef145d8d45d5a2e33d76a0a8d21 (diff) | |
download | systemd-75db32dcd810cb6f4bc0148c3e280c5c04203491.tar.gz |
journal: Move more pattern matching logic into pcre2-util
To avoid having "#if HAVE_PCRE2" all throughout the code, let's
confine the pcre2 header specific stuff to pcre2-util.c. Instead of
exposing all the individual symbols from pcre2, let's only expose
three high level functions that do all we need:
- pcre2_pattern_compile(): Compile the regex
- pcre2_pattern_matches(): Check if the compiled regex matches a message
- pcre2_pattern_free(): Free the compiled regex
We expose the compiled pcre2 pattern (which is of type pcre2_code *) as
a void pointer to avoid having to include pcre2.h in all code where we
work with compiled pcre2 patterns. For readability, we typedef void
to pcre2_pattern and use that as the type specifier for compiled pcre2
patterns.
Diffstat (limited to 'src/shared/pcre2-util.c')
-rw-r--r-- | src/shared/pcre2-util.c | 119 |
1 files changed, 116 insertions, 3 deletions
diff --git a/src/shared/pcre2-util.c b/src/shared/pcre2-util.c index 80f6cec342..998dab0491 100644 --- a/src/shared/pcre2-util.c +++ b/src/shared/pcre2-util.c @@ -14,8 +14,10 @@ pcre2_code* (*sym_pcre2_compile)(PCRE2_SPTR, PCRE2_SIZE, uint32_t, int *, PCRE2_ int (*sym_pcre2_get_error_message)(int, PCRE2_UCHAR *, PCRE2_SIZE); int (*sym_pcre2_match)(const pcre2_code *, PCRE2_SPTR, PCRE2_SIZE, PCRE2_SIZE, uint32_t, pcre2_match_data *, pcre2_match_context *); PCRE2_SIZE* (*sym_pcre2_get_ovector_pointer)(pcre2_match_data *); +#endif int dlopen_pcre2(void) { +#if HAVE_PCRE2 /* So here's something weird: PCRE2 actually renames the symbols exported by the library via C * macros, so that the exported symbols carry a suffix "_8" but when used from C the suffix is * gone. In the argument list below we ignore this mangling. Surprisingly (at least to me), we @@ -33,12 +35,123 @@ int dlopen_pcre2(void) { DLSYM_ARG(pcre2_get_error_message), DLSYM_ARG(pcre2_match), DLSYM_ARG(pcre2_get_ovector_pointer)); +#else + return log_error_errno(SYNTHETIC_ERRNO(EOPNOTSUPP), "PCRE2 support is not compiled in."); +#endif } +int pattern_compile_and_log(const char *pattern, PatternCompileCase case_, pcre2_code **ret) { +#if HAVE_PCRE2 + PCRE2_SIZE erroroffset; + pcre2_code *p; + unsigned flags = 0; + int errorcode, r; + + assert(pattern); + + r = dlopen_pcre2(); + if (r < 0) + return r; + + if (case_ == PATTERN_COMPILE_CASE_INSENSITIVE) + flags = PCRE2_CASELESS; + else if (case_ == PATTERN_COMPILE_CASE_AUTO) { + _cleanup_(sym_pcre2_match_data_freep) pcre2_match_data *md = NULL; + bool has_case; + _cleanup_(sym_pcre2_code_freep) pcre2_code *cs = NULL; + + md = sym_pcre2_match_data_create(1, NULL); + if (!md) + return log_oom(); + + r = pattern_compile_and_log("[[:upper:]]", PATTERN_COMPILE_CASE_SENSITIVE, &cs); + if (r < 0) + return r; + + r = sym_pcre2_match(cs, (PCRE2_SPTR8) pattern, PCRE2_ZERO_TERMINATED, 0, 0, md, NULL); + has_case = r >= 0; + + flags = !has_case * PCRE2_CASELESS; + } + + log_debug("Doing case %s matching based on %s", + flags & PCRE2_CASELESS ? "insensitive" : "sensitive", + case_ != PATTERN_COMPILE_CASE_AUTO ? "request" : "pattern casing"); + + p = sym_pcre2_compile((PCRE2_SPTR8) pattern, + PCRE2_ZERO_TERMINATED, flags, &errorcode, &erroroffset, NULL); + if (!p) { + unsigned char buf[LINE_MAX]; + + r = sym_pcre2_get_error_message(errorcode, buf, sizeof buf); + + return log_error_errno(SYNTHETIC_ERRNO(EINVAL), + "Bad pattern \"%s\": %s", pattern, + r < 0 ? "unknown error" : (char *)buf); + } + + if (ret) + *ret = p; + + return 0; #else + return log_error_errno(SYNTHETIC_ERRNO(EOPNOTSUPP), "PCRE2 support is not compiled in."); +#endif +} -int dlopen_pcre2(void) { - return log_error_errno(SYNTHETIC_ERRNO(EOPNOTSUPP), - "PCRE2 support is not compiled in."); +int pattern_matches_and_log(pcre2_code *compiled_pattern, const char *message, size_t size, size_t *ret_ovec) { +#if HAVE_PCRE2 + _cleanup_(sym_pcre2_match_data_freep) pcre2_match_data *md = NULL; + int r; + + assert(compiled_pattern); + assert(message); + /* pattern_compile_and_log() must be called before this function is called and that function already + * dlopens pcre2 so we can assert on it being available here. */ + assert(pcre2_dl); + + md = sym_pcre2_match_data_create(1, NULL); + if (!md) + return log_oom(); + + r = sym_pcre2_match(compiled_pattern, + (const unsigned char *)message, + size, + 0, /* start at offset 0 in the subject */ + 0, /* default options */ + md, + NULL); + if (r == PCRE2_ERROR_NOMATCH) + return false; + if (r < 0) { + unsigned char buf[LINE_MAX]; + + r = sym_pcre2_get_error_message(r, buf, sizeof(buf)); + return log_error_errno(SYNTHETIC_ERRNO(EINVAL), "Pattern matching failed: %s", + r < 0 ? "unknown error" : (char*) buf); + } + + if (ret_ovec) { + ret_ovec[0] = sym_pcre2_get_ovector_pointer(md)[0]; + ret_ovec[1] = sym_pcre2_get_ovector_pointer(md)[1]; + } + + return true; +#else + return log_error_errno(SYNTHETIC_ERRNO(EOPNOTSUPP), "PCRE2 support is not compiled in."); +#endif } + +void *pattern_free(pcre2_code *p) { +#if HAVE_PCRE2 + if (!p) + return NULL; + + assert(pcre2_dl); + sym_pcre2_code_free(p); + return NULL; +#else + assert(p == NULL); + return NULL; #endif +} |