diff --git a/lib/common/strings.c b/lib/common/strings.c index e24df546cc..106eb276f2 100644 --- a/lib/common/strings.c +++ b/lib/common/strings.c @@ -1,513 +1,517 @@ /* * Copyright 2004-2019 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #ifndef _GNU_SOURCE # define _GNU_SOURCE #endif #include #include #include #include #include #include char * crm_itoa_stack(int an_int, char *buffer, size_t len) { if (buffer != NULL) { snprintf(buffer, len, "%d", an_int); } return buffer; } long long crm_int_helper(const char *text, char **end_text) { long long result = -1; char *local_end_text = NULL; int saved_errno = 0; errno = 0; if (text != NULL) { #ifdef ANSI_ONLY if (end_text != NULL) { result = strtol(text, end_text, 10); } else { result = strtol(text, &local_end_text, 10); } #else if (end_text != NULL) { result = strtoll(text, end_text, 10); } else { result = strtoll(text, &local_end_text, 10); } #endif saved_errno = errno; if (errno == EINVAL) { crm_err("Conversion of %s failed", text); result = -1; } else if (errno == ERANGE) { crm_err("Conversion of %s was clipped: %lld", text, result); } else if (errno != 0) { crm_perror(LOG_ERR, "Conversion of %s failed", text); + + } else if (local_end_text == text) { + crm_err("Text contained no digits: %s", text); + result = -1; } if (local_end_text != NULL && local_end_text[0] != '\0') { crm_err("Characters left over after parsing '%s': '%s'", text, local_end_text); } errno = saved_errno; } return result; } /*! * \brief Parse a long long integer value from a string * * \param[in] text The string to parse * \param[in] default_text Default string to parse if text is NULL * * \return Parsed value on success, -1 (and set errno) on error */ long long crm_parse_ll(const char *text, const char *default_text) { if (text == NULL) { text = default_text; if (text == NULL) { crm_err("No default conversion value supplied"); errno = EINVAL; return -1; } } return crm_int_helper(text, NULL); } /*! * \brief Parse an integer value from a string * * \param[in] text The string to parse * \param[in] default_text Default string to parse if text is NULL * * \return Parsed value on success, INT_MIN or INT_MAX (and set errno to ERANGE) * if parsed value is out of integer range, otherwise -1 (and set errno) */ int crm_parse_int(const char *text, const char *default_text) { long long result = crm_parse_ll(text, default_text); if (result < INT_MIN) { // If errno is ERANGE, crm_parse_ll() has already logged a message if (errno != ERANGE) { crm_err("Conversion of %s was clipped: %lld", text, result); errno = ERANGE; } return INT_MIN; } else if (result > INT_MAX) { // If errno is ERANGE, crm_parse_ll() has already logged a message if (errno != ERANGE) { crm_err("Conversion of %s was clipped: %lld", text, result); errno = ERANGE; } return INT_MAX; } return (int) result; } /*! * \internal * \brief Parse a milliseconds value (without units) from a string * * \param[in] text String to parse * * \return Milliseconds on success, 0 otherwise (and errno will be set) */ guint crm_parse_ms(const char *text) { if (text) { long long ms = crm_int_helper(text, NULL); if ((ms < 0) || (ms > G_MAXUINT)) { errno = ERANGE; } return errno? 0 : (guint) ms; } return 0; } gboolean safe_str_neq(const char *a, const char *b) { if (a == b) { return FALSE; } else if (a == NULL || b == NULL) { return TRUE; } else if (strcasecmp(a, b) == 0) { return FALSE; } return TRUE; } gboolean crm_is_true(const char *s) { gboolean ret = FALSE; if (s != NULL) { crm_str_to_boolean(s, &ret); } return ret; } int crm_str_to_boolean(const char *s, int *ret) { if (s == NULL) { return -1; } else if (strcasecmp(s, "true") == 0 || strcasecmp(s, "on") == 0 || strcasecmp(s, "yes") == 0 || strcasecmp(s, "y") == 0 || strcasecmp(s, "1") == 0) { *ret = TRUE; return 1; } else if (strcasecmp(s, "false") == 0 || strcasecmp(s, "off") == 0 || strcasecmp(s, "no") == 0 || strcasecmp(s, "n") == 0 || strcasecmp(s, "0") == 0) { *ret = FALSE; return 1; } return -1; } char * crm_strip_trailing_newline(char *str) { int len; if (str == NULL) { return str; } for (len = strlen(str) - 1; len >= 0 && str[len] == '\n'; len--) { str[len] = '\0'; } return str; } gboolean crm_str_eq(const char *a, const char *b, gboolean use_case) { if (use_case) { return g_strcmp0(a, b) == 0; /* TODO - Figure out which calls, if any, really need to be case independent */ } else if (a == b) { return TRUE; } else if (a == NULL || b == NULL) { /* shouldn't be comparing NULLs */ return FALSE; } else if (strcasecmp(a, b) == 0) { return TRUE; } return FALSE; } static inline const char * null2emptystr(const char *); static inline const char * null2emptystr(const char *input) { return (input == NULL) ? "" : input; } /*! * \brief Check whether a string starts with a certain sequence * * \param[in] str String to check * \param[in] match Sequence to match against beginning of \p str * * \return \c TRUE if \p str begins with match, \c FALSE otherwise * \note This is equivalent to !strncmp(s, prefix, strlen(prefix)) * but is likely less efficient when prefix is a string literal * if the compiler optimizes away the strlen() at compile time, * and more efficient otherwise. */ bool crm_starts_with(const char *str, const char *prefix) { const char *s = str; const char *p = prefix; if (!s || !p) { return FALSE; } while (*s && *p) { if (*s++ != *p++) { return FALSE; } } return (*p == 0); } static inline int crm_ends_with_internal(const char *, const char *, gboolean); static inline int crm_ends_with_internal(const char *s, const char *match, gboolean as_extension) { if ((s == NULL) || (match == NULL)) { return 0; } else { size_t slen, mlen; if (match[0] != '\0' && (as_extension /* following commented out for inefficiency: || strchr(&match[1], match[0]) == NULL */)) return !strcmp(null2emptystr(strrchr(s, match[0])), match); if ((mlen = strlen(match)) == 0) return 1; slen = strlen(s); return ((slen >= mlen) && !strcmp(s + slen - mlen, match)); } } /*! * \internal * \brief Check whether a string ends with a certain sequence * * \param[in] s String to check * \param[in] match Sequence to match against end of \p s * * \return \c TRUE if \p s ends (verbatim, i.e., case sensitively) * with match (including empty string), \c FALSE otherwise * * \see crm_ends_with_ext() */ gboolean crm_ends_with(const char *s, const char *match) { return crm_ends_with_internal(s, match, FALSE); } /*! * \internal * \brief Check whether a string ends with a certain "extension" * * \param[in] s String to check * \param[in] match Extension to match against end of \p s, that is, * its first character must not occur anywhere * in the rest of that very sequence (example: file * extension where the last dot is its delimiter, * e.g., ".html"); incorrect results may be * returned otherwise. * * \return \c TRUE if \p s ends (verbatim, i.e., case sensitively) * with "extension" designated as \p match (including empty * string), \c FALSE otherwise * * \note Main incentive to prefer this function over \c crm_ends_with * where possible is the efficiency (at the cost of added * restriction on \p match as stated; the complexity class * remains the same, though: BigO(M+N) vs. BigO(M+2N)). * * \see crm_ends_with() */ gboolean crm_ends_with_ext(const char *s, const char *match) { return crm_ends_with_internal(s, match, TRUE); } /* * This re-implements g_str_hash as it was prior to glib2-2.28: * * https://gitlab.gnome.org/GNOME/glib/commit/354d655ba8a54b754cb5a3efb42767327775696c * * Note that the new g_str_hash is presumably a *better* hash (it's actually * a correct implementation of DJB's hash), but we need to preserve existing * behaviour, because the hash key ultimately determines the "sort" order * when iterating through GHashTables, which affects allocation of scores to * clone instances when iterating through rsc->allowed_nodes. It (somehow) * also appears to have some minor impact on the ordering of a few * pseudo_event IDs in the transition graph. */ guint g_str_hash_traditional(gconstpointer v) { const signed char *p; guint32 h = 0; for (p = v; *p != '\0'; p++) h = (h << 5) - h + *p; return h; } /* used with hash tables where case does not matter */ gboolean crm_strcase_equal(gconstpointer a, gconstpointer b) { return crm_str_eq((const char *) a, (const char *) b, FALSE); } guint crm_strcase_hash(gconstpointer v) { const signed char *p; guint32 h = 0; for (p = v; *p != '\0'; p++) h = (h << 5) - h + g_ascii_tolower(*p); return h; } static void copy_str_table_entry(gpointer key, gpointer value, gpointer user_data) { if (key && value && user_data) { g_hash_table_insert((GHashTable*)user_data, strdup(key), strdup(value)); } } GHashTable * crm_str_table_dup(GHashTable *old_table) { GHashTable *new_table = NULL; if (old_table) { new_table = crm_str_table_new(); g_hash_table_foreach(old_table, copy_str_table_entry, new_table); } return new_table; } char * add_list_element(char *list, const char *value) { int len = 0; int last = 0; if (value == NULL) { return list; } if (list) { last = strlen(list); } len = last + 2; /* +1 space, +1 EOS */ len += strlen(value); list = realloc_safe(list, len); sprintf(list + last, " %s", value); return list; } bool crm_compress_string(const char *data, int length, int max, char **result, unsigned int *result_len) { int rc; char *compressed = NULL; char *uncompressed = strdup(data); #ifdef CLOCK_MONOTONIC struct timespec after_t; struct timespec before_t; #endif if(max == 0) { max = (length * 1.1) + 600; /* recommended size */ } #ifdef CLOCK_MONOTONIC clock_gettime(CLOCK_MONOTONIC, &before_t); #endif compressed = calloc(max, sizeof(char)); CRM_ASSERT(compressed); *result_len = max; rc = BZ2_bzBuffToBuffCompress(compressed, result_len, uncompressed, length, CRM_BZ2_BLOCKS, 0, CRM_BZ2_WORK); free(uncompressed); if (rc != BZ_OK) { crm_err("Compression of %d bytes failed: %s " CRM_XS " bzerror=%d", length, bz2_strerror(rc), rc); free(compressed); return FALSE; } #ifdef CLOCK_MONOTONIC clock_gettime(CLOCK_MONOTONIC, &after_t); crm_trace("Compressed %d bytes into %d (ratio %d:1) in %.0fms", length, *result_len, length / (*result_len), difftime (after_t.tv_sec, before_t.tv_sec) * 1000 + (after_t.tv_nsec - before_t.tv_nsec) / 1e6); #else crm_trace("Compressed %d bytes into %d (ratio %d:1)", length, *result_len, length / (*result_len)); #endif *result = compressed; return TRUE; } /*! * \brief Compare two strings alphabetically (case-insensitive) * * \param[in] a First string to compare * \param[in] b Second string to compare * * \return 0 if strings are equal, -1 if a < b, 1 if a > b * * \note Usable as a GCompareFunc with g_list_sort(). * NULL is considered less than non-NULL. */ gint crm_alpha_sort(gconstpointer a, gconstpointer b) { if (!a && !b) { return 0; } else if (!a) { return -1; } else if (!b) { return 1; } return strcasecmp(a, b); } char * crm_strdup_printf(char const *format, ...) { va_list ap; int len = 0; char *string = NULL; va_start(ap, format); len = vasprintf (&string, format, ap); CRM_ASSERT(len > 0); va_end(ap); return string; }