diff --git a/src/backend/utils/adt/formatting.c b/src/backend/utils/adt/formatting.c index ca3c48d..3ef10dc 100644 --- a/src/backend/utils/adt/formatting.c +++ b/src/backend/utils/adt/formatting.c @@ -317,10 +317,6 @@ static const char *const numth[] = {"st", "nd", "rd", "th", NULL}; * Flags & Options: * ---------- */ -#define ONE_UPPER 1 /* Name */ -#define ALL_UPPER 2 /* NAME */ -#define ALL_LOWER 3 /* name */ - #define MAX_MONTH_LEN 9 #define MAX_MON_LEN 3 #define MAX_DAY_LEN 9 @@ -1068,9 +1064,9 @@ static int from_char_parse_int_len(int *dest, char **src, const int len, FormatNode *node, bool *have_error); static int from_char_parse_int(int *dest, char **src, FormatNode *node, bool *have_error); -static int seq_search(char *name, const char *const *array, int type, int max, int *len); +static int seq_search(const char *name, const char *const *array, int *len); static int from_char_seq_search(int *dest, char **src, - const char *const *array, int type, int max, + const char *const *array, int max, FormatNode *node, bool *have_error); static void do_to_timestamp(text *date_txt, text *fmt, bool std, struct pg_tm *tm, fsec_t *fsec, int *fprec, @@ -2454,70 +2450,57 @@ from_char_parse_int(int *dest, char **src, FormatNode *node, bool *have_error) } /* ---------- - * Sequential search with to upper/lower conversion + * Sequentially search null-terminated "array" for a case-insensitive match + * to the initial character(s) of "name". + * + * Returns array index of match, or -1 for no match. + * + * *len is set to the length of the match, or 0 for no match. + * + * Case-insensitivity is defined per pg_ascii_toupper, so this is only + * suitable for comparisons to ASCII strings. * ---------- */ static int -seq_search(char *name, const char *const *array, int type, int max, int *len) +seq_search(const char *name, const char *const *array, int *len) { - const char *p; + unsigned char firstc; const char *const *a; - char *n; - int last, - i; *len = 0; + /* empty string can't match anything */ if (!*name) return -1; - /* set first char */ - if (type == ONE_UPPER || type == ALL_UPPER) - *name = pg_toupper((unsigned char) *name); - else if (type == ALL_LOWER) - *name = pg_tolower((unsigned char) *name); + /* we handle first char specially to gain some speed */ + firstc = pg_ascii_toupper((unsigned char) *name); - for (last = 0, a = array; *a != NULL; a++) + for (a = array; *a != NULL; a++) { + int i; + const char *p; + const char *n; + /* compare first chars */ - if (*name != **a) + if (pg_ascii_toupper((unsigned char) **a) != firstc) continue; for (i = 1, p = *a + 1, n = name + 1;; n++, p++, i++) { - /* search fragment (max) only */ - if (max && i == max) - { - *len = i; - return a - array; - } - /* full size */ + /* return success if we matched whole array entry */ if (*p == '\0') { *len = i; return a - array; } - /* Not found in array 'a' */ + /* else, must have another character in "name" ... */ if (*n == '\0') break; - /* - * Convert (but convert new chars only) - */ - if (i > last) - { - if (type == ONE_UPPER || type == ALL_LOWER) - *n = pg_tolower((unsigned char) *n); - else if (type == ALL_UPPER) - *n = pg_toupper((unsigned char) *n); - last = i; - } - -#ifdef DEBUG_TO_FROM_CHAR - elog(DEBUG_elog_output, "N: %c, P: %c, A: %s (%s)", - *n, *p, *a, name); -#endif - if (*n != *p) + /* ... and it must match */ + if (pg_ascii_toupper((unsigned char) *p) != + pg_ascii_toupper((unsigned char) *n)) break; } } @@ -2526,8 +2509,8 @@ seq_search(char *name, const char *const *array, int type, int max, int *len) } /* - * Perform a sequential search in 'array' for text matching the first 'max' - * characters of the source string. + * Perform a sequential search in 'array' for an entry matching the first + * character(s) of the 'src' string case-insensitively. * * If a match is found, copy the array index of the match into the integer * pointed to by 'dest', advance 'src' to the end of the part of the string @@ -2535,19 +2518,28 @@ seq_search(char *name, const char *const *array, int type, int max, int *len) * * If the string doesn't match, throw an error if 'have_error' is NULL, * otherwise set '*have_error' and return -1. + * + * 'max' and 'node' are used only for error reports: 'max' is the max + * number of bytes of the string to include in the error report, and + * node->key->name identifies what we were searching for. */ static int -from_char_seq_search(int *dest, char **src, const char *const *array, int type, +from_char_seq_search(int *dest, char **src, const char *const *array, int max, FormatNode *node, bool *have_error) { int len; - *dest = seq_search(*src, array, type, max, &len); + /* Assert that no call gives us a 'max' too large for the error buffer */ + Assert(max <= DCH_MAX_ITEM_SIZ); + + *dest = seq_search(*src, array, &len); + if (len <= 0) { char copy[DCH_MAX_ITEM_SIZ + 1]; - Assert(max <= DCH_MAX_ITEM_SIZ); + /* Use multibyte-aware truncation to avoid generating a bogus string */ + max = pg_mbcliplen(*src, strlen(*src), max); strlcpy(copy, *src, max + 1); RETURN_ERROR(ereport(ERROR, @@ -3279,7 +3271,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, case DCH_a_m: case DCH_p_m: from_char_seq_search(&value, &s, ampm_strings_long, - ALL_UPPER, n->key->len, n, have_error); + n->key->len, n, have_error); CHECK_ERROR; from_char_set_int(&out->pm, value % 2, n, have_error); CHECK_ERROR; @@ -3290,7 +3282,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, case DCH_am: case DCH_pm: from_char_seq_search(&value, &s, ampm_strings, - ALL_UPPER, n->key->len, n, have_error); + n->key->len, n, have_error); CHECK_ERROR; from_char_set_int(&out->pm, value % 2, n, have_error); CHECK_ERROR; @@ -3403,7 +3395,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, case DCH_a_d: case DCH_b_c: from_char_seq_search(&value, &s, adbc_strings_long, - ALL_UPPER, n->key->len, n, have_error); + n->key->len, n, have_error); CHECK_ERROR; from_char_set_int(&out->bc, value % 2, n, have_error); CHECK_ERROR; @@ -3413,7 +3405,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, case DCH_ad: case DCH_bc: from_char_seq_search(&value, &s, adbc_strings, - ALL_UPPER, n->key->len, n, have_error); + n->key->len, n, have_error); CHECK_ERROR; from_char_set_int(&out->bc, value % 2, n, have_error); CHECK_ERROR; @@ -3421,7 +3413,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, case DCH_MONTH: case DCH_Month: case DCH_month: - from_char_seq_search(&value, &s, months_full, ONE_UPPER, + from_char_seq_search(&value, &s, months_full, MAX_MONTH_LEN, n, have_error); CHECK_ERROR; from_char_set_int(&out->mm, value + 1, n, have_error); @@ -3430,7 +3422,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, case DCH_MON: case DCH_Mon: case DCH_mon: - from_char_seq_search(&value, &s, months, ONE_UPPER, + from_char_seq_search(&value, &s, months, MAX_MON_LEN, n, have_error); CHECK_ERROR; from_char_set_int(&out->mm, value + 1, n, have_error); @@ -3444,7 +3436,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, case DCH_DAY: case DCH_Day: case DCH_day: - from_char_seq_search(&value, &s, days, ONE_UPPER, + from_char_seq_search(&value, &s, days, MAX_DAY_LEN, n, have_error); CHECK_ERROR; from_char_set_int(&out->d, value, n, have_error); @@ -3454,7 +3446,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, case DCH_DY: case DCH_Dy: case DCH_dy: - from_char_seq_search(&value, &s, days, ONE_UPPER, + from_char_seq_search(&value, &s, days, MAX_DY_LEN, n, have_error); CHECK_ERROR; from_char_set_int(&out->d, value, n, have_error); @@ -3572,7 +3564,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, break; case DCH_RM: from_char_seq_search(&value, &s, rm_months_upper, - ALL_UPPER, MAX_RM_LEN, n, have_error); + MAX_RM_LEN, n, have_error); CHECK_ERROR; from_char_set_int(&out->mm, MONTHS_PER_YEAR - value, n, have_error); @@ -3580,7 +3572,7 @@ DCH_from_char(FormatNode *node, char *in, TmFromChar *out, bool std, break; case DCH_rm: from_char_seq_search(&value, &s, rm_months_lower, - ALL_LOWER, MAX_RM_LEN, n, have_error); + MAX_RM_LEN, n, have_error); CHECK_ERROR; from_char_set_int(&out->mm, MONTHS_PER_YEAR - value, n, have_error);