Fix multibyte issue in ltree_strncasecmp().

Previously, the API for ltree_strncasecmp() took two inputs but only
one length (that of the smaller input). It truncated the larger input
to that length, but that could break a multibyte sequence.

Change the API to be a check for prefix equality (possibly
case-insensitive) instead, which is all that's needed by the
callers. Also, provide the lengths of both inputs.

Reviewed-by: Chao Li <li.evan.chao@gmail.com>
Reviewed-by: Peter Eisentraut <peter@eisentraut.org>
Discussion: https://postgr.es/m/5f65b85740197ba6249ea507cddf609f84a6188b.camel%40j-davis.com
Backpatch-through: 14

diff --git a/contrib/ltree/lquery_op.c b/contrib/ltree/lquery_op.c
index a6466f575fd7df486ec06c05412e563526a25e88..0b39d64a83973f00aa88ef6be1c7e27311f6a73e 100644 (file)
--- a/contrib/ltree/lquery_op.c
+++ b/contrib/ltree/lquery_op.c
@@ -41,7 +41,8 @@ getlexeme(char *start, char *end, int *len)
 }
 
 bool
-compare_subnode(ltree_level *t, char *qn, int len, int (*cmpptr) (const char *, const char *, size_t), bool anyend)
+compare_subnode(ltree_level *t, char *qn, int len,
+                               ltree_prefix_eq_func prefix_eq, bool anyend)
 {
        char       *endt = t->name + t->len;
        char       *endq = qn + len;
@@ -57,7 +58,7 @@ compare_subnode(ltree_level *t, char *qn, int len, int (*cmpptr) (const char *,
                while ((tn = getlexeme(tn, endt, &lent)) != NULL)
                {
                        if ((lent == lenq || (lent > lenq && anyend)) &&
-                               (*cmpptr) (qn, tn, lenq) == 0)
+                               (*prefix_eq) (qn, lenq, tn, lent))
                        {
 
                                isok = true;
@@ -74,14 +75,29 @@ compare_subnode(ltree_level *t, char *qn, int len, int (*cmpptr) (const char *,
        return true;
 }
 
-int
-ltree_strncasecmp(const char *a, const char *b, size_t s)
+/*
+ * Check if 'a' is a prefix of 'b'.
+ */
+bool
+ltree_prefix_eq(const char *a, size_t a_sz, const char *b, size_t b_sz)
+{
+       if (a_sz > b_sz)
+               return false;
+       else
+               return (strncmp(a, b, a_sz) == 0);
+}
+
+/*
+ * Case-insensitive check if 'a' is a prefix of 'b'.
+ */
+bool
+ltree_prefix_eq_ci(const char *a, size_t a_sz, const char *b, size_t b_sz)
 {
-       char       *al = str_tolower(a, s, DEFAULT_COLLATION_OID);
-       char       *bl = str_tolower(b, s, DEFAULT_COLLATION_OID);
-       int                     res;
+       char       *al = str_tolower(a, a_sz, DEFAULT_COLLATION_OID);
+       char       *bl = str_tolower(b, b_sz, DEFAULT_COLLATION_OID);
+       bool            res;
 
-       res = strncmp(al, bl, s);
+       res = (strncmp(al, bl, a_sz) == 0);
 
        pfree(al);
        pfree(bl);
@@ -109,19 +125,19 @@ checkLevel(lquery_level *curq, ltree_level *curt)
 
        for (int i = 0; i < curq->numvar; i++)
        {
-               int                     (*cmpptr) (const char *, const char *, size_t);
+               ltree_prefix_eq_func prefix_eq;
 
-               cmpptr = (curvar->flag & LVAR_INCASE) ? ltree_strncasecmp : strncmp;
+               prefix_eq = (curvar->flag & LVAR_INCASE) ? ltree_prefix_eq_ci : ltree_prefix_eq;
 
                if (curvar->flag & LVAR_SUBLEXEME)
                {
-                       if (compare_subnode(curt, curvar->name, curvar->len, cmpptr,
+                       if (compare_subnode(curt, curvar->name, curvar->len, prefix_eq,
                                                                (curvar->flag & LVAR_ANYEND)))
                                return success;
                }
                else if ((curvar->len == curt->len ||
                                  (curt->len > curvar->len && (curvar->flag & LVAR_ANYEND))) &&
-                                (*cmpptr) (curvar->name, curt->name, curvar->len) == 0)
+                                (*prefix_eq) (curvar->name, curvar->len, curt->name, curt->len))
                        return success;
 
                curvar = LVAR_NEXT(curvar);

diff --git a/contrib/ltree/ltree.h b/contrib/ltree/ltree.h
index 5e0761641d32ac5db8e925a6c34d699eb21705ec..78478dec173d47a392ab001441e5932ae7d5404a 100644 (file)
--- a/contrib/ltree/ltree.h
+++ b/contrib/ltree/ltree.h
@@ -157,6 +157,8 @@ typedef struct
        char            data[FLEXIBLE_ARRAY_MEMBER];
 } ltxtquery;
 
+typedef bool (*ltree_prefix_eq_func) (const char *, size_t, const char *, size_t);
+
 #define HDRSIZEQT              MAXALIGN(VARHDRSZ + sizeof(int32))
 #define COMPUTESIZE(size,lenofoperand) ( HDRSIZEQT + (size) * sizeof(ITEM) + (lenofoperand) )
 #define LTXTQUERY_TOO_BIG(size,lenofoperand) \
@@ -208,9 +210,10 @@ bool               ltree_execute(ITEM *curitem, void *checkval,
 int                    ltree_compare(const ltree *a, const ltree *b);
 bool           inner_isparent(const ltree *c, const ltree *p);
 bool           compare_subnode(ltree_level *t, char *qn, int len,
-                                                       int (*cmpptr) (const char *, const char *, size_t), bool anyend);
+                                                       ltree_prefix_eq_func prefix_eq, bool anyend);
 ltree     *lca_inner(ltree **a, int len);
-int                    ltree_strncasecmp(const char *a, const char *b, size_t s);
+bool           ltree_prefix_eq(const char *a, size_t a_sz, const char *b, size_t b_sz);
+bool           ltree_prefix_eq_ci(const char *a, size_t a_sz, const char *b, size_t b_sz);
 
 /* fmgr macros for ltree objects */
 #define DatumGetLtreeP(X)                      ((ltree *) PG_DETOAST_DATUM(X))

diff --git a/contrib/ltree/ltxtquery_op.c b/contrib/ltree/ltxtquery_op.c
index 002102c9c75b69466d79852448ad488a2905890a..3dcbab2c4846017bf3fa30ed8c28d73f40ef38ad 100644 (file)
--- a/contrib/ltree/ltxtquery_op.c
+++ b/contrib/ltree/ltxtquery_op.c
@@ -58,19 +58,19 @@ checkcondition_str(void *checkval, ITEM *val)
        ltree_level *level = LTREE_FIRST(((CHKVAL *) checkval)->node);
        int                     tlen = ((CHKVAL *) checkval)->node->numlevel;
        char       *op = ((CHKVAL *) checkval)->operand + val->distance;
-       int                     (*cmpptr) (const char *, const char *, size_t);
+       ltree_prefix_eq_func prefix_eq;
 
-       cmpptr = (val->flag & LVAR_INCASE) ? ltree_strncasecmp : strncmp;
+       prefix_eq = (val->flag & LVAR_INCASE) ? ltree_prefix_eq_ci : ltree_prefix_eq;
        while (tlen > 0)
        {
                if (val->flag & LVAR_SUBLEXEME)
                {
-                       if (compare_subnode(level, op, val->length, cmpptr, (val->flag & LVAR_ANYEND)))
+                       if (compare_subnode(level, op, val->length, prefix_eq, (val->flag & LVAR_ANYEND)))
                                return true;
                }
                else if ((val->length == level->len ||
                                  (level->len > val->length && (val->flag & LVAR_ANYEND))) &&
-                                (*cmpptr) (op, level->name, val->length) == 0)
+                                (*prefix_eq) (op, val->length, level->name, level->len))
                        return true;
 
                tlen--;