Fix bogus tree-flattening logic in QTNTernary().

QTNTernary() contains logic to flatten, eg, '(a & b) & c' into 'a & b & c',
which is all well and good, but it tries to do that to NOT nodes as well,
so that '!!a' gets changed to '!a'.  Explicitly restrict the conversion to
be done only on AND and OR nodes, and add a test case illustrating the bug.

In passing, provide some comments for the sadly naked functions in
tsquery_util.c, and simplify some baroque logic in QTNFree(), which
I think may have been leaking some items it intended to free.

Noted while investigating a complaint from Andreas Seltenreich.
Back-patch to all supported versions.

diff --git a/src/backend/utils/adt/tsquery_util.c b/src/backend/utils/adt/tsquery_util.c
index 0f338aa653d8e8a9336ad53ca0fc05f4cb9f4969..9add11f65741975bf6b5902e638a83709987c081 100644 (file)
--- a/src/backend/utils/adt/tsquery_util.c
+++ b/src/backend/utils/adt/tsquery_util.c
@@ -17,6 +17,9 @@
 #include "tsearch/ts_utils.h"
 #include "miscadmin.h"
 
+/*
+ * Build QTNode tree for a tsquery given in QueryItem array format.
+ */
 QTNode *
 QT2QTN(QueryItem *in, char *operand)
 {
@@ -50,6 +53,12 @@ QT2QTN(QueryItem *in, char *operand)
    return node;
 }
 
+/*
+ * Free a QTNode tree.
+ *
+ * Referenced "word" and "valnode" items are freed if marked as transient
+ * by flags.
+ */
 void
 QTNFree(QTNode *in)
 {
@@ -62,26 +71,27 @@ QTNFree(QTNode *in)
    if (in->valnode->type == QI_VAL && in->word && (in->flags & QTN_WORDFREE) != 0)
        pfree(in->word);
 
-   if (in->child)
+   if (in->valnode->type == QI_OPR)
    {
-       if (in->valnode)
-       {
-           if (in->valnode->type == QI_OPR && in->nchild > 0)
-           {
-               int         i;
-
-               for (i = 0; i < in->nchild; i++)
-                   QTNFree(in->child[i]);
-           }
-           if (in->flags & QTN_NEEDFREE)
-               pfree(in->valnode);
-       }
-       pfree(in->child);
+       int         i;
+
+       for (i = 0; i < in->nchild; i++)
+           QTNFree(in->child[i]);
    }
+   if (in->child)
+       pfree(in->child);
+
+   if (in->flags & QTN_NEEDFREE)
+       pfree(in->valnode);
 
    pfree(in);
 }
 
+/*
+ * Sort comparator for QTNodes.
+ *
+ * The sort order is somewhat arbitrary.
+ */
 int
 QTNodeCompare(QTNode *an, QTNode *bn)
 {
@@ -135,12 +145,19 @@ QTNodeCompare(QTNode *an, QTNode *bn)
    }
 }
 
+/*
+ * qsort comparator for QTNode pointers.
+ */
 static int
 cmpQTN(const void *a, const void *b)
 {
    return QTNodeCompare(*(QTNode *const *) a, *(QTNode *const *) b);
 }
 
+/*
+ * Canonicalize a QTNode tree by sorting the children of AND/OR nodes
+ * into an arbitrary but well-defined order.
+ */
 void
 QTNSort(QTNode *in)
 {
@@ -158,13 +175,16 @@ QTNSort(QTNode *in)
        qsort((void *) in->child, in->nchild, sizeof(QTNode *), cmpQTN);
 }
 
+/*
+ * Are two QTNode trees equal according to QTNodeCompare?
+ */
 bool
 QTNEq(QTNode *a, QTNode *b)
 {
    uint32      sign = a->sign & b->sign;
 
    if (!(sign == a->sign && sign == b->sign))
-       return 0;
+       return false;
 
    return (QTNodeCompare(a, b) == 0) ? true : false;
 }
@@ -190,14 +210,17 @@ QTNTernary(QTNode *in)
    for (i = 0; i < in->nchild; i++)
        QTNTernary(in->child[i]);
 
+   /* Only AND and OR are associative, so don't flatten other node types */
+   if (in->valnode->qoperator.oper != OP_AND &&
+       in->valnode->qoperator.oper != OP_OR)
+       return;
+
    for (i = 0; i < in->nchild; i++)
    {
        QTNode     *cc = in->child[i];
 
-       /* OP_Phrase isn't associative */
        if (cc->valnode->type == QI_OPR &&
-           in->valnode->qoperator.oper == cc->valnode->qoperator.oper &&
-           in->valnode->qoperator.oper != OP_PHRASE)
+           in->valnode->qoperator.oper == cc->valnode->qoperator.oper)
        {
            int         oldnchild = in->nchild;
 
@@ -236,9 +259,6 @@ QTNBinary(QTNode *in)
    for (i = 0; i < in->nchild; i++)
        QTNBinary(in->child[i]);
 
-   if (in->nchild <= 2)
-       return;
-
    while (in->nchild > 2)
    {
        QTNode     *nn = (QTNode *) palloc0(sizeof(QTNode));
@@ -263,8 +283,9 @@ QTNBinary(QTNode *in)
 }
 
 /*
- * Count the total length of operand string in tree, including '\0'-
- * terminators.
+ * Count the total length of operand strings in tree (including '\0'-
+ * terminators) and the total number of nodes.
+ * Caller must initialize *sumlen and *nnode to zeroes.
  */
 static void
 cntsize(QTNode *in, int *sumlen, int *nnode)
@@ -293,6 +314,10 @@ typedef struct
    char       *curoperand;
 } QTN2QTState;
 
+/*
+ * Recursively convert a QTNode tree into flat tsquery format.
+ * Caller must have allocated arrays of the correct size.
+ */
 static void
 fillQT(QTN2QTState *state, QTNode *in)
 {
@@ -330,6 +355,9 @@ fillQT(QTN2QTState *state, QTNode *in)
    }
 }
 
+/*
+ * Build flat tsquery from a QTNode tree.
+ */
 TSQuery
 QTN2QT(QTNode *in)
 {
@@ -358,6 +386,11 @@ QTN2QT(QTNode *in)
    return out;
 }
 
+/*
+ * Copy a QTNode tree.
+ *
+ * Modifiable copies of the words and valnodes are made, too.
+ */
 QTNode *
 QTNCopy(QTNode *in)
 {
@@ -393,6 +426,9 @@ QTNCopy(QTNode *in)
    return out;
 }
 
+/*
+ * Clear the specified flag bit(s) in all nodes of a QTNode tree.
+ */
 void
 QTNClearFlags(QTNode *in, uint32 flags)
 {

diff --git a/src/test/regress/expected/tsearch.out b/src/test/regress/expected/tsearch.out
index 129d06ef07ea430a0e05603ffc7849fcb5a79a4a..55d6a85738785e3bc1f9458ec9a4f6b610b5c508 100644 (file)
--- a/src/test/regress/expected/tsearch.out
+++ b/src/test/regress/expected/tsearch.out
@@ -1184,6 +1184,13 @@ SELECT ts_rewrite('foo & bar & qq & new & york',  'new & york'::tsquery, 'big &
  'foo' & 'bar' & 'qq' & ( 'city' & 'new' & 'york' | 'nyc' | 'big' & 'apple' )
 (1 row)
 
+SELECT ts_rewrite(ts_rewrite('new & !york ', 'york', '!jersey'),
+                  'jersey', 'mexico');
+     ts_rewrite     
+--------------------
+ 'new' & !!'mexico'
+(1 row)
+
 SELECT ts_rewrite('moscow', 'SELECT keyword, sample FROM test_tsquery'::text );
      ts_rewrite      
 ---------------------

diff --git a/src/test/regress/sql/tsearch.sql b/src/test/regress/sql/tsearch.sql
index 65eb9438d4941212552c94ec05356ebd0c3ead5f..afd990e6969197ca0e95b1263b204090f0734934 100644 (file)
--- a/src/test/regress/sql/tsearch.sql
+++ b/src/test/regress/sql/tsearch.sql
@@ -402,6 +402,8 @@ SELECT COUNT(*) FROM test_tsquery WHERE keyword >  'new & york';
 RESET enable_seqscan;
 
 SELECT ts_rewrite('foo & bar & qq & new & york',  'new & york'::tsquery, 'big & apple | nyc | new & york & city');
+SELECT ts_rewrite(ts_rewrite('new & !york ', 'york', '!jersey'),
+                  'jersey', 'mexico');
 
 SELECT ts_rewrite('moscow', 'SELECT keyword, sample FROM test_tsquery'::text );
 SELECT ts_rewrite('moscow & hotel', 'SELECT keyword, sample FROM test_tsquery'::text );