1.*.4|3|2.*{1}
(1 row)
+SELECT 'foo.bar{,}.!a*|b{1,}.c{,44}.d{3,4}'::lquery;
+ lquery
+------------------------------------
+ foo.bar{,}.!a*|b{1,}.c{,44}.d{3,4}
+(1 row)
+
SELECT 'qwerty%@*.tu'::lquery;
lquery
--------------
SELECT 'a.b.c.d.e'::ltree ~ '*.!d.*';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '*.!d';
SELECT 'a.b.c.d.e'::ltree ~ '*.!e.*';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ 'a.*.!e';
SELECT 'a.b.c.d.e'::ltree ~ 'a.*.!d.*';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ 'a.*.!f.*';
SELECT 'a.b.c.d.e'::ltree ~ '*.a.*.!d.*';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '*.a.!d.*';
SELECT 'a.b.c.d.e'::ltree ~ '*.a.*.!d.*';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '*.!b.*';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '*.!b.c.*';
SELECT 'a.b.c.d.e'::ltree ~ '*.!b.*.c.*';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '!b.*.c.*';
SELECT 'a.b.c.d.e'::ltree ~ '*{1}.!b.*{1}.!c.*.e';
?column?
----------
- t
+ f
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ 'a.!b.*{1}.!c.*.e';
?column?
----------
- t
+ f
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '!b.*{1}.!c.*.e';
?column?
----------
- t
+ f
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '*.!b.*{1}.!c.*.e';
?column?
----------
- t
+ f
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '*.!b.*.!c.*.e';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '!b.!c.*';
SELECT 'a.b.c.d.e'::ltree ~ '*{1}.!b.*{1}.!c.*';
?column?
----------
- t
+ f
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ 'a.!b.*{1}.!c.*';
?column?
----------
- t
+ f
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '!b.*{1}.!c.*';
?column?
----------
- t
+ f
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ '*.!b.*{1}.!c.*';
SELECT 'a.b.c.d.e'::ltree ~ '*.!b.*.!c.*';
?column?
----------
- f
+ t
(1 row)
SELECT 'a.b.c.d.e'::ltree ~ 'a.*{2}.*{2}';
f
(1 row)
+SELECT '5.0.1.0'::ltree ~ '5.!0.!0.0';
+ ?column?
+----------
+ f
+(1 row)
+
+SELECT 'a.b'::ltree ~ '!a.!a';
+ ?column?
+----------
+ f
+(1 row)
+
+SELECT 'a.b.c.d.e'::ltree ~ 'a{,}';
+ ?column?
+----------
+ f
+(1 row)
+
+SELECT 'a.b.c.d.e'::ltree ~ 'a{1,}.*';
+ ?column?
+----------
+ t
+(1 row)
+
+SELECT 'a.b.c.d.e'::ltree ~ 'a{,}.!a{,}';
+ ?column?
+----------
+ t
+(1 row)
+
+SELECT 'a.b.c.d.a'::ltree ~ 'a{,}.!a{,}';
+ ?column?
+----------
+ f
+(1 row)
+
+SELECT 'a.b.c.d.a'::ltree ~ 'a{,2}.!a{1,}';
+ ?column?
+----------
+ f
+(1 row)
+
+SELECT 'a.b.c.d.e'::ltree ~ 'a{,2}.!a{1,}';
+ ?column?
+----------
+ t
+(1 row)
+
+SELECT 'a.b.c.d.e'::ltree ~ '!x{,}';
+ ?column?
+----------
+ t
+(1 row)
+
+SELECT 'a.b.c.d.e'::ltree ~ '!c{,}';
+ ?column?
+----------
+ f
+(1 row)
+
+SELECT 'a.b.c.d.e'::ltree ~ '!c{0,3}.!a{2,}';
+ ?column?
+----------
+ t
+(1 row)
+
+SELECT 'a.b.c.d.e'::ltree ~ '!c{0,3}.!d{2,}.*';
+ ?column?
+----------
+ t
+(1 row)
+
SELECT 'QWER_TY'::ltree ~ 'q%@*';
?column?
----------
#include "catalog/pg_collation.h"
#include "ltree.h"
+#include "miscadmin.h"
#include "utils/formatting.h"
PG_FUNCTION_INFO_V1(ltq_regex);
#define NEXTVAL(x) ( (lquery*)( (char*)(x) + INTALIGN( VARSIZE(x) ) ) )
-typedef struct
-{
- lquery_level *q;
- int nq;
- ltree_level *t;
- int nt;
- int posq;
- int post;
-} FieldNot;
-
static char *
getlexeme(char *start, char *end, int *len)
{
}
/*
- * See if a (non-star) lquery_level matches an ltree_level
+ * See if an lquery_level matches an ltree_level
*
- * Does not consider level's possible LQL_NOT flag.
+ * This accounts for all flags including LQL_NOT, but does not
+ * consider repetition counts.
*/
static bool
checkLevel(lquery_level *curq, ltree_level *curt)
{
- int (*cmpptr) (const char *, const char *, size_t);
lquery_variant *curvar = LQL_FIRST(curq);
- int i;
+ bool success;
+
+ success = (curq->flag & LQL_NOT) ? false : true;
+
+ /* numvar == 0 means '*' which matches anything */
+ if (curq->numvar == 0)
+ return success;
- for (i = 0; i < curq->numvar; i++)
+ for (int i = 0; i < curq->numvar; i++)
{
+ int (*cmpptr) (const char *, const char *, size_t);
+
cmpptr = (curvar->flag & LVAR_INCASE) ? ltree_strncasecmp : strncmp;
if (curvar->flag & LVAR_SUBLEXEME)
{
- if (compare_subnode(curt, curvar->name, curvar->len, cmpptr, (curvar->flag & LVAR_ANYEND)))
- return true;
+ if (compare_subnode(curt, curvar->name, curvar->len, cmpptr,
+ (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)
- {
+ return success;
- return true;
- }
curvar = LVAR_NEXT(curvar);
}
- return false;
+ return !success;
}
/*
-void
-printFieldNot(FieldNot *fn ) {
- while(fn->q) {
- elog(NOTICE,"posQ:%d lenQ:%d posT:%d lenT:%d", fn->posq,fn->nq,fn->post,fn->nt);
- fn++;
- }
-}
-*/
-
-/*
- * Try to match an lquery (of query_numlevel items) to an ltree (of
- * tree_numlevel items)
- *
- * If the query contains any NOT flags, "ptr" must point to a FieldNot
- * workspace initialized with ptr->q == NULL. Otherwise it can be NULL.
- * (LQL_NOT flags will be ignored if ptr == NULL.)
- *
- * high_pos is the last ltree position the first lquery item is allowed
- * to match at; it should be zero for external calls.
- *
- * force_advance must be false except in internal recursive calls.
+ * Try to match an lquery (of qlen items) to an ltree (of tlen items)
*/
static bool
-checkCond(lquery_level *curq, int query_numlevel,
- ltree_level *curt, int tree_numlevel,
- FieldNot *ptr,
- uint32 high_pos,
- bool force_advance)
+checkCond(lquery_level *curq, int qlen,
+ ltree_level *curt, int tlen)
{
- uint32 low_pos = 0, /* first allowed ltree position for match */
- cur_tpos = 0; /* ltree position of curt */
- int tlen = tree_numlevel, /* counts of remaining items */
- qlen = query_numlevel;
- lquery_level *prevq = NULL;
-
- /* advance curq (setting up prevq) if requested */
- if (force_advance)
- {
- Assert(qlen > 0);
- prevq = curq;
- curq = LQL_NEXT(curq);
- qlen--;
- }
+ /* Since this function recurses, it could be driven to stack overflow */
+ check_stack_depth();
- while (tlen > 0 && qlen > 0)
- {
- if (curq->numvar)
- {
- /* Current query item is not '*' */
- ltree_level *prevt = curt;
-
- /* skip tree items that must be ignored due to prior * items */
- while (cur_tpos < low_pos)
- {
- curt = LEVEL_NEXT(curt);
- tlen--;
- cur_tpos++;
- if (tlen == 0)
- return false;
- if (ptr && ptr->q)
- ptr->nt++;
- }
+ /* Pathological patterns could take awhile, too */
+ CHECK_FOR_INTERRUPTS();
- if (ptr && (curq->flag & LQL_NOT))
- {
- /* Deal with a NOT item */
- if (!(prevq && prevq->numvar == 0))
- prevq = curq;
- if (ptr->q == NULL)
- {
- ptr->t = prevt;
- ptr->q = prevq;
- ptr->nt = 1;
- ptr->nq = 1 + ((prevq == curq) ? 0 : 1);
- ptr->posq = query_numlevel - qlen - ((prevq == curq) ? 0 : 1);
- ptr->post = cur_tpos;
- }
- else
- {
- ptr->nt++;
- ptr->nq++;
- }
-
- if (qlen == 1 && ptr->q->numvar == 0)
- ptr->nt = tree_numlevel - ptr->post;
- curt = LEVEL_NEXT(curt);
- tlen--;
- cur_tpos++;
- if (high_pos < cur_tpos)
- high_pos++;
- }
- else
- {
- /* Not a NOT item, check for normal match */
- bool isok = false;
-
- while (cur_tpos <= high_pos && tlen > 0 && !isok)
- {
- isok = checkLevel(curq, curt);
- curt = LEVEL_NEXT(curt);
- tlen--;
- cur_tpos++;
- if (isok && prevq && prevq->numvar == 0 &&
- tlen > 0 && cur_tpos <= high_pos)
- {
- FieldNot tmpptr;
-
- if (ptr)
- memcpy(&tmpptr, ptr, sizeof(FieldNot));
- if (checkCond(prevq, qlen + 1,
- curt, tlen,
- (ptr) ? &tmpptr : NULL,
- high_pos - cur_tpos,
- true))
- return true;
- }
- if (!isok && ptr && ptr->q)
- ptr->nt++;
- }
- if (!isok)
- return false;
-
- if (ptr && ptr->q)
- {
- if (checkCond(ptr->q, ptr->nq,
- ptr->t, ptr->nt,
- NULL,
- 0,
- false))
- return false;
- ptr->q = NULL;
- }
- low_pos = cur_tpos;
- high_pos = cur_tpos;
- }
- }
+ /* Loop while we have query items to consider */
+ while (qlen > 0)
+ {
+ int low,
+ high;
+ lquery_level *nextq;
+
+ /*
+ * Get min and max repetition counts for this query item, dealing with
+ * the backwards-compatibility hack that the low/high fields aren't
+ * meaningful for non-'*' items unless LQL_COUNT is set.
+ */
+ if ((curq->flag & LQL_COUNT) || curq->numvar == 0)
+ low = curq->low, high = curq->high;
else
- {
- /* Current query item is '*' */
- low_pos += curq->low;
-
- if (low_pos > tree_numlevel)
- return false;
+ low = high = 1;
- high_pos = Min(high_pos + curq->high, tree_numlevel);
+ /*
+ * We may limit "high" to the remaining text length; this avoids
+ * separate tests below.
+ */
+ if (high > tlen)
+ high = tlen;
- if (ptr && ptr->q)
- {
- ptr->nq++;
- if (qlen == 1)
- ptr->nt = tree_numlevel - ptr->post;
- }
- }
+ /* Fail if a match of required number of items is impossible */
+ if (high < low)
+ return false;
- prevq = curq;
- curq = LQL_NEXT(curq);
+ /*
+ * Recursively check the rest of the pattern against each possible
+ * start point following some of this item's match(es).
+ */
+ nextq = LQL_NEXT(curq);
qlen--;
- }
- /* Fail if we've already run out of ltree items */
- if (low_pos > tree_numlevel || tree_numlevel > high_pos)
- return false;
-
- /* Remaining lquery items must be NOT or '*' items */
- while (qlen > 0)
- {
- if (curq->numvar)
- {
- if (!(curq->flag & LQL_NOT))
- return false;
- }
- else
+ for (int matchcnt = 0; matchcnt < high; matchcnt++)
{
- low_pos += curq->low;
+ /*
+ * If we've consumed an acceptable number of matches of this item,
+ * and the rest of the pattern matches beginning here, we're good.
+ */
+ if (matchcnt >= low && checkCond(nextq, qlen, curt, tlen))
+ return true;
- if (low_pos > tree_numlevel)
+ /*
+ * Otherwise, try to match one more text item to this query item.
+ */
+ if (!checkLevel(curq, curt))
return false;
- high_pos = Min(high_pos + curq->high, tree_numlevel);
+ curt = LEVEL_NEXT(curt);
+ tlen--;
}
- curq = LQL_NEXT(curq);
- qlen--;
+ /*
+ * Once we've consumed "high" matches, we can succeed only if the rest
+ * of the pattern matches beginning here. Loop around (if you prefer,
+ * think of this as tail recursion).
+ */
+ curq = nextq;
}
- /* Fail if trailing '*'s require more ltree items than we have */
- if (low_pos > tree_numlevel || tree_numlevel > high_pos)
- return false;
-
- /* Finish pending NOT check, if any */
- if (ptr && ptr->q &&
- checkCond(ptr->q, ptr->nq,
- ptr->t, ptr->nt,
- NULL,
- 0,
- false))
- return false;
-
- return true;
+ /*
+ * Once we're out of query items, we match only if there's no remaining
+ * text either.
+ */
+ return (tlen == 0);
}
Datum
{
ltree *tree = PG_GETARG_LTREE_P(0);
lquery *query = PG_GETARG_LQUERY_P(1);
- bool res = false;
-
- if (query->flag & LQUERY_HASNOT)
- {
- FieldNot fn;
-
- fn.q = NULL;
+ bool res;
- res = checkCond(LQUERY_FIRST(query), query->numlevel,
- LTREE_FIRST(tree), tree->numlevel,
- &fn,
- 0,
- false);
- }
- else
- {
- res = checkCond(LQUERY_FIRST(query), query->numlevel,
- LTREE_FIRST(tree), tree->numlevel,
- NULL,
- 0,
- false);
- }
+ res = checkCond(LQUERY_FIRST(query), query->numlevel,
+ LTREE_FIRST(tree), tree->numlevel);
PG_FREE_IF_COPY(tree, 0);
PG_FREE_IF_COPY(query, 1);
/*
* In an lquery_level, "flag" contains the union of the variants' flags
* along with possible LQL_xxx flags; so those bit sets can't overlap.
+ *
+ * "low" and "high" are nominally the minimum and maximum number of matches.
+ * However, for backwards compatibility with pre-v13 on-disk lqueries,
+ * non-'*' levels (those with numvar > 0) only have valid low/high if the
+ * LQL_COUNT flag is set; otherwise those fields are zero, but the behavior
+ * is as if they were both 1.
*/
typedef struct
{
uint16 totallen; /* total length of this level, in bytes */
uint16 flag; /* see LQL_xxx and LVAR_xxx flags */
uint16 numvar; /* number of variants; 0 means '*' */
- uint16 low; /* minimum repeat count for '*' */
- uint16 high; /* maximum repeat count for '*' */
+ uint16 low; /* minimum repeat count */
+ uint16 high; /* maximum repeat count */
/* Array of maxalign'd lquery_variant structs follows: */
char variants[FLEXIBLE_ARRAY_MEMBER];
} lquery_level;
#define LQL_FIRST(x) ( (lquery_variant*)( ((char*)(x))+LQL_HDRSIZE ) )
#define LQL_NOT 0x10 /* level has '!' (NOT) prefix */
+#define LQL_COUNT 0x20 /* level is non-'*' and has repeat counts */
#ifdef LOWER_NODE
#define FLG_CANLOOKSIGN(x) ( ( (x) & ( LQL_NOT | LVAR_ANYEND | LVAR_SUBLEXEME ) ) == 0 )
state = LQPRS_WAITVAR;
}
+ else if (charlen == 1 && t_iseq(ptr, '{'))
+ {
+ lptr->len = ptr - lptr->start -
+ ((lptr->flag & LVAR_SUBLEXEME) ? 1 : 0) -
+ ((lptr->flag & LVAR_INCASE) ? 1 : 0) -
+ ((lptr->flag & LVAR_ANYEND) ? 1 : 0);
+ if (lptr->wlen > LTREE_LABEL_MAX_CHARS)
+ ereport(ERROR,
+ (errcode(ERRCODE_NAME_TOO_LONG),
+ errmsg("label string is too long"),
+ errdetail("Label length is %d, must be at most %d, at character %d.",
+ lptr->wlen, LTREE_LABEL_MAX_CHARS,
+ pos)));
+
+ curqlevel->flag |= LQL_COUNT;
+ state = LQPRS_WAITFNUM;
+ }
else if (charlen == 1 && t_iseq(ptr, '.'))
{
lptr->len = ptr - lptr->start -
state = LQPRS_WAITFNUM;
else if (charlen == 1 && t_iseq(ptr, '.'))
{
+ /* We only get here for '*', so these are correct defaults */
curqlevel->low = 0;
curqlevel->high = LTREE_MAX_LEVELS;
curqlevel = NEXTLEV(curqlevel);
{
totallen++;
if (curqlevel->numvar)
+ {
totallen += 1 + (curqlevel->numvar * 4) + curqlevel->totallen;
+ if (curqlevel->flag & LQL_COUNT)
+ totallen += 2 * 11 + 3;
+ }
else
totallen += 2 * 11 + 4;
curqlevel = LQL_NEXT(curqlevel);
}
}
else
+ {
+ *ptr = '*';
+ ptr++;
+ }
+
+ if ((curqlevel->flag & LQL_COUNT) || curqlevel->numvar == 0)
{
if (curqlevel->low == curqlevel->high)
{
- sprintf(ptr, "*{%d}", curqlevel->low);
+ sprintf(ptr, "{%d}", curqlevel->low);
}
else if (curqlevel->low == 0)
{
if (curqlevel->high == LTREE_MAX_LEVELS)
{
- *ptr = '*';
- *(ptr + 1) = '\0';
+ if (curqlevel->numvar == 0)
+ {
+ /* This is default for '*', so print nothing */
+ *ptr = '\0';
+ }
+ else
+ sprintf(ptr, "{,}");
}
else
- sprintf(ptr, "*{,%d}", curqlevel->high);
+ sprintf(ptr, "{,%d}", curqlevel->high);
}
else if (curqlevel->high == LTREE_MAX_LEVELS)
{
- sprintf(ptr, "*{%d,}", curqlevel->low);
+ sprintf(ptr, "{%d,}", curqlevel->low);
}
else
- sprintf(ptr, "*{%d,%d}", curqlevel->low, curqlevel->high);
+ sprintf(ptr, "{%d,%d}", curqlevel->low, curqlevel->high);
ptr = strchr(ptr, '\0');
}
SELECT '1.*.4|3|2.*{,4}'::lquery;
SELECT '1.*.4|3|2.*{1,}'::lquery;
SELECT '1.*.4|3|2.*{1}'::lquery;
+SELECT 'foo.bar{,}.!a*|b{1,}.c{,44}.d{3,4}'::lquery;
SELECT 'qwerty%@*.tu'::lquery;
SELECT nlevel('1.2.3.4');
SELECT 'a.b.c.d.e'::ltree ~ 'a.*{1}.*{2}.e';
SELECT 'a.b.c.d.e'::ltree ~ 'a.*{1}.*{4}';
SELECT 'a.b.c.d.e'::ltree ~ 'a.*{5}.*';
+SELECT '5.0.1.0'::ltree ~ '5.!0.!0.0';
+SELECT 'a.b'::ltree ~ '!a.!a';
+
+SELECT 'a.b.c.d.e'::ltree ~ 'a{,}';
+SELECT 'a.b.c.d.e'::ltree ~ 'a{1,}.*';
+SELECT 'a.b.c.d.e'::ltree ~ 'a{,}.!a{,}';
+SELECT 'a.b.c.d.a'::ltree ~ 'a{,}.!a{,}';
+SELECT 'a.b.c.d.a'::ltree ~ 'a{,2}.!a{1,}';
+SELECT 'a.b.c.d.e'::ltree ~ 'a{,2}.!a{1,}';
+SELECT 'a.b.c.d.e'::ltree ~ '!x{,}';
+SELECT 'a.b.c.d.e'::ltree ~ '!c{,}';
+SELECT 'a.b.c.d.e'::ltree ~ '!c{0,3}.!a{2,}';
+SELECT 'a.b.c.d.e'::ltree ~ '!c{0,3}.!d{2,}.*';
SELECT 'QWER_TY'::ltree ~ 'q%@*';
SELECT 'QWER_TY'::ltree ~ 'Q_t%@*';
lquery represents a regular-expression-like pattern
for matching ltree values. A simple word matches that
label within a path. A star symbol (*) matches zero
- or more labels. For example:
+ or more labels. These can be joined with dots to form a pattern that
+ must match the whole label path. For example:
foo Match the exact label path foo
*.foo.* Match any label path containing the label foo
- Star symbols can also be quantified to restrict how many labels
- they can match:
+ Both star symbols and simple words can be quantified to restrict how many
+ labels they can match:
*{n} Match exactly n labels
*{n,} Match at least n labels
*{n,m} Match at least n but not more than m labels
-*{,m} Match at most m labels — same as *{0,m}
+*{,m} Match at most m labels — same as *{0,m}
+foo{n,m} Match at least n but not more than m occurrences of foo
+foo{,} Match any number of occurrences of foo, including zero
+ In the absence of any explicit quantifier, the default for a star symbol
+ is to match any number of labels (that is, {,}) while
+ the default for a non-star item is to match exactly once (that
+ is, {1}).
There are several modifiers that can be put at the end of a non-star
- label in lquery to make it match more than just the exact match:
+ lquery item to make it match more than just the exact match:
@ Match case-insensitively, for example a@ matches A
* Match any label with this prefix, for example foo* matches foobar
- Also, you can write several possibly-modified labels separated with
- | (OR) to match any of those labels, and you can put
- ! (NOT) at the start to match any label that doesn't
- match any of the alternatives.
+ Also, you can write several possibly-modified non-star items separated with
+ | (OR) to match any of those items, and you can put
+ ! (NOT) at the start of a non-star group to match any
+ label that doesn't match any of the alternatives. A quantifier, if any,
+ goes at the end of the group; it means some number of matches for the
+ group as a whole (that is, some number of labels matching or not matching
+ any of the alternatives).
Here's an annotated example of lquery:
-Top.*{0,2}.sport*@.!football|tennis.Russ*|Spain
-a. b. c. d. e.
+Top.*{0,2}.sport*@.!football|tennis{1,}.Russ*|Spain
+a. b. c. d. e.
This query will match any label path that:
- then a label not matching football nor
- tennis
+ then has one or more labels, none of which
+ match football nor tennis
Top.Collections.Pictures.Astronomy.Astronauts
(7 rows)
-ltreetest=> SELECT path FROM test WHERE path ~ '*.!pictures@.*.Astronomy.*';
path
------------------------------------
Top.Science.Astronomy
projects / postgresql.git / commitdiff