Here is new version that adds following changes:
4. Function is now strict, per discussion.
5. Return empty array in case when XPath expression detects nothing
(previously, NULL was returned in such case), per discussion.
6. (bugfix) Work with fragments with prologue: select xpath('/a',
'<?xml version="1.0"?><a /><b />'); // now XML datum is always wrapped
with dummy <x>...</x>, XML prologue simply goes away (if any).
7. Some cleanup.
On 4/4/07, Nikolay Samokhvalov <[EMAIL PROTECTED]> wrote:
The patch attached contains following changes:
1. The function name xmlpath() is changed to xpath();
2. Approach of passing of namepspace mappings to the function is changed:
now that array (the 3rd argument) should be a 2-dimentional array with the
length of the second axis = 2 (e.g., ARRAY[ARRAY['a1', 'http://a1'],
ARRAY['a2', 'http://a2'], ARRAY['a2', 'http://a2']]);
3. Description of xpath() function in docs (I'm sorry for possible
mistakes in English and docbook formatting, please check it...)
On 4/3/07, Bruce Momjian <[EMAIL PROTECTED]> wrote:
> Nikolay Samokhvalov wrote:
> > On 3/22/07, Peter Eisentraut < [EMAIL PROTECTED]> wrote:
> > > Bruce Momjian wrote:
> > > > Patch applied.
> > >
> > > This code seems to think that if an xml datum starts with "<?xml" it's
a
> > > document. That is completely bogus.
> >
> > Agreed. I'll fix it.
>
> Nikolay, it has been a week, and I have see no fixes from you in
> response to requests from Peter. If the patch doesn't arrive in 1-2
> days, I will have to revert your patch and it will be kept for 8.4.
> Feature freeze was April 1, 2007.
>
> --
> Bruce Momjian <[EMAIL PROTECTED]> http://momjian.us
> EnterpriseDB
http://www.enterprisedb.com
>
> + If your life is a hard drive, Christ can be your backup. +
>
--
Best regards,
Nikolay
--
Best regards,
Nikolay
Index: doc/src/sgml/func.sgml
===================================================================
RCS file: /projects/cvsroot/pgsql/doc/src/sgml/func.sgml,v
retrieving revision 1.377
diff -u -r1.377 func.sgml
--- doc/src/sgml/func.sgml 7 Apr 2007 03:58:08 -0000 1.377
+++ doc/src/sgml/func.sgml 10 Apr 2007 11:51:34 -0000
@@ -8107,20 +8107,56 @@
a database system. These capabilities include import/export,
indexing, searching, transforming, and <acronym>XML</> to
<acronym>SQL</> mapping. <productname>PostgreSQL</> supports some
- but not all of these <acronym>XML</> capabilities. For an
- overview of <acronym>XML</> use in databases, see <ulink
- url="http://www.rpbourret.com/xml/XMLAndDatabases.htm";></>.
+ but not all of these <acronym>XML</> capabilities. One of such capability
+ is <function>xpath</function> function which allows to evaluate XPath 1.0
+ expressions against any XML value. For an overview of XML use in databases,
+ see <ulink url="http://www.rpbourret.com/xml/XMLAndDatabases.htm";></>.
</para>
+ <sect3>
+ <title><literal>xpath</literal></title>
+
+ <indexterm>
+ <primary>xpath</primary>
+ </indexterm>
+
+<synopsis>
+<function>xpath</function>(<replaceable>xpath</replaceable>, <replaceable>xml</replaceable><optional>, <replaceable>array</replaceable></optional>)
+</synopsis>
+
+ <para>
+ The function <function>xpath</function> evaluates XPath expression against
+ XML value and produces an array of XML values as a result. The third
+ argument of the function is an array of namespace mappings. This array
+ should be two-dimentional array with the length of the second axis being
+ equal to 2 (e.g., it should be an array of arrays, every of which constists
+ of exactly 2 elements, name and URI).
+ </para>
+
+ <para>
+ Example:
+<screen><![CDATA[
+SELECT xpath('/my:a/text()', '<my:a xmlns:my="http://example.com";>test</my:a>', ARRAY[ARRAY['my', 'http://example.com']]);
+ xpath
+--------
+ {test}
+(1 row)
+]]></screen>
+ </para>
+ </sect3>
+
+ <sect3>
+ <title>Other capabilities</title>
+
<variablelist>
<varlistentry>
<term>Indexing</term>
<listitem>
<para>
- <filename>contrib/xml2/</> functions can be used in expression
- indexes to index specific <acronym>XML</> fields. To index the
- full contents of <acronym>XML</> documents, the full-text
+ To improve performance of XPath evaluation, one can use functional indexes
+ over <function>xpath</function> with constant XPath expression.
+ To index the full contents of <acronym>XML</> documents, the full-text
indexing tool <filename>contrib/tsearch2/</> can be used. Of
course, Tsearch2 indexes have no <acronym>XML</> awareness so
additional <filename>contrib/xml2/</> checks should be added to
@@ -8134,9 +8170,10 @@
<listitem>
<para>
- XPath searches are implemented using <filename>contrib/xml2/</>.
- It processes <acronym>XML</> text documents and returns results
- based on the requested query.
+ The basic XPath function is <function>xpath</function> which is built in.
+ Additional XPath functions are implemented using
+ <filename>contrib/xml2/</>.
+
</para>
</listitem>
</varlistentry>
@@ -8165,6 +8202,8 @@
</listitem>
</varlistentry>
</variablelist>
+
+ </sect3>
</sect2>
</sect1>
Index: src/test/regress/sql/xml.sql
===================================================================
RCS file: /projects/cvsroot/pgsql/src/test/regress/sql/xml.sql,v
retrieving revision 1.14
diff -u -r1.14 xml.sql
--- src/test/regress/sql/xml.sql 22 Mar 2007 20:26:30 -0000 1.14
+++ src/test/regress/sql/xml.sql 10 Apr 2007 11:51:34 -0000
@@ -147,9 +147,9 @@
-- Text XPath expressions evaluation
-SELECT xmlpath('/value', data) FROM xmltest;
-SELECT xmlpath(NULL, NULL) IS NULL FROM xmltest;
-SELECT xmlpath('', '<!-- error -->');
-SELECT xmlpath('//text()', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>');
-SELECT xmlpath('//loc:piece/@id', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>', ARRAY[ARRAY['loc'], ARRAY['http://127.0.0.1']]);
-SELECT xmlpath('//b', '<a>one <b>two</b> three <b>etc</b></a>');
+SELECT xpath('/value', data) FROM xmltest;
+SELECT xpath(NULL, NULL) IS NULL FROM xmltest;
+SELECT xpath('', '<!-- error -->');
+SELECT xpath('//text()', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>');
+SELECT xpath('//loc:piece/@id', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>', ARRAY[ARRAY['loc', 'http://127.0.0.1']]);
+SELECT xpath('//b', '<a>one <b>two</b> three <b>etc</b></a>');
Index: src/include/catalog/pg_proc.h
===================================================================
RCS file: /projects/cvsroot/pgsql/src/include/catalog/pg_proc.h,v
retrieving revision 1.454
diff -u -r1.454 pg_proc.h
--- src/include/catalog/pg_proc.h 2 Apr 2007 03:49:40 -0000 1.454
+++ src/include/catalog/pg_proc.h 10 Apr 2007 11:51:34 -0000
@@ -4114,9 +4114,9 @@
DATA(insert OID = 2938 ( database_to_xml_and_xmlschema PGNSP PGUID 12 100 0 f f t f s 3 142 "16 16 25" _null_ _null_ "{nulls,tableforest,targetns}" database_to_xml_and_xmlschema - _null_ ));
DESCR("map database contents and structure to XML and XML Schema");
-DATA(insert OID = 2931 ( xmlpath PGNSP PGUID 12 1 0 f f f f i 3 143 "25 142 1009" _null_ _null_ _null_ xmlpath - _null_ ));
+DATA(insert OID = 2931 ( xpath PGNSP PGUID 12 1 0 f f t f i 3 143 "25 142 1009" _null_ _null_ _null_ xpath - _null_ ));
DESCR("evaluate XPath expression, with namespaces support");
-DATA(insert OID = 2932 ( xmlpath PGNSP PGUID 14 1 0 f f f f i 2 143 "25 142" _null_ _null_ _null_ "select pg_catalog.xmlpath($1, $2, NULL)" - _null_ ));
+DATA(insert OID = 2932 ( xpath PGNSP PGUID 14 1 0 f f t f i 2 143 "25 142" _null_ _null_ _null_ "select pg_catalog.xpath($1, $2, ''{}''::_text)" - _null_ ));
DESCR("evaluate XPath expression");
/* uuid */
Index: src/include/utils/xml.h
===================================================================
RCS file: /projects/cvsroot/pgsql/src/include/utils/xml.h,v
retrieving revision 1.18
diff -u -r1.18 xml.h
--- src/include/utils/xml.h 1 Apr 2007 09:00:26 -0000 1.18
+++ src/include/utils/xml.h 10 Apr 2007 11:51:34 -0000
@@ -36,7 +36,7 @@
extern Datum texttoxml(PG_FUNCTION_ARGS);
extern Datum xmltotext(PG_FUNCTION_ARGS);
extern Datum xmlvalidate(PG_FUNCTION_ARGS);
-extern Datum xmlpath(PG_FUNCTION_ARGS);
+extern Datum xpath(PG_FUNCTION_ARGS);
extern Datum table_to_xml(PG_FUNCTION_ARGS);
extern Datum query_to_xml(PG_FUNCTION_ARGS);
Index: src/backend/utils/adt/xml.c
===================================================================
RCS file: /projects/cvsroot/pgsql/src/backend/utils/adt/xml.c,v
retrieving revision 1.42
diff -u -r1.42 xml.c
--- src/backend/utils/adt/xml.c 6 Apr 2007 04:21:43 -0000 1.42
+++ src/backend/utils/adt/xml.c 10 Apr 2007 11:51:34 -0000
@@ -2996,12 +2996,12 @@
* As we have no support of XQuery sequences yet, this functions seems
* to be the most useful one (array of XML functions plays a role of
* some kind of substritution for XQuery sequences).
-
+ *
* Workaround here: we parse XML data in different way to allow XPath for
* fragments (see "XPath for fragment" TODO comment inside).
*/
Datum
-xmlpath(PG_FUNCTION_ARGS)
+xpath(PG_FUNCTION_ARGS)
{
#ifdef USE_LIBXML
ArrayBuildState *astate = NULL;
@@ -3013,11 +3013,11 @@
int32 len, xpath_len;
xmlChar *string, *xpath_expr;
bool res_is_null = FALSE;
- int i;
+ int i, res_nitems;
xmltype *data;
text *xpath_expr_text;
ArrayType *namespaces;
- int *dims, ndims, ns_count = 0, bitmask = 1;
+ int *dims, ndim, ns_count = 0, bitmask = 1;
char *ptr;
bits8 *bitmap;
char **ns_names = NULL, **ns_uris = NULL;
@@ -3025,32 +3025,30 @@
bool typbyval;
char typalign;
- /* the function is not strict, we must check first two args */
- if (PG_ARGISNULL(0) || PG_ARGISNULL(1))
- PG_RETURN_NULL();
-
xpath_expr_text = PG_GETARG_TEXT_P(0);
data = PG_GETARG_XML_P(1);
+ namespaces = PG_GETARG_ARRAYTYPE_P(2);
- /* Namespace mappings passed as text[].
- * Assume that 2-dimensional array has been passed,
- * the 1st subarray is array of names, the 2nd -- array of URIs,
- * example: ARRAY[ARRAY['myns', 'myns2'], ARRAY['http://example.com', 'http://example2.com']].
+ /*
+ * Namespace mappings are passed as text[].
+ * If empty array is passed (ndim = 0, "0-dimentional"), then there is no namespace mappings.
+ * Else, 2-dimentional array with length of the second axis being equal to 2 should be passed,
+ * i.e. every subarray contain 2 elements, the first element defining name, the second one -- URI.
+ * Example: ARRAY[ARRAY['myns', 'http://example.com'], ARRAY['myns2', 'http://example2.com']].
*/
- if (!PG_ARGISNULL(2))
+ ndim = ARR_NDIM(namespaces);
+ if (ndim != 0)
{
- namespaces = PG_GETARG_ARRAYTYPE_P(2);
- ndims = ARR_NDIM(namespaces);
dims = ARR_DIMS(namespaces);
/* Sanity check */
- if (ndims != 2)
+ if (ndim != 2 || dims[1] != 2)
ereport(ERROR, (errmsg("invalid array passed for namespace mappings"),
- errdetail("Only 2-dimensional array may be used for namespace mappings.")));
+ errdetail("Only 2-dimentional array with length of the second axis being equal to 2 may be used for namespace mappings.")));
Assert(ARR_ELEMTYPE(namespaces) == TEXTOID);
- ns_count = ArrayGetNItems(ndims, dims) / 2;
+ ns_count = ArrayGetNItems(ndim, dims) / 2; /* number of NS mappings */
get_typlenbyvalalign(ARR_ELEMTYPE(namespaces),
&typlen, &typbyval, &typalign);
ns_names = (char **) palloc(ns_count * sizeof(char *));
@@ -3058,18 +3056,17 @@
ptr = ARR_DATA_PTR(namespaces);
bitmap = ARR_NULLBITMAP(namespaces);
bitmask = 1;
-
for (i = 0; i < ns_count * 2; i++)
{
if (bitmap && (*bitmap & bitmask) == 0)
- ereport(ERROR, (errmsg("neither namespace nor URI may be NULL"))); /* TODO: better message */
+ ereport(ERROR, (errmsg("neither namespace nor URI may be NULL")));
else
{
- if (i < ns_count)
- ns_names[i] = DatumGetCString(DirectFunctionCall1(textout,
+ if (i % 2 == 0)
+ ns_names[i / 2] = DatumGetCString(DirectFunctionCall1(textout,
PointerGetDatum(ptr)));
else
- ns_uris[i - ns_count] = DatumGetCString(DirectFunctionCall1(textout,
+ ns_uris[i / 2] = DatumGetCString(DirectFunctionCall1(textout,
PointerGetDatum(ptr)));
ptr = att_addlength_pointer(ptr, typlen, ptr);
ptr = (char *) att_align_nominal(ptr, typalign);
@@ -3093,31 +3090,38 @@
if (xpath_len == 0)
ereport(ERROR, (errmsg("empty XPath expression")));
- if (xmlStrncmp((xmlChar *) VARDATA(data), (xmlChar *) "<?xml", 5) == 0)
+ /*
+ * To handle both document and fragment, regardless of the fact that XML datum has
+ * single root (XML well-formedness), we wrap XML datum with dummy element
+ * (<x>...</x>) and extend XPath expression accordingly. To do it, throw away
+ * XML prolog, if any.
+ */
+ if ((len > 4) && xmlStrncmp((xmlChar *) VARDATA(data), (xmlChar *) "<?xml", 5) == 0)
{
- string = palloc(len + 1);
- memcpy(string, VARDATA(data), len);
- string[len] = '\0';
- xpath_expr = palloc(xpath_len + 1);
- memcpy(xpath_expr, VARDATA(xpath_expr_text), xpath_len);
- xpath_expr[xpath_len] = '\0';
+ i = 5;
+ while ((i < len) && (('?' != (VARDATA(data))[i - 1]) || ('>' != (VARDATA(data))[i])))
+ i++;
+
+ if (i == len)
+ xml_ereport(ERROR, ERRCODE_INTERNAL_ERROR,
+ "could not parse XML data");
+
+ ++i;
+ string = xmlStrncatNew((xmlChar *) "<x>", (xmlChar *) VARDATA(data) + i, len - i);
}
else
- {
- /* use "<x>...</x>" as dummy root element to enable XPath for fragments */
- /* TODO: (XPath for fragment) find better solution to work with XML fragment! */
string = xmlStrncatNew((xmlChar *) "<x>", (xmlChar *) VARDATA(data), len);
- string = xmlStrncat(string, (xmlChar *) "</x>", 5);
- len += 7;
- xpath_expr = xmlStrncatNew((xmlChar *) "/x", (xmlChar *) VARDATA(xpath_expr_text), xpath_len);
- len += 2;
- }
+
+ string = xmlStrncat(string, (xmlChar *) "</x>", 5);
+ len += 7;
+ xpath_expr = xmlStrncatNew((xmlChar *) "/x", (xmlChar *) VARDATA(xpath_expr_text), xpath_len);
+ xpath_len += 2;
xml_init();
PG_TRY();
{
- /* redundant XML parsing (two parsings for the same value in the same session are possible) */
+ /* redundant XML parsing (two parsings for the same value during one command execution are possible) */
ctxt = xmlNewParserCtxt();
if (ctxt == NULL)
xml_ereport(ERROR, ERRCODE_INTERNAL_ERROR,
@@ -3152,15 +3156,15 @@
if (xpathobj == NULL)
ereport(ERROR, (errmsg("could not create XPath object")));
+ /*
+ * Return empty array in cases when nothing is found
+ */
if (xpathobj->nodesetval == NULL)
- res_is_null = TRUE;
-
- if (!res_is_null && xpathobj->nodesetval->nodeNr == 0)
- /* TODO maybe empty array should be here, not NULL? (if so -- fix segfault) */
- /*PG_RETURN_ARRAYTYPE_P(makeArrayResult(astate, CurrentMemoryContext));*/
- res_is_null = TRUE;
+ res_nitems = 0;
+ else
+ res_nitems = xpathobj->nodesetval->nodeNr;
- if (!res_is_null)
+ if (!res_is_null && res_nitems)
for (i = 0; i < xpathobj->nodesetval->nodeNr; i++)
{
Datum elem;
@@ -3196,13 +3200,11 @@
PG_END_TRY();
if (res_is_null)
- {
PG_RETURN_NULL();
- }
+ else if (res_nitems == 0)
+ PG_RETURN_ARRAYTYPE_P(construct_empty_array(XMLOID));
else
- {
PG_RETURN_ARRAYTYPE_P(makeArrayResult(astate, CurrentMemoryContext));
- }
#else
NO_XML_SUPPORT();
return 0;
Index: src/test/regress/expected/xml_1.out
===================================================================
RCS file: /projects/cvsroot/pgsql/src/test/regress/expected/xml_1.out,v
retrieving revision 1.16
diff -u -r1.16 xml_1.out
--- src/test/regress/expected/xml_1.out 5 Apr 2007 13:53:24 -0000 1.16
+++ src/test/regress/expected/xml_1.out 10 Apr 2007 11:51:34 -0000
@@ -326,29 +326,29 @@
(2 rows)
-- Text XPath expressions evaluation
-SELECT xmlpath('/value', data) FROM xmltest;
- xmlpath
----------
+SELECT xpath('/value', data) FROM xmltest;
+ xpath
+-------
(0 rows)
-SELECT xmlpath(NULL, NULL) IS NULL FROM xmltest;
-ERROR: unsupported XML feature
-DETAIL: This functionality requires libxml support.
-HINT: You need to re-compile PostgreSQL using --with-libxml.
-CONTEXT: SQL function "xmlpath" statement 1
-SELECT xmlpath('', '<!-- error -->');
+SELECT xpath(NULL, NULL) IS NULL FROM xmltest;
+ ?column?
+----------
+(0 rows)
+
+SELECT xpath('', '<!-- error -->');
ERROR: unsupported XML feature
DETAIL: This functionality requires libxml support.
HINT: You need to re-compile PostgreSQL using --with-libxml.
-SELECT xmlpath('//text()', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>');
+SELECT xpath('//text()', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>');
ERROR: unsupported XML feature
DETAIL: This functionality requires libxml support.
HINT: You need to re-compile PostgreSQL using --with-libxml.
-SELECT xmlpath('//loc:piece/@id', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>', ARRAY[ARRAY['loc'], ARRAY['http://127.0.0.1']]);
+SELECT xpath('//loc:piece/@id', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>', ARRAY[ARRAY['loc', 'http://127.0.0.1']]);
ERROR: unsupported XML feature
DETAIL: This functionality requires libxml support.
HINT: You need to re-compile PostgreSQL using --with-libxml.
-SELECT xmlpath('//b', '<a>one <b>two</b> three <b>etc</b></a>');
+SELECT xpath('//b', '<a>one <b>two</b> three <b>etc</b></a>');
ERROR: unsupported XML feature
DETAIL: This functionality requires libxml support.
HINT: You need to re-compile PostgreSQL using --with-libxml.
Index: src/test/regress/expected/xml.out
===================================================================
RCS file: /projects/cvsroot/pgsql/src/test/regress/expected/xml.out,v
retrieving revision 1.17
diff -u -r1.17 xml.out
--- src/test/regress/expected/xml.out 22 Mar 2007 20:26:30 -0000 1.17
+++ src/test/regress/expected/xml.out 10 Apr 2007 11:51:34 -0000
@@ -402,37 +402,37 @@
(9 rows)
-- Text XPath expressions evaluation
-SELECT xmlpath('/value', data) FROM xmltest;
- xmlpath
+SELECT xpath('/value', data) FROM xmltest;
+ xpath
----------------------
{<value>one</value>}
{<value>two</value>}
(2 rows)
-SELECT xmlpath(NULL, NULL) IS NULL FROM xmltest;
+SELECT xpath(NULL, NULL) IS NULL FROM xmltest;
?column?
----------
t
t
(2 rows)
-SELECT xmlpath('', '<!-- error -->');
+SELECT xpath('', '<!-- error -->');
ERROR: empty XPath expression
-CONTEXT: SQL function "xmlpath" statement 1
-SELECT xmlpath('//text()', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>');
- xmlpath
+CONTEXT: SQL function "xpath" statement 1
+SELECT xpath('//text()', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>');
+ xpath
----------------
{"number one"}
(1 row)
-SELECT xmlpath('//loc:piece/@id', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>', ARRAY[ARRAY['loc'], ARRAY['http://127.0.0.1']]);
- xmlpath
----------
+SELECT xpath('//loc:piece/@id', '<local:data xmlns:local="http://127.0.0.1";><local:piece id="1">number one</local:piece><local:piece id="2" /></local:data>', ARRAY[ARRAY['loc', 'http://127.0.0.1']]);
+ xpath
+-------
{1,2}
(1 row)
-SELECT xmlpath('//b', '<a>one <b>two</b> three <b>etc</b></a>');
- xmlpath
+SELECT xpath('//b', '<a>one <b>two</b> three <b>etc</b></a>');
+ xpath
-------------------------
{<b>two</b>,<b>etc</b>}
(1 row)
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
