shithub: xml-9atom

ref: 47a288683dbdc55c5df9b9f65db9909de6021af0
dir: /xpath/

View raw version
.TH XPATH 2
.SH NAME
xmllookpath
\- XPath support
.SH SYNOPSIS
.de PB
.PP
.ft L
.nf
..
.PB
#include <u.h>
#include <libc.h>
#include <xml.h>
#include <xpath.h>
.PB
enum {
	Xelem = 1,
	Xstring = 2,
	Xnum = 3,
}
.PB
struct XpResult {
	int type; 	/* type of XpResult */
	int error;	/* 1 if error. Check errstr */
	int num;  	/* number of results */
	union {   	/* array of results */
		char **strings;	/* if type == Xstring */
		Elem **elems;  	/* if type == Xelems */
		int   *numbers;	/* if type == Xnum */
	};
	...
};
.PB
.PD 0
.ta +\w'\fLXpResult 'u
XpResult	xmllookpath(Elem *ep, char *xpath)
.SH DESCRIPTION
.PP
.I Libxpath
is an extension library to
.IR libxml .
It provides XPath functionality for looking up certain nodes in an
existing in-memory XML DOM model.
.PP
.I Xmllookpath
is the main function for querying the XML document using an XPath string.
It's using
.I ep
as the reference element within the DOM model.
.PP
The resulting
.I XpResult
holds the typed results as an array depending on the type.
The allocated array will hold
.I num
items.
If
.I error
is set,
.I errstr
contains the description of the error.
.SH SOURCE
/sys/src/libxpath
.SH "SEE ALSO"
.IR xml (2),
.IR errstr (2).
.SH BUGS
The current implementation of XPath is incomplete and very limited.
It should be possible to extend the parser to cover a full set of functionality.