The XML::XPath::Context class is used to evaluate XPath expressions. Generally, you should not directly use this class, but instead use the XML::Document#find and XML::Node#find methods.
doc = XML::Document.string('<header>content</header>')
context = XPath::Context.new(doc)
context.node = doc.root
context.register_namespaces_from_node(doc.root)
nodes = context.find('/header')
Methods
public class
public instance
Attributes
| doc | [R] |
Public class methods
Creates a new XPath context for the specified document. The context can then be used to evaluate an XPath expression.
doc = XML::Document.string('<header><first>hi</first></header>')
context = XPath::Context.new(doc)
nodes = XPath::Object.new('//first', context)
nodes.length == 1
static VALUE rxml_xpath_context_initialize(VALUE self, VALUE node)
{
xmlDocPtr xdoc;
VALUE document;
if (rb_obj_is_kind_of(node, cXMLNode) == Qtrue)
{
document = rb_funcall(node, rb_intern("doc"), 0);
if (NIL_P(document))
rb_raise(rb_eTypeError, "Supplied node must belong to a document.");
}
else if (rb_obj_is_kind_of(node, cXMLDocument) == Qtrue)
{
document = node;
}
else
{
rb_raise(rb_eTypeError, "Supplied argument must be a document or node.");
}
Data_Get_Struct(document, xmlDoc, xdoc);
DATA_PTR(self) = xmlXPathNewContext(xdoc);
/* Save the doc as an attribute, this will expose it to Ruby's GC. */
rb_ivar_set(self, DOC_ATTRIBUTE, document);
return self;
}
Public instance methods
Disables an XPath::Context’s built-in cache.
static VALUE
rxml_xpath_context_disable_cache(VALUE self)
{
xmlXPathContextPtr xctxt;
Data_Get_Struct(self, xmlXPathContext, xctxt);
if (xmlXPathContextSetCache(xctxt, 0, 0, 0) == -1)
rxml_raise(&xmlLastError);
return self;
}
Enables an XPath::Context’s built-in cache. If the cache is enabled then XPath objects will be cached internally for reuse. The size parameter controls sets the maximum number of XPath objects that will be cached per XPath object type (node-set, string, number, boolean, and misc objects). Set size to nil to use the default cache size of 100.
static VALUE
rxml_xpath_context_enable_cache(int argc, VALUE *argv, VALUE self)
{
xmlXPathContextPtr xctxt;
VALUE size;
int value = -1;
Data_Get_Struct(self, xmlXPathContext, xctxt);
if (rb_scan_args(argc, argv, "01", &size) == 1)
{
value = NUM2INT(size);
}
if (xmlXPathContextSetCache(xctxt, 1, value, 0) == -1)
rxml_raise(&xmlLastError);
return self;
}
Executes the provided xpath function. The result depends on the execution of the xpath statement. It may be true, false, a number, a string or a node set.
static VALUE rxml_xpath_context_find(VALUE self, VALUE xpath_expr)
{
xmlXPathContextPtr xctxt;
xmlXPathObjectPtr xobject;
xmlXPathCompExprPtr xcompexpr;
VALUE result;
Data_Get_Struct(self, xmlXPathContext, xctxt);
if (TYPE(xpath_expr) == T_STRING)
{
VALUE expression = rb_check_string_type(xpath_expr);
xobject = xmlXPathEval((xmlChar*) StringValueCStr(expression), xctxt);
}
else if (rb_obj_is_kind_of(xpath_expr, cXMLXPathExpression))
{
Data_Get_Struct(xpath_expr, xmlXPathCompExpr, xcompexpr);
xobject = xmlXPathCompiledEval(xcompexpr, xctxt);
}
else
{
rb_raise(rb_eTypeError,
"Argument should be an intance of a String or XPath::Expression");
}
if (xobject == NULL)
{
/* xmlLastError is different than xctxt->lastError. Use
xmlLastError since it has the message set while xctxt->lastError
does not. */
xmlErrorPtr xerror = xmlGetLastError();
rxml_raise(xerror);
}
switch (xobject->type)
{
case XPATH_NODESET:
result = rxml_xpath_object_wrap(xctxt->doc, xobject);
break;
case XPATH_BOOLEAN:
result = (xobject->boolval != 0) ? Qtrue : Qfalse;
xmlXPathFreeObject(xobject);
break;
case XPATH_NUMBER:
result = rb_float_new(xobject->floatval);
xmlXPathFreeObject(xobject);
break;
case XPATH_STRING:
result = rxml_str_new2((const char*)xobject->stringval, xctxt->doc->encoding);
xmlXPathFreeObject(xobject);
break;
default:
result = Qnil;
xmlXPathFreeObject(xobject);
}
return result;
}
Set the current node used by the XPath engine
doc = XML::Document.string('<header><first>hi</first></header>')
context.node = doc.root.first
static VALUE rxml_xpath_context_node_set(VALUE self, VALUE node)
{
xmlXPathContextPtr xctxt;
xmlNodePtr xnode;
Data_Get_Struct(self, xmlXPathContext, xctxt);
Data_Get_Struct(node, xmlNode, xnode);
xctxt->node = xnode;
return node;
}
Register the specified namespace URI with the specified prefix in this context.
context.register_namespace('xi', 'http://www.w3.org/2001/XInclude')
static VALUE rxml_xpath_context_register_namespace(VALUE self, VALUE prefix, VALUE uri)
{
xmlXPathContextPtr ctxt;
Data_Get_Struct(self, xmlXPathContext, ctxt);
/* Prefix could be a symbol. */
prefix = rb_obj_as_string(prefix);
if (xmlXPathRegisterNs(ctxt, (xmlChar*) StringValuePtr(prefix),
(xmlChar*) StringValuePtr(uri)) == 0)
{
return (Qtrue);
}
else
{
/* Should raise an exception, IMHO (whose?, why shouldnt it? -danj)*/
rb_warning("register namespace failed");
return (Qfalse);
}
}
Register the specified namespaces in this context. There are three different forms that libxml accepts. These include a string, an array of strings, or a hash table:
context.register_namespaces('xi:http://www.w3.org/2001/XInclude')
context.register_namespaces(['xlink:http://www.w3.org/1999/xlink',
'xi:http://www.w3.org/2001/XInclude')
context.register_namespaces('xlink' => 'http://www.w3.org/1999/xlink',
'xi' => 'http://www.w3.org/2001/XInclude')
static VALUE rxml_xpath_context_register_namespaces(VALUE self, VALUE nslist)
{
char *cp;
long i;
VALUE rprefix, ruri;
xmlXPathContextPtr xctxt;
Data_Get_Struct(self, xmlXPathContext, xctxt);
/* Need to loop through the 2nd argument and iterate through the
* list of namespaces that we want to allow */
switch (TYPE(nslist))
{
case T_STRING:
cp = strchr(StringValuePtr(nslist), (int) ':');
if (cp == NULL)
{
rprefix = nslist;
ruri = Qnil;
}
else
{
rprefix = rb_str_new(StringValuePtr(nslist), (int) ((long) cp
- (long) StringValuePtr(nslist)));
ruri = rxml_str_new2(&cp[1], xctxt->doc->encoding);
}
/* Should test the results of this */
rxml_xpath_context_register_namespace(self, rprefix, ruri);
break;
case T_ARRAY:
for (i = 0; i < RARRAY_LEN(nslist); i++)
{
rxml_xpath_context_register_namespaces(self, RARRAY_PTR(nslist)[i]);
}
break;
case T_HASH:
rb_hash_foreach(nslist, iterate_ns_hash, self);
break;
default:
rb_raise(
rb_eArgError,
"Invalid argument type, only accept string, array of strings, or an array of arrays");
}
return self;
}
Helper method to read in namespaces defined on a node.
doc = XML::Document.string('<header><first>hi</first></header>')
context = XPath::Context.new(doc)
context.register_namespaces_from_node(doc.root)
static VALUE rxml_xpath_context_register_namespaces_from_node(VALUE self,
VALUE node)
{
xmlXPathContextPtr xctxt;
xmlNodePtr xnode;
xmlNsPtr *xnsArr;
Data_Get_Struct(self, xmlXPathContext, xctxt);
if (rb_obj_is_kind_of(node, cXMLDocument) == Qtrue)
{
xmlDocPtr xdoc;
Data_Get_Struct(node, xmlDoc, xdoc);
xnode = xmlDocGetRootElement(xdoc);
}
else if (rb_obj_is_kind_of(node, cXMLNode) == Qtrue)
{
Data_Get_Struct(node, xmlNode, xnode);
}
else
{
rb_raise(rb_eTypeError, "The first argument must be a document or node.");
}
xnsArr = xmlGetNsList(xnode->doc, xnode);
if (xnsArr)
{
xmlNsPtr xns = *xnsArr;
while (xns)
{
/* If there is no prefix, then this is the default namespace.
Skip it for now. */
if (xns->prefix)
{
VALUE prefix = rxml_str_new2((const char*)xns->prefix, xctxt->doc->encoding);
VALUE uri = rxml_str_new2((const char*)xns->href, xctxt->doc->encoding);
rxml_xpath_context_register_namespace(self, prefix, uri);
}
xns = xns->next;
}
xmlFree(xnsArr);
}
return self;
}