fastxml 0.1.91

data/README

@@ -0,0 +1,9 @@
+= FastXml, a simple fast xml library using libxml and libxslt
+
+== Overview
+
+FastXml is: 
+
+# not standalone, it *requires libxml* and *libxslt*
+# it attempts to provide the speediest xml parsing library available for ruby
+# it provides an hpricot-like syntax for xml parsing and xslt processing

data/ext/Makefile

@@ -0,0 +1,153 @@
+
+SHELL = /bin/sh
+
+#### Start of system configuration section. ####
+
+srcdir = .
+topdir = /opt/local/lib/ruby/1.8/i686-darwin9.2.2
+hdrdir = $(topdir)
+VPATH = $(srcdir):$(topdir):$(hdrdir)
+prefix = $(DESTDIR)/opt/local
+exec_prefix = $(prefix)
+sitedir = $(prefix)/lib/ruby/site_ruby
+rubylibdir = $(libdir)/ruby/$(ruby_version)
+docdir = $(datarootdir)/doc/$(PACKAGE)
+dvidir = $(docdir)
+datarootdir = $(prefix)/share
+archdir = $(rubylibdir)/$(arch)
+sbindir = $(exec_prefix)/sbin
+psdir = $(docdir)
+vendordir = $(prefix)/lib/ruby/vendor_ruby
+localedir = $(datarootdir)/locale
+htmldir = $(docdir)
+datadir = $(datarootdir)
+includedir = $(prefix)/include
+infodir = $(datarootdir)/info
+sysconfdir = $(prefix)/etc
+mandir = $(DESTDIR)/opt/local/share/man
+libdir = $(exec_prefix)/lib
+sharedstatedir = $(prefix)/com
+oldincludedir = $(DESTDIR)/usr/include
+pdfdir = $(docdir)
+sitearchdir = $(sitelibdir)/$(sitearch)
+vendorarchdir = $(vendorlibdir)/$(vendorarch)
+bindir = $(exec_prefix)/bin
+localstatedir = $(prefix)/var
+vendorlibdir = $(vendordir)/$(ruby_version)
+sitelibdir = $(sitedir)/$(ruby_version)
+libexecdir = $(exec_prefix)/libexec
+
+CC = /usr/bin/gcc-4.0
+LIBRUBY = $(LIBRUBY_SO)
+LIBRUBY_A = lib$(RUBY_SO_NAME)-static.a
+LIBRUBYARG_SHARED = -l$(RUBY_SO_NAME)
+LIBRUBYARG_STATIC = -l$(RUBY_SO_NAME)-static
+
+RUBY_EXTCONF_H = 
+CFLAGS   =  -fno-common -O2  -fno-common -pipe -fno-common -Wall -I/opt/local/include/libxml2 
+INCFLAGS = -I. -I. -I/opt/local/lib/ruby/1.8/i686-darwin9.2.2 -I. -I/usr/include/libxml2
+CPPFLAGS =  -I/opt/local/include
+CXXFLAGS = $(CFLAGS) 
+DLDFLAGS = -L. -L/opt/local/lib -L/opt/local/lib -lxml2 -lz -lpthread -liconv -lm  
+LDSHARED = cc -dynamic -bundle -undefined suppress -flat_namespace
+AR = ar
+EXEEXT = 
+
+RUBY_INSTALL_NAME = ruby
+RUBY_SO_NAME = ruby
+arch = i686-darwin9.2.2
+sitearch = i686-darwin9.2.2
+vendorarch = i686-darwin9.2.2
+ruby_version = 1.8
+ruby = /opt/local/bin/ruby
+RUBY = $(ruby)
+RM = rm -f
+MAKEDIRS = mkdir -p
+INSTALL = /usr/bin/install
+INSTALL_PROG = $(INSTALL) -m 0755
+INSTALL_DATA = $(INSTALL) -m 644
+COPY = cp
+
+#### End of system configuration section. ####
+
+preload = 
+
+libpath = . $(libdir)
+LIBPATH =  -L"." -L"$(libdir)"
+DEFFILE = 
+
+CLEANFILES = mkmf.log
+DISTCLEANFILES = 
+
+extout = 
+extout_prefix = 
+target_prefix = 
+LOCAL_LIBS = 
+LIBS = $(LIBRUBYARG_SHARED) -lxslt -lxslt -lxml2 -lxml2  -lpthread -ldl -lobjc  
+SRCS = fastxml.c fastxml_attrlist.c fastxml_doc.c fastxml_node.c fastxml_nodelist.c
+OBJS = fastxml.o fastxml_attrlist.o fastxml_doc.o fastxml_node.o fastxml_nodelist.o
+TARGET = fastxml
+DLLIB = $(TARGET).bundle
+EXTSTATIC = 
+STATIC_LIB = 
+
+RUBYCOMMONDIR = $(sitedir)$(target_prefix)
+RUBYLIBDIR    = $(sitelibdir)$(target_prefix)
+RUBYARCHDIR   = $(sitearchdir)$(target_prefix)
+
+TARGET_SO     = $(DLLIB)
+CLEANLIBS     = $(TARGET).bundle $(TARGET).il? $(TARGET).tds $(TARGET).map
+CLEANOBJS     = *.o *.a *.s[ol] *.pdb *.exp *.bak
+
+all:		$(DLLIB)
+static:		$(STATIC_LIB)
+
+clean:
+		@-$(RM) $(CLEANLIBS) $(CLEANOBJS) $(CLEANFILES)
+
+distclean:	clean
+		@-$(RM) Makefile $(RUBY_EXTCONF_H) conftest.* mkmf.log
+		@-$(RM) core ruby$(EXEEXT) *~ $(DISTCLEANFILES)
+
+realclean:	distclean
+install: install-so install-rb
+
+install-so: $(RUBYARCHDIR)
+install-so: $(RUBYARCHDIR)/$(DLLIB)
+$(RUBYARCHDIR)/$(DLLIB): $(DLLIB)
+	$(INSTALL_PROG) $(DLLIB) $(RUBYARCHDIR)
+install-rb: pre-install-rb install-rb-default
+install-rb-default: pre-install-rb-default
+pre-install-rb: Makefile
+pre-install-rb-default: Makefile
+$(RUBYARCHDIR):
+	$(MAKEDIRS) $@
+
+site-install: site-install-so site-install-rb
+site-install-so: install-so
+site-install-rb: install-rb
+
+.SUFFIXES: .c .m .cc .cxx .cpp .C .o
+
+.cc.o:
+	$(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) -c $<
+
+.cxx.o:
+	$(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) -c $<
+
+.cpp.o:
+	$(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) -c $<
+
+.C.o:
+	$(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) -c $<
+
+.c.o:
+	$(CC) $(INCFLAGS) $(CPPFLAGS) $(CFLAGS) -c $<
+
+$(DLLIB): $(OBJS)
+	@-$(RM) $@
+	$(LDSHARED) -o $@ $(OBJS) $(LIBPATH) $(DLDFLAGS) $(LOCAL_LIBS) $(LIBS)
+
+
+
+$(OBJS): ruby.h defines.h

data/ext/extconf.rb

@@ -0,0 +1,13 @@
+require 'mkmf'
+ext_name = 'fastxml'
+dir_config ext_name
+find_header( 'libxml/tree.h', "/usr/include/libxml2", "/usr/local/include/libxml2", "/opt/local/include/libxml2"  )
+find_header( 'libxslt/xslt.h', "/usr/include/libxslt", "/usr/local/include/libxslt", "/opt/local/include/libxslt"  )
+find_library( 'xml2', "xmlInitParser", "/usr/lib", "/usr/local/lib", "/opt/local/lib" )
+have_library( 'xml2', 'xmlInitParser', 'libxml/parser.h' )
+find_library( 'xslt', "xmlInitParser", "/usr/lib", "/usr/local/lib", "/opt/local/lib" )
+have_library( 'xslt', 'xsltParseStylesheetFile', 'libxslt/xslt.h' )
+$LDFLAGS << ' %s' % (`xml2-config --libs`).chomp
+$CFLAGS << ' -Wall %s' % (`xml2-config --cflags`).chomp 
+
+create_makefile ext_name

data/ext/fastxml.c

@@ -0,0 +1,280 @@
+/*
+ *  $Id$
+ */
+#define fastxml_c
+#include "fastxml.h"
+#include "fastxml_node.h"
+#include "fastxml_doc.h"
+#include "fastxml_nodelist.h"
+#include "fastxml_attrlist.h"
+
+VALUE rb_cFastXmlDoc;
+VALUE rb_cFastXmlNode;
+VALUE rb_cFastXmlNodeList;
+VALUE rb_cFastXmlAttrList;
+VALUE rb_sValidateDtd;
+VALUE rb_sForgivingParse;
+VALUE rb_sHtmlParse;
+ID s_readlines;
+ID s_to_s;
+
+void Init_fastxml()
+{
+    if (xmlHasFeature(XML_WITH_TREE) == 0)
+        rb_raise( rb_eRuntimeError, "libxml not built with tree support" );
+
+    if (xmlHasFeature(XML_WITH_XPATH) == 0)
+        rb_raise( rb_eRuntimeError, "libxml not built with xpath support" );
+
+	s_readlines = rb_intern("readlines");
+	s_to_s = rb_intern("to_s");
+
+    xmlInitParser();
+    xmlXPathInit();
+    VALUE rb_mFastXml = rb_define_module( "FastXml" );
+    rb_define_const( rb_mFastXml, "LIBXML_VERSION", rb_str_new2( LIBXML_DOTTED_VERSION ) );
+    rb_cFastXmlDoc = rb_define_class_under( rb_mFastXml, "Doc", rb_cObject );        
+    rb_cFastXmlNode = rb_define_class_under( rb_mFastXml, "Node", rb_cObject );
+    rb_cFastXmlNodeList = rb_define_class_under( rb_mFastXml, "NodeList", rb_cObject );
+	rb_cFastXmlAttrList = rb_define_class_under( rb_mFastXml, "AttrList", rb_cObject );
+
+    /* setting symbols */
+    rb_sValidateDtd = ID2SYM( rb_intern("validate") );
+    rb_sForgivingParse = ID2SYM( rb_intern("forgiving") );
+    rb_sHtmlParse = ID2SYM( rb_intern("html") );
+    
+    /* Doc */
+    rb_define_method( rb_cFastXmlDoc, "initialize", fastxml_doc_initialize, -1 );
+    rb_define_method( rb_cFastXmlDoc, "search", fastxml_doc_search, 1 );
+    rb_define_method( rb_cFastXmlDoc, "to_s", fastxml_doc_to_s, 0 );
+    rb_define_method( rb_cFastXmlDoc, "root", fastxml_doc_root, 0 );
+	rb_define_method( rb_cFastXmlDoc, "transform", fastxml_doc_transform, 1 );
+	rb_define_method( rb_cFastXmlDoc, "stylesheet=", fastxml_doc_stylesheet_set, 1 );
+	rb_define_method( rb_cFastXmlDoc, "stylesheet", fastxml_doc_stylesheet, 0 );
+	rb_define_method( rb_cFastXmlDoc, "children", fastxml_doc_children, 0 );
+    rb_define_method( rb_cFastXmlDoc, "inspect", fastxml_doc_inspect, 0 );
+    
+    /* Node */
+    rb_define_method( rb_cFastXmlNode, "initialize", fastxml_node_initialize, 0 );
+    rb_define_method( rb_cFastXmlNode, "search", fastxml_node_search, 1 );
+    rb_define_method( rb_cFastXmlNode, "to_s", fastxml_node_to_s, 0 );
+    rb_define_method( rb_cFastXmlNode, "name", fastxml_node_name, 0 );
+    rb_define_method( rb_cFastXmlNode, "content", fastxml_node_value, 0 );
+    rb_define_method( rb_cFastXmlNode, "content=", fastxml_node_value_set, 1 );
+    rb_define_method( rb_cFastXmlNode, "inner_xml", fastxml_node_innerxml, 0 );
+	rb_define_method( rb_cFastXmlNode, "xpath", fastxml_node_xpath, 0 );
+	rb_define_method( rb_cFastXmlNode, "attr", fastxml_node_attr, 0 );
+	rb_define_method( rb_cFastXmlNode, "children", fastxml_node_children, 0 );
+	rb_define_method( rb_cFastXmlNode, "next", fastxml_node_next, 0 );	
+	rb_define_method( rb_cFastXmlNode, "prev", fastxml_node_prev, 0 );	
+	rb_define_method( rb_cFastXmlNode, "parent", fastxml_node_parent, 0 );	
+    rb_define_method( rb_cFastXmlNode, "inspect", fastxml_node_inspect, 0 );
+
+	/* NodeList */
+    rb_include_module( rb_cFastXmlNodeList, rb_mEnumerable );	
+    rb_define_method( rb_cFastXmlNodeList, "initialize", fastxml_nodelist_initialize, 0 );
+    rb_define_method( rb_cFastXmlNodeList, "length", fastxml_nodelist_length, 0 );
+    rb_define_method( rb_cFastXmlNodeList, "each", fastxml_nodelist_each, 0 );
+    rb_define_method( rb_cFastXmlNodeList, "entry", fastxml_nodelist_entry, 1 );
+	rb_define_method( rb_cFastXmlNodeList, "to_ary", fastxml_nodelist_entry, 0 );
+	
+	/* AttrList */
+	rb_include_module( rb_cFastXmlAttrList, rb_mEnumerable );
+	rb_define_method( rb_cFastXmlAttrList, "initialize", fastxml_attrlist_initialize, 0 );
+	rb_define_method( rb_cFastXmlAttrList, "[]", fastxml_attrlist_indexer, 1 );
+	rb_define_method( rb_cFastXmlAttrList, "[]=", fastxml_attrlist_indexer_set, 2 );
+	
+	rb_require( "lib/fastxml_lib" );
+
+}
+
+
+
+
+void fastxml_data_mark( fxml_data_t *data )
+{
+    /* do nothing */
+}
+
+void fastxml_data_free( fxml_data_t *data )
+{
+    if (data != NULL)
+    {
+		if (data->xpath_obj != NULL) 
+			xmlXPathFreeObject( data->xpath_obj );			
+	
+        if (data->doc != NULL && data->node == NULL && data->list == NULL && data->xpath_obj == NULL) 
+            xmlFreeDoc( data->doc );
+
+        // the doc free will cleanup the nodes
+
+		data->xpath_obj = NULL;
+		data->list = NULL;
+        data->doc = NULL;
+		data->node = NULL;
+        free( data );
+    }
+    data = NULL;
+}
+
+VALUE fastxml_data_alloc( VALUE klass )
+{
+    return Qnil;
+}
+
+VALUE fastxml_raw_node_to_my_obj(xmlNodePtr cur, fxml_data_t *chld)
+{
+    VALUE dv_chld, new_tmp;
+    chld->node = cur;
+    chld->doc = cur->doc; 
+
+    new_tmp = rb_class_new_instance( 0, 0, rb_cFastXmlNode ); 
+    dv_chld = Data_Wrap_Struct( rb_cObject, fastxml_data_mark, fastxml_data_free, chld );
+    rb_iv_set( new_tmp, "@lxml_doc", dv_chld );	
+	
+	return new_tmp;	
+}
+
+VALUE fastxml_raw_node_to_obj(xmlNodePtr cur)
+{
+    fxml_data_t *chld = ALLOC(fxml_data_t);
+    memset( chld, 0, sizeof(fxml_data_t) );
+	return fastxml_raw_node_to_my_obj( cur, chld );
+}
+
+VALUE fastxml_nodelist_to_obj(xmlNodePtr root, int len)
+{
+    VALUE ret, dv_chld;
+    xmlNodePtr cur = root;
+	fxml_data_t *ndlst = ALLOC(fxml_data_t);
+	memset( ndlst, 0, sizeof(fxml_data_t) );
+	
+	ndlst->list_len = len;
+	ndlst->list = cur;
+	ret = rb_class_new_instance( 0, 0, rb_cFastXmlNodeList ); 
+    dv_chld = Data_Wrap_Struct( rb_cObject, fastxml_data_mark, fastxml_data_free, ndlst );
+    rb_iv_set( ret, "@lxml_doc", dv_chld );
+   
+	return ret;
+}
+
+VALUE fastxml_nodeset_to_obj(xmlXPathObjectPtr raw_xpath_obj, fxml_data_t *data)
+{
+    VALUE ret, dv_chld;
+	fxml_data_t *ndlst = ALLOC(fxml_data_t);
+	memset( ndlst, 0, sizeof(fxml_data_t) );
+	
+	ndlst->xpath_obj = raw_xpath_obj;
+	ndlst->list_len = raw_xpath_obj->nodesetval->nodeNr;
+	ret = rb_class_new_instance( 0, 0, rb_cFastXmlNodeList ); 
+    dv_chld = Data_Wrap_Struct( rb_cObject, fastxml_data_mark, fastxml_data_free, ndlst );
+    rb_iv_set( ret, "@lxml_doc", dv_chld );
+   
+	return ret;
+}
+
+VALUE munge_xpath_namespace( VALUE orig_expr, xmlChar *root_ns )
+{
+	VALUE path_bits = rb_str_split( orig_expr, "/" );
+	VALUE ns_prefix = rb_str_new2( (const char*)root_ns );
+	VALUE ns_indic = rb_str_new2( ":" );
+	VALUE slash = rb_str_new2( "/" );
+	VALUE path_bit, str_idx;
+	VALUE ret_ary = rb_ary_new();
+	long i;
+	
+	rb_str_append( ns_prefix, ns_indic );
+    for (i=0; i<RARRAY(path_bits)->len; i++) {
+        path_bit = RARRAY(path_bits)->ptr[i];
+		
+		if (RSTRING(path_bit)->len > 0) {
+			str_idx = rb_funcall( path_bit, rb_intern( "index" ), 1, ns_indic );
+			if (str_idx == Qnil || str_idx == Qfalse) // didn't find the :, so it looks like we don't have a namespace
+				path_bit = rb_str_plus( ns_prefix, path_bit );
+		}	
+		
+		rb_ary_push( ret_ary, path_bit );
+    }
+	
+	return rb_ary_join( ret_ary, slash );
+}
+
+VALUE fastxml_xpath_search(VALUE self, VALUE raw_xpath, VALUE blk)
+{
+    VALUE ret, dv, xpath_s;
+	xmlXPathCompExprPtr xpath_xpr;
+    xmlXPathContextPtr xpath_ctx; 
+    xmlXPathObjectPtr xpath_obj;     
+    fxml_data_t *data;
+    xmlChar *xpath_expr;
+	xmlNodePtr root = NULL;
+	xmlNsPtr *ns_list = NULL;
+	xmlNsPtr *cur_ns = NULL;
+	xmlChar *root_ns = NULL;
+	int ns_cnt = 0;
+
+    if (NIL_P(raw_xpath)) 
+        rb_raise(rb_eArgError, "nil passed as xpath");
+
+    dv = rb_iv_get( self, "@lxml_doc" );    
+    Data_Get_Struct( dv, fxml_data_t, data ); 
+
+    xpath_ctx = xmlXPathNewContext( data->doc );
+    if (xpath_ctx == NULL) 
+        rb_raise( rb_eRuntimeError, "unable to create xpath context" );
+
+	root = data->node;
+	if (root == NULL)
+		root = xmlDocGetRootElement( data->doc );
+		
+	xpath_ctx->node = root;
+	cur_ns = ns_list = xmlGetNsList( data->doc, root );
+	while (cur_ns != NULL && (*cur_ns) != NULL) { 
+		xmlXPathRegisterNs( xpath_ctx, (*cur_ns)->prefix, (*cur_ns)->href );
+		cur_ns++;
+	}
+
+    if (ns_list != NULL) {
+	    xpath_ctx->namespaces = ns_list;
+	    xpath_ctx->nsNr = ns_cnt;
+    }
+	
+	xpath_s = rb_obj_as_string( raw_xpath );
+	if (root->ns != NULL) { // we have a base namespace, this is going to get "interesting"
+		root_ns = (xmlChar*)root->ns->prefix;
+		if (root_ns == NULL) 
+			root_ns = (xmlChar*)"myFunkyLittleRootNsNotToBeUseByAnyoneElseIHope"; 
+            // alternatives? how do other xpath processors handle root/default namespaces?
+
+		xmlXPathRegisterNs( xpath_ctx, root_ns, root->ns->href );
+		// need to update the xpath expression
+		xpath_s = munge_xpath_namespace( xpath_s, root_ns );
+		xpath_ctx->nsNr++;
+	}
+	
+	xpath_expr = (xmlChar*)RSTRING(xpath_s)->ptr;
+	xpath_xpr = xmlXPathCompile( xpath_expr );
+	if (xpath_xpr == NULL) {
+		xmlXPathFreeContext( xpath_ctx );
+		xmlFree( ns_list );		
+		rb_raise( rb_eRuntimeError, "unable to evaluate xpath expression" );
+	}	
+
+	xpath_obj = xmlXPathCompiledEval( xpath_xpr, xpath_ctx );
+    if (xpath_obj == NULL) {
+        rb_raise( rb_eRuntimeError, "unable to evaluate xpath expression" );
+		xmlXPathFreeCompExpr( xpath_xpr );
+        xmlXPathFreeContext( xpath_ctx ); 
+		xmlFree( ns_list );
+        return Qnil;
+    }    
+
+    ret = fastxml_nodeset_to_obj( xpath_obj, data );
+
+	xmlFree( ns_list );
+	xmlXPathFreeCompExpr( xpath_xpr );
+    xmlXPathFreeContext( xpath_ctx ); 
+
+    return ret;	
+}
+