lib/nokogiri/xml/sax/legacy_handlers.rb @@ -1,5 +1,13 @@ #include <nokogiri.h> +#define STRING_OR_NULL(str) \ + ({ \ + RTEST(str) ? StringValuePtr(str) : NULL \ + }) + +#define RBSTR_OR_QNIL(_str, rb_enc) \ + (_str ? NOKOGIRI_STR_NEW2(_str, STRING_OR_NULL(rb_enc)) : Qnil) + /* * call-seq: * parse_memory(data) @@ -89,7 +97,7 @@ static void start_element(void * ctx, const xmlChar *name, const xmlChar **atts) if(atts) { while((attr = atts[i]) != NULL) { rb_funcall(attributes, rb_intern("<<"), 1, - NOKOGIRI_STR_NEW2(attr, RTEST(enc) ? StringValuePtr(enc) : NULL) + NOKOGIRI_STR_NEW2(attr, STRING_OR_NULL(enc)) ); i++; } @@ -98,7 +106,7 @@ static void start_element(void * ctx, const xmlChar *name, const xmlChar **atts) rb_funcall( doc, rb_intern("start_element"), 2, - NOKOGIRI_STR_NEW2(name, RTEST(enc) ? StringValuePtr(enc) : NULL), + NOKOGIRI_STR_NEW2(name, STRING_OR_NULL(enc)), attributes ); } @@ -109,19 +117,51 @@ static void end_element(void * ctx, const xmlChar *name) VALUE MAYBE_UNUSED(enc) = rb_iv_get(self, "@encoding"); VALUE doc = rb_funcall(self, rb_intern("document"), 0); rb_funcall(doc, rb_intern("end_element"), 1, - NOKOGIRI_STR_NEW2(name, RTEST(enc) ? StringValuePtr(enc) : NULL) + NOKOGIRI_STR_NEW2(name, STRING_OR_NULL(enc)) ); } -/** - * start_element_ns was borrowed heavily from libxml-ruby. - */ +static VALUE attributes_as_list( + VALUE self, + int nb_attributes, + const xmlChar ** attributes) +{ + VALUE list = rb_ary_new2(nb_attributes); + VALUE MAYBE_UNUSED(enc) = rb_iv_get(self, "@encoding"); + + VALUE attr_klass = rb_const_get(cNokogiriXmlSaxParser, rb_intern("Attribute")); + if (attributes) { + /* Each attribute is an array of [localname, prefix, URI, value, end] */ + int i; + for (i = 0; i < nb_attributes * 5; i += 5) { + VALUE attribute = rb_funcall(attr_klass, rb_intern("new"), 4, + /* localname */ + RBSTR_OR_QNIL(attributes[i + 0], enc), + + /* prefix */ + RBSTR_OR_QNIL(attributes[i + 1], enc), + + /* URI */ + RBSTR_OR_QNIL(attributes[i + 2], enc), + + /* value */ + NOKOGIRI_STR_NEW((const char*)attributes[i+3], + (attributes[i+4] - attributes[i+3]), + STRING_OR_NULL(enc)) + ); + rb_ary_push(list, attribute); + } + } + + return list; +} + static void start_element_ns ( void * ctx, const xmlChar * localname, const xmlChar * prefix, - const xmlChar * URI, + const xmlChar * uri, int nb_namespaces, const xmlChar ** namespaces, int nb_attributes, @@ -132,52 +172,42 @@ start_element_ns ( VALUE doc = rb_funcall(self, rb_intern("document"), 0); VALUE MAYBE_UNUSED(enc) = rb_iv_get(self, "@encoding"); - VALUE attrHash = rb_hash_new(); - VALUE nsHash = rb_hash_new(); + VALUE attribute_list = attributes_as_list(self, nb_attributes, attributes); - if (attributes) - { - /* Each attribute is an array of [localname, prefix, URI, value, end] */ - int i; - for (i = 0; i < nb_attributes * 5; i += 5) - { - rb_hash_aset( attrHash, - NOKOGIRI_STR_NEW2((const char*)attributes[i+0], RTEST(enc) ? StringValuePtr(enc) : NULL), - NOKOGIRI_STR_NEW((const char*)attributes[i+3], (attributes[i+4] - attributes[i+3]), RTEST(enc) ? StringValuePtr(enc) : NULL)); - } - } + VALUE ns_list = rb_ary_new2(nb_namespaces); - if (namespaces) - { + if (namespaces) { int i; for (i = 0; i < nb_namespaces * 2; i += 2) { - rb_hash_aset( nsHash, - namespaces[i+0] ? NOKOGIRI_STR_NEW2((const char*)namespaces[i+0], RTEST(enc) ? StringValuePtr(enc) : NULL) : Qnil, - namespaces[i+1] ? NOKOGIRI_STR_NEW2((const char*)namespaces[i+1], RTEST(enc) ? StringValuePtr(enc) : NULL) : Qnil); + rb_ary_push(ns_list, + rb_ary_new3(2, + RBSTR_OR_QNIL(namespaces[i + 0], enc), + RBSTR_OR_QNIL(namespaces[i + 1], enc) + ) + ); } } rb_funcall( doc, - rb_intern("start_element_ns"), + rb_intern("start_element_namespace"), 5, - NOKOGIRI_STR_NEW2(localname, RTEST(enc) ? StringValuePtr(enc) : NULL), - attrHash, - prefix ? NOKOGIRI_STR_NEW2(prefix, RTEST(enc) ? StringValuePtr(enc) : NULL) : Qnil, - URI ? NOKOGIRI_STR_NEW2(URI, RTEST(enc) ? StringValuePtr(enc) : NULL) : Qnil, - nsHash + NOKOGIRI_STR_NEW2(localname, STRING_OR_NULL(enc)), + attribute_list, + RBSTR_OR_QNIL(prefix, enc), + RBSTR_OR_QNIL(uri, enc), + ns_list ); - /* Call start element if it's there' */ - if (rb_respond_to(doc, rb_intern("start_element"))) - { - VALUE name = - NOKOGIRI_STR_NEW2(localname, RTEST(enc) ? StringValuePtr(enc) : NULL); - VALUE attrArray = rb_funcall(attrHash, rb_intern("to_a"), 0); - attrArray = rb_funcall(attrArray, rb_intern("flatten"), 0); - rb_funcall(doc, rb_intern("start_element"), 2, name, attrArray); - } - + rb_funcall( self, + rb_intern("start_element_namespace"), + 5, + NOKOGIRI_STR_NEW2(localname, STRING_OR_NULL(enc)), + attribute_list, + RBSTR_OR_QNIL(prefix, enc), + RBSTR_OR_QNIL(uri, enc), + ns_list + ); } /** @@ -188,26 +218,23 @@ end_element_ns ( void * ctx, const xmlChar * localname, const xmlChar * prefix, - const xmlChar * URI) + const xmlChar * uri) { VALUE self = (VALUE)ctx; VALUE doc = rb_funcall(self, rb_intern("document"), 0); VALUE MAYBE_UNUSED(enc) = rb_iv_get(self, "@encoding"); - rb_funcall(doc, rb_intern("end_element_ns"), 3, - NOKOGIRI_STR_NEW2(localname, RTEST(enc) ? StringValuePtr(enc) : NULL), - prefix ? NOKOGIRI_STR_NEW2(prefix, RTEST(enc) ? StringValuePtr(enc) : NULL) : Qnil, - URI ? NOKOGIRI_STR_NEW2(URI, RTEST(enc) ? StringValuePtr(enc) : NULL) : Qnil + rb_funcall(doc, rb_intern("end_element_namespace"), 3, + NOKOGIRI_STR_NEW2(localname, STRING_OR_NULL(enc)), + RBSTR_OR_QNIL(prefix, enc), + RBSTR_OR_QNIL(uri, enc) ); - /* Call end element for old-times sake */ - if (rb_respond_to(doc, rb_intern("end_element"))) - { - VALUE name = - NOKOGIRI_STR_NEW2(localname, RTEST(enc) ? StringValuePtr(enc) : NULL); - rb_funcall(doc, rb_intern("end_element"), 1, name); - } - + rb_funcall(self, rb_intern("end_element_namespace"), 3, + NOKOGIRI_STR_NEW2(localname, STRING_OR_NULL(enc)), + RBSTR_OR_QNIL(prefix, enc), + RBSTR_OR_QNIL(uri, enc) + ); } static void characters_func(void * ctx, const xmlChar * ch, int len) @@ -215,7 +242,7 @@ static void characters_func(void * ctx, const xmlChar * ch, int len) VALUE self = (VALUE)ctx; VALUE MAYBE_UNUSED(enc) = rb_iv_get(self, "@encoding"); VALUE doc = rb_funcall(self, rb_intern("document"), 0); - VALUE str = NOKOGIRI_STR_NEW(ch, len, RTEST(enc) ? StringValuePtr(enc):NULL); + VALUE str = NOKOGIRI_STR_NEW(ch, len, STRING_OR_NULL(enc)); rb_funcall(doc, rb_intern("characters"), 1, str); } @@ -224,7 +251,7 @@ static void comment_func(void * ctx, const xmlChar * value) VALUE self = (VALUE)ctx; VALUE MAYBE_UNUSED(enc) = rb_iv_get(self, "@encoding"); VALUE doc = rb_funcall(self, rb_intern("document"), 0); - VALUE str = NOKOGIRI_STR_NEW2(value, RTEST(enc) ? StringValuePtr(enc):NULL); + VALUE str = NOKOGIRI_STR_NEW2(value, STRING_OR_NULL(enc)); rb_funcall(doc, rb_intern("comment"), 1, str); } @@ -241,7 +268,7 @@ static void warning_func(void * ctx, const char *msg, ...) va_end(args); rb_funcall(doc, rb_intern("warning"), 1, - NOKOGIRI_STR_NEW2(message, RTEST(enc) ? StringValuePtr(enc) : NULL) + NOKOGIRI_STR_NEW2(message, STRING_OR_NULL(enc)) ); free(message); } @@ -259,7 +286,7 @@ static void error_func(void * ctx, const char *msg, ...) va_end(args); rb_funcall(doc, rb_intern("error"), 1, - NOKOGIRI_STR_NEW2(message, RTEST(enc) ? StringValuePtr(enc) : NULL) + NOKOGIRI_STR_NEW2(message, STRING_OR_NULL(enc)) ); free(message); } @@ -270,7 +297,7 @@ static void cdata_block(void * ctx, const xmlChar * value, int len) VALUE MAYBE_UNUSED(enc) = rb_iv_get(self, "@encoding"); VALUE doc = rb_funcall(self, rb_intern("document"), 0); VALUE string = - NOKOGIRI_STR_NEW(value, len, RTEST(enc) ? StringValuePtr(enc) : NULL); + NOKOGIRI_STR_NEW(value, len, STRING_OR_NULL(enc)); rb_funcall(doc, rb_intern("cdata_block"), 1, string); } ext/nokogiri/xml_sax_parser.c @@ -1,3 +1,4 @@ require 'nokogiri/xml/sax/document' +require 'nokogiri/xml/sax/legacy_handlers' require 'nokogiri/xml/sax/parser' require 'nokogiri/xml/sax/push_parser' lib/nokogiri/xml/sax.rb @@ -93,11 +93,11 @@ module Nokogiri ### # Called at the beginning of an element # +name+ is the element name - # +attrs+ is a hash of attributes + # +attrs+ is a list of attributes # +prefix+ is the namespace prefix for the element # +uri+ is the associated namespace URI # +ns+ is a hash of namespace prefix:urls associated with the element - def start_element_ns name, attrs = {}, prefix = nil, uri = nil, ns = {} + def start_element_namespace name, attrs = {}, prefix = nil, uri = nil, ns = {} end ### @@ -105,7 +105,7 @@ module Nokogiri # +name+ is the element's name # +prefix+ is the namespace prefix associated with the element # +uri+ is the associated namespace URI - def end_element_ns name, prefix = nil, uri = nil + def end_element_namespace name, prefix = nil, uri = nil end ### lib/nokogiri/xml/sax/document.rb @@ -30,6 +30,9 @@ module Nokogiri # For more information about SAX parsers, see Nokogiri::XML::SAX. Also # see Nokogiri::XML::SAX::Document for the available events. class Parser + class Attribute < Struct.new(:localname, :prefix, :uri, :value) + end + # Encodinds this parser supports ENCODINGS = { 'NONE' => 0, # No char encoding detected @@ -67,6 +70,7 @@ module Nokogiri def initialize(doc = Nokogiri::XML::SAX::Document.new, encoding = 'ASCII') @encoding = encoding @document = doc + @warned = false end ### @@ -95,6 +99,9 @@ module Nokogiri raise Errno::EISDIR if File.directory?(filename) native_parse_file filename end + + private + include Nokogiri::XML::SAX::LegacyHandlers end end end lib/nokogiri/xml/sax/parser.rb @@ -54,6 +54,9 @@ module Nokogiri def finish write '', true end + + private + include Nokogiri::XML::SAX::LegacyHandlers end end end lib/nokogiri/xml/sax/push_parser.rb @@ -19,7 +19,7 @@ module XSD # :nodoc: # require 'xsd/xmlparser/nokogiri' # # obj = AvlPortType.new - # obj.getLatestByRoute(obj.getAgencies.item, 8).item.each do |bus| + # obj.getLatestByRoute(obj.getAgencies.first, 8).each do |bus| # p "#{bus.routeID}, #{bus.longitude}, #{bus.latitude}" # end # @@ -62,7 +62,7 @@ module XSD # :nodoc: characters string end - %w{ start_document start_element_ns end_element_ns end_document comment }.each do |name| + %w{ start_document start_element_namespace end_element_namespace end_document comment }.each do |name| class_eval %{ def #{name}(*args); end } end add_factory(self) lib/xsd/xmlparser/nokogiri.rb @@ -56,10 +56,10 @@ module Nokogiri module SAX class TestCase < Nokogiri::TestCase class Doc < XML::SAX::Document - attr_reader :start_elements, :start_elements_ns, :start_document_called - attr_reader :end_elements, :end_elements_ns, :end_document_called - attr_reader :data, :comments, :cdata_blocks - attr_reader :errors, :warnings + attr_reader :start_elements, :start_document_called + attr_reader :end_elements, :end_document_called + attr_reader :data, :comments, :cdata_blocks, :start_elements_namespace + attr_reader :errors, :warnings, :end_elements_namespace def start_document @start_document_called = true @@ -86,8 +86,8 @@ module Nokogiri super end - def start_element_ns *args - (@start_elements_ns ||= []) << args + def start_element_namespace *args + (@start_elements_namespace ||= []) << args super end @@ -96,8 +96,8 @@ module Nokogiri super end - def end_element_ns *args - (@end_elements_ns ||= []) << args + def end_element_namespace *args + (@end_elements_namespace ||= []) << args super end test/helper.rb @@ -9,15 +9,86 @@ module Nokogiri @parser = XML::SAX::Parser.new(Doc.new) end + def test_namespace_declaration_order_is_saved + @parser.parse <<-eoxml +<root xmlns:foo='http://foo.example.com/' xmlns='http://example.com/'> + <a foo:bar='hello' /> +</root> + eoxml + assert_equal 2, @parser.document.start_elements_namespace.length + el = @parser.document.start_elements_namespace.first + namespaces = el.last + assert_equal ['foo', 'http://foo.example.com/'], namespaces.first + assert_equal [nil, 'http://example.com/'], namespaces.last + end + def test_bad_document_calls_error_handler @parser.parse('<foo><bar></foo>') assert @parser.document.errors assert @parser.document.errors.length > 0 end + def test_namespace_are_super_fun_to_parse + @parser.parse <<-eoxml +<root xmlns:foo='http://foo.example.com/'> + <a foo:bar='hello' /> + <b xmlns:foo='http://bar.example.com/'> + <a foo:bar='hello' /> + </b> + <foo:bar>hello world</foo:bar> +</root> + eoxml + assert @parser.document.start_elements_namespace.length > 0 + el = @parser.document.start_elements_namespace[1] + assert_equal 'a', el.first + assert_equal 1, el[1].length + + attribute = el[1].first + assert_equal 'bar', attribute.localname + assert_equal 'foo', attribute.prefix + assert_equal 'hello', attribute.value + assert_equal 'http://foo.example.com/', attribute.uri + end + + def test_sax_v1_namespace_attribute_declarations + @parser.parse <<-eoxml +<root xmlns:foo='http://foo.example.com/' xmlns='http://example.com/'> + <a foo:bar='hello' /> + <b xmlns:foo='http://bar.example.com/'> + <a foo:bar='hello' /> + </b> + <foo:bar>hello world</foo:bar> +</root> + eoxml + assert @parser.document.start_elements.length > 0 + elm = @parser.document.start_elements.first + assert_equal 'root', elm.first + assert elm[1].include? ['xmlns:foo', 'http://foo.example.com/'] + assert elm[1].include? ['xmlns', 'http://example.com/'] + end + + def test_sax_v1_namespace_nodes + @parser.parse <<-eoxml +<root xmlns:foo='http://foo.example.com/' xmlns='http://example.com/'> + <a foo:bar='hello' /> + <b xmlns:foo='http://bar.example.com/'> + <a foo:bar='hello' /> + </b> + <foo:bar>hello world</foo:bar> +</root> + eoxml + assert_equal 5, @parser.document.start_elements.length + assert @parser.document.start_elements.map { |se| + se.first + }.include? 'foo:bar' + assert @parser.document.end_elements.map { |se| + se.first + }.include? 'foo:bar' + end + def test_start_is_called_without_namespace @parser.parse('<foo:f><bar></foo:f>') - assert_equal ['f', 'bar'], + assert_equal ['foo:f', 'bar'], @parser.document.start_elements.map { |x| x.first } end test/xml/sax/test_parser.rb @@ -43,12 +43,16 @@ module Nokogiri <stream:stream xmlns='jabber:client' xmlns:stream='http://etherx.jabber.org/streams' version='1.0' size='large'></stream:stream> eoxml - assert_equal [[ 'stream', - {'version' => '1.0', 'size' => 'large'}, - 'stream', - 'http://etherx.jabber.org/streams', - {nil => 'jabber:client', 'stream' => 'http://etherx.jabber.org/streams'}]], - @parser.document.start_elements_ns + assert_equal 1, @parser.document.start_elements_namespace.length + el = @parser.document.start_elements_namespace.first + + assert_equal 'stream', el.first + assert_equal 2, el[1].length + assert_equal [['version', '1.0'], ['size', 'large']], + el[1].map { |x| [x.localname, x.value] } + + assert_equal 'stream', el[2] + assert_equal 'http://etherx.jabber.org/streams', el[3] @parser.finish end @@ -58,7 +62,7 @@ module Nokogiri eoxml assert_equal [['stream', 'stream', 'http://etherx.jabber.org/streams']], - @parser.document.end_elements_ns + @parser.document.end_elements_namespace @parser.finish end test/xml/sax/test_push_parser.rb a69f9957de45845f5bacb721e7e10b8acd818431 Aaron Patterson aaron.patterson@gmail.com http://github.com/tenderlove/nokogiri/commit/507b912cc4cad68a98d1a3ff37195d180b08c71a 507b912cc4cad68a98d1a3ff37195d180b08c71a 2009-06-20T19:33:02-07:00 2009-06-20T19:33:02-07:00 Squashed commit of the following: commit cb697a95616d240d2a7d81c284e920753f890231 Author: Aaron Patterson <aaron.patterson@gmail.com> Date: Sat Jun 20 19:32:40 2009 -0700 making sure namespace declaration order is preserved commit 51e1d483215e98fa17217cf86683973f1b9ab896 Author: Aaron Patterson <aaron.patterson@gmail.com> Date: Sat Jun 20 19:00:48 2009 -0700 making sax v1 parser work properly commit d4f3091b31a03a78067a67b8dd0e50d83a2500c8 Author: Aaron Patterson <aaron.patterson@gmail.com> Date: Sat Jun 20 17:54:36 2009 -0700 intermediate fixes d54cc5bb0080fe4f4a73bb16228ad74766d4cc0b Aaron Patterson aaron.patterson@gmail.com