caring-nokogiri 1.4.1.pre1

data/lib/nokogiri/ffi/xml/xpath.rb

@@ -0,0 +1,19 @@
+module Nokogiri
+  module XML
+    class XPath
+      
+      attr_accessor :cstruct # :nodoc:
+
+      def node_set # :nodoc:
+        ptr = cstruct[:nodesetval] if cstruct[:nodesetval]
+        ptr = LibXML.xmlXPathNodeSetCreate(nil) if ptr.null?
+
+        set = XML::NodeSet.new(@document)
+        set.cstruct = LibXML::XmlNodeSet.new(ptr)
+        set.document = @document
+        set
+      end
+
+    end
+  end
+end

data/lib/nokogiri/ffi/xml/xpath_context.rb

@@ -0,0 +1,135 @@
+module Nokogiri
+  module XML
+    class XPathContext
+
+      attr_accessor :cstruct # :nodoc:
+
+      def register_ns(prefix, uri) # :nodoc:
+        LibXML.xmlXPathRegisterNs(cstruct, prefix, uri)
+      end
+
+      def evaluate(search_path, xpath_handler=nil) # :nodoc:
+        lookup = nil # to keep lambda in scope long enough to avoid a possible GC tragedy
+        query = search_path.to_s
+
+        if xpath_handler
+          lookup = lambda do |ctx, name, uri|
+            return nil unless xpath_handler.respond_to?(name)
+            ruby_funcall name, xpath_handler
+          end
+          LibXML.xmlXPathRegisterFuncLookup(cstruct, lookup, nil);
+        end
+
+        exception_handler = lambda do |ctx, error|
+          raise XPath::SyntaxError.wrap(error)
+        end
+        LibXML.xmlResetLastError()
+        LibXML.xmlSetStructuredErrorFunc(nil, exception_handler)
+
+        generic_exception_handler = lambda do |ctx, msg|
+          raise RuntimeError.new(msg) # TODO: varargs
+        end
+        LibXML.xmlSetGenericErrorFunc(nil, generic_exception_handler)
+
+        xpath_ptr = LibXML.xmlXPathEvalExpression(query, cstruct)
+
+        LibXML.xmlSetStructuredErrorFunc(nil, nil)
+        LibXML.xmlSetGenericErrorFunc(nil, nil)
+
+        if xpath_ptr.null?
+          error = LibXML.xmlGetLastError()
+          raise XPath::SyntaxError.wrap(error)
+        end
+
+        xpath = XML::XPath.new
+        xpath.cstruct = LibXML::XmlXpathObject.new(xpath_ptr)
+        xpath.document = cstruct[:doc]
+        xpath
+      end
+
+      def self.new(node) # :nodoc:
+        LibXML.xmlXPathInit()
+
+        ptr = LibXML.xmlXPathNewContext(node.cstruct[:doc])
+
+        ctx = allocate
+        ctx.cstruct = LibXML::XmlXpathContext.new(ptr)
+        ctx.cstruct[:node] = node.cstruct
+        ctx
+      end
+
+      private
+
+      #
+      #  returns a lambda that will call the handler function with marshalled parameters
+      #
+      def ruby_funcall(name, xpath_handler) # :nodoc:
+        lambda do |ctx, nargs|
+          parser_context = LibXML::XmlXpathParserContext.new(ctx)
+          context = parser_context.context
+          doc = context.doc.ruby_doc
+
+          params = []
+
+          nargs.times do |j|
+            obj = LibXML::XmlXpathObject.new(LibXML.valuePop(ctx))
+            case obj[:type]
+            when LibXML::XmlXpathObject::XPATH_STRING
+              params.unshift obj[:stringval]
+            when LibXML::XmlXpathObject::XPATH_BOOLEAN
+              params.unshift obj[:boolval] == 1
+            when LibXML::XmlXpathObject::XPATH_NUMBER
+              params.unshift obj[:floatval]
+            when LibXML::XmlXpathObject::XPATH_NODESET
+              ns_ptr = LibXML::XmlNodeSet.new(obj[:nodesetval])
+              set = NodeSet.allocate
+              set.cstruct = ns_ptr
+              params.unshift set
+            else
+              char_ptr = params.unshift LibXML.xmlXPathCastToString(obj)
+              string = char_ptr.read_string
+              LibXML.xmlFree(char_ptr)
+              string
+            end
+          end
+
+          result = xpath_handler.send(name, *params)
+
+          case result.class.to_s
+          when Fixnum.to_s, Float.to_s, Bignum.to_s
+            LibXML.xmlXPathReturnNumber(ctx, result)
+          when String.to_s
+            LibXML.xmlXPathReturnString(
+              ctx,
+              LibXML.xmlXPathWrapCString(result)
+              )
+          when TrueClass.to_s
+            LibXML.xmlXPathReturnTrue(ctx)
+          when FalseClass.to_s
+            LibXML.xmlXPathReturnFalse(ctx)
+          when NilClass.to_s
+            ;
+          when Array.to_s
+            node_set = XML::NodeSet.new(doc, result)
+            LibXML.xmlXPathReturnNodeSet(
+              ctx,
+              LibXML.xmlXPathNodeSetMerge(nil, node_set.cstruct)
+              )
+          else
+            if result.is_a?(XML::NodeSet)
+              LibXML.xmlXPathReturnNodeSet(
+                ctx,
+                LibXML.xmlXPathNodeSetMerge(nil, result.cstruct)
+                )
+            else
+              raise RuntimeError.new("Invalid return type #{result.class.inspect}")
+            end
+          end
+
+          nil
+        end # lambda
+      end # ruby_funcall
+
+    end
+  end
+end

data/lib/nokogiri/ffi/xslt/stylesheet.rb

@@ -0,0 +1,47 @@
+module Nokogiri
+  module XSLT
+    class Stylesheet
+
+      attr_accessor :cstruct # :nodoc:
+
+      def self.parse_stylesheet_doc(document) # :nodoc:
+        LibXML.exsltRegisterAll
+
+        generic_exception_handler = lambda do |ctx, msg|
+          raise RuntimeError.new(msg) # TODO: varargs
+        end
+        LibXML.xsltSetGenericErrorFunc(nil, generic_exception_handler)
+
+        ss = LibXML.xsltParseStylesheetDoc(LibXML.xmlCopyDoc(document.cstruct, 1)) # 1 => recursive
+
+        LibXML.xsltSetGenericErrorFunc(nil, nil)
+
+        obj = allocate
+        obj.cstruct = LibXML::XsltStylesheet.new(ss)
+        obj
+      end
+
+      def serialize(document) # :nodoc:
+        buf_ptr = FFI::MemoryPointer.new :pointer
+        buf_len = FFI::MemoryPointer.new :int
+        LibXML.xsltSaveResultToString(buf_ptr, buf_len, document.cstruct, cstruct)
+        buf = Nokogiri::LibXML::XmlAlloc.new(buf_ptr.read_pointer)
+        buf.pointer.read_string(buf_len.read_int)
+      end
+
+      def transform(document, params=[]) # :nodoc:
+        param_arr = FFI::MemoryPointer.new(:pointer, params.length + 1)
+        params.each_with_index do |param, j|
+          param_arr[j].put_pointer(0, FFI::MemoryPointer.from_string(param.to_s))
+        end
+        param_arr[params.length].put_pointer(0,nil)
+
+        ptr = LibXML.xsltApplyStylesheet(cstruct, document.cstruct, param_arr)
+        raise(RuntimeError, "could not perform xslt transform on document") if ptr.null?
+
+        XML::Document.wrap(ptr)
+      end
+
+    end
+  end
+end

data/lib/nokogiri/html.rb

@@ -0,0 +1,35 @@
+require 'nokogiri/html/entity_lookup'
+require 'nokogiri/html/document'
+require 'nokogiri/html/document_fragment'
+require 'nokogiri/html/sax/parser_context'
+require 'nokogiri/html/sax/parser'
+require 'nokogiri/html/element_description'
+
+module Nokogiri
+  class << self
+    ###
+    # Parse HTML.  Convenience method for Nokogiri::HTML::Document.parse
+    def HTML thing, url = nil, encoding = nil, options = XML::ParseOptions::DEFAULT_HTML, &block
+      Nokogiri::HTML::Document.parse(thing, url, encoding, options, &block)
+    end
+  end
+
+  module HTML
+    class << self
+      ###
+      # Parse HTML.  Convenience method for Nokogiri::HTML::Document.parse
+      def parse thing, url = nil, encoding = nil, options = XML::ParseOptions::DEFAULT_HTML, &block
+        Document.parse(thing, url, encoding, options, &block)
+      end
+
+      ####
+      # Parse a fragment from +string+ in to a NodeSet.
+      def fragment string
+        HTML::DocumentFragment.parse(string)
+      end
+    end
+
+    # Instance of Nokogiri::HTML::EntityLookup
+    NamedCharacters = EntityLookup.new
+  end
+end

data/lib/nokogiri/html/builder.rb

@@ -0,0 +1,35 @@
+module Nokogiri
+  module HTML
+    ###
+    # Nokogiri HTML builder is used for building HTML documents.  It is very
+    # similar to the Nokogiri::XML::Builder.  In fact, you should go read the
+    # documentation for Nokogiri::XML::Builder before reading this
+    # documentation.
+    #
+    # == Synopsis:
+    #
+    # Create an HTML document with a body that has an onload attribute, and a
+    # span tag with a class of "bold" that has content of "Hello world".
+    #
+    #   builder = Nokogiri::HTML::Builder.new do |doc|
+    #     doc.html {
+    #       doc.body(:onload => 'some_func();') {
+    #         doc.span.bold {
+    #           doc.text "Hello world"
+    #         }
+    #       }
+    #     }
+    #   end
+    #   puts builder.to_html
+    #
+    # The HTML builder inherits from the XML builder, so make sure to read the
+    # Nokogiri::XML::Builder documentation.
+    class Builder < Nokogiri::XML::Builder
+      ###
+      # Convert the builder to HTML
+      def to_html
+        @doc.to_html
+      end
+    end
+  end
+end

data/lib/nokogiri/html/document.rb

@@ -0,0 +1,88 @@
+module Nokogiri
+  module HTML
+    class Document < Nokogiri::XML::Document
+      ###
+      # Get the meta tag encoding for this document.  If there is no meta tag,
+      # then nil is returned
+      def meta_encoding
+        return nil unless meta = css('meta').find { |node|
+          node['http-equiv'] =~ /Content-Type/i
+        }
+
+        /charset\s*=\s*([\w\d-]+)/i.match(meta['content'])[1]
+      end
+
+      ###
+      # Set the meta tag encoding for this document.  If there is no meta 
+      # content tag, nil is returned and the encoding is not set.
+      def meta_encoding= encoding
+        return nil unless meta = css('meta').find { |node|
+          node['http-equiv'] =~ /Content-Type/i
+        }
+
+        meta['content'] = "text/html; charset=%s" % encoding
+        encoding
+      end
+
+      ####
+      # Serialize Node using +options+.  Save options can also be set using a
+      # block. See SaveOptions.
+      #
+      # These two statements are equivalent:
+      #
+      #  node.serialize(:encoding => 'UTF-8', :save_with => FORMAT | AS_XML)
+      #
+      # or
+      #
+      #   node.serialize(:encoding => 'UTF-8') do |config|
+      #     config.format.as_xml
+      #   end
+      #
+      def serialize options = {}, &block
+        options[:save_with] ||= XML::Node::SaveOptions::FORMAT |
+            XML::Node::SaveOptions::AS_HTML |
+            XML::Node::SaveOptions::NO_DECLARATION |
+            XML::Node::SaveOptions::NO_EMPTY_TAGS
+        super
+      end
+
+      ####
+      # Create a Nokogiri::XML::DocumentFragment from +tags+
+      def fragment tags = nil
+        DocumentFragment.new(self, tags)
+      end
+
+      class << self
+        ###
+        # Parse HTML.  +thing+ may be a String, or any object that
+        # responds to _read_ and _close_ such as an IO, or StringIO.
+        # +url+ is resource where this document is located.  +encoding+ is the
+        # encoding that should be used when processing the document. +options+
+        # is a number that sets options in the parser, such as
+        # Nokogiri::XML::ParseOptions::RECOVER.  See the constants in
+        # Nokogiri::XML::ParseOptions.
+        def parse string_or_io, url = nil, encoding = nil, options = XML::ParseOptions::DEFAULT_HTML, &block
+
+          options = Nokogiri::XML::ParseOptions.new(options) if Fixnum === options
+          # Give the options to the user
+          yield options if block_given?
+
+          if string_or_io.respond_to?(:encoding)
+            encoding ||= string_or_io.encoding.name
+          end
+
+          if string_or_io.respond_to?(:read)
+            url ||= string_or_io.respond_to?(:path) ? string_or_io.path : nil
+            return read_io(string_or_io, url, encoding, options.to_i)
+          end
+
+          # read_memory pukes on empty docs
+          return new if string_or_io.nil? or string_or_io.empty?
+
+          read_memory(string_or_io, url, encoding, options.to_i)
+        end
+      end
+
+    end
+  end
+end

data/lib/nokogiri/html/document_fragment.rb

@@ -0,0 +1,15 @@
+module Nokogiri
+  module HTML
+    class DocumentFragment < Nokogiri::XML::DocumentFragment
+
+      class << self
+        ####
+        # Create a Nokogiri::XML::DocumentFragment from +tags+
+        def parse tags
+          self.new(HTML::Document.new, tags)
+        end
+      end
+
+    end
+  end
+end

data/lib/nokogiri/html/element_description.rb

@@ -0,0 +1,23 @@
+module Nokogiri
+  module HTML
+    class ElementDescription
+      ###
+      # Is this element a block element?
+      def block?
+        !inline?
+      end
+
+      ###
+      # Convert this description to a string
+      def to_s
+        "#{name}: #{description}"
+      end
+
+      ###
+      # Inspection information
+      def inspect
+        "#<#{self.class.name}: #{name} #{description}>"
+      end
+    end
+  end
+end

data/lib/nokogiri/html/entity_lookup.rb

@@ -0,0 +1,13 @@
+module Nokogiri
+  module HTML
+    class EntityDescription < Struct.new(:value, :name, :description); end
+
+    class EntityLookup
+      ###
+      # Look up entity with +name+
+      def [] name
+        (val = get(name)) && val.value
+      end
+    end
+  end
+end

data/lib/nokogiri/html/sax/parser.rb

@@ -0,0 +1,48 @@
+module Nokogiri
+  module HTML
+    ###
+    # Nokogiri lets you write a SAX parser to process HTML but get HTML
+    # correction features.
+    #
+    # See Nokogiri::HTML::SAX::Parser for a basic example of using a
+    # SAX parser with HTML.
+    #
+    # For more information on SAX parsers, see Nokogiri::XML::SAX
+    module SAX
+      ###
+      # This class lets you perform SAX style parsing on HTML with HTML
+      # error correction.
+      #
+      # Here is a basic usage example:
+      #
+      #   class MyDoc < Nokogiri::XML::SAX::Document
+      #     def start_element name, attributes = []
+      #       puts "found a #{name}"
+      #     end
+      #   end
+      #
+      #   parser = Nokogiri::HTML::SAX::Parser.new(MyDoc.new)
+      #   parser.parse(File.read(ARGV[0], 'rb'))
+      #
+      # For more information on SAX parsers, see Nokogiri::XML::SAX
+      class Parser < Nokogiri::XML::SAX::Parser
+        ###
+        # Parse html stored in +data+ using +encoding+
+        def parse_memory data, encoding = 'UTF-8'
+          raise ArgumentError unless data
+          return unless data.length > 0
+          ParserContext.memory(data, encoding).parse_with self
+        end
+
+        ###
+        # Parse a file with +filename+
+        def parse_file filename, encoding = 'UTF-8'
+          raise ArgumentError unless filename
+          raise Errno::ENOENT unless File.exists?(filename)
+          raise Errno::EISDIR if File.directory?(filename)
+          ParserContext.file(filename, encoding).parse_with self
+        end
+      end
+    end
+  end
+end