class Nokogiri::XML::SAX::PushParser

PushParser can parse a document that is fed to it manually. It must be given a SAX::Document object which will be called with SAX events as the document is being parsed.

Calling #<< writes XML to the parser, calling any SAX callbacks it can.

#finish tells the parser that the document is finished and calls the end_document SAX method.

Example:

parser = PushParser.new(Class.new(XML::SAX::Document) {
  def start_document
    puts "start document called"
  end
}.new)
parser << "<div>hello<"
parser << "/div>"
parser.finish

Attributes

document[RW]

The Nokogiri::XML::SAX::Document on which the PushParser will be operating

Public Class Methods

new(doc = XML::SAX::Document.new, file_name = nil, encoding = 'UTF-8') click to toggle source

Create a new PushParser with doc as the SAX Document, providing an optional file_name and encoding

# File lib/nokogiri/xml/sax/push_parser.rb, line 34
def initialize(doc = XML::SAX::Document.new, file_name = nil, encoding = 'UTF-8')
  @document = doc
  @encoding = encoding
  @sax_parser = XML::SAX::Parser.new(doc)

  ## Create our push parser context
  initialize_native(@sax_parser, file_name)
end

Public Instance Methods

<<(chunk, last_chunk = false)
Alias for: write
finish() click to toggle source

Finish the parsing. This method is only necessary for Nokogiri::XML::SAX::Document#end_document to be called.

# File lib/nokogiri/xml/sax/push_parser.rb, line 54
def finish
  write '', true
end
options() click to toggle source
static VALUE get_options(VALUE self)
{
  xmlParserCtxtPtr ctx;
  Data_Get_Struct(self, xmlParserCtxt, ctx);

  return INT2NUM(ctx->options);
}
options=(p1) click to toggle source
static VALUE set_options(VALUE self, VALUE options)
{
  xmlParserCtxtPtr ctx;
  Data_Get_Struct(self, xmlParserCtxt, ctx);

  if (xmlCtxtUseOptions(ctx, (int)NUM2INT(options)) != 0)
    rb_raise(rb_eRuntimeError, "Cannot set XML parser context options");

  return Qnil;
}
write(chunk, last_chunk = false) click to toggle source

Write a chunk of XML to the PushParser. Any callback methods that can be called will be called immediately.

# File lib/nokogiri/xml/sax/push_parser.rb, line 46
def write chunk, last_chunk = false
  native_write(chunk, last_chunk)
end
Also aliased as: <<

Private Instance Methods

initialize_native(xml_sax, filename) click to toggle source

Initialize the push parser with xml_sax using filename

static VALUE initialize_native(VALUE self, VALUE _xml_sax, VALUE _filename)
{
  xmlSAXHandlerPtr sax;
  const char * filename = NULL;
  xmlParserCtxtPtr ctx;

  Data_Get_Struct(_xml_sax, xmlSAXHandler, sax);

  if(_filename != Qnil) filename = StringValuePtr(_filename);

  ctx = xmlCreatePushParserCtxt(
      sax,
      NULL,
      NULL,
      0,
      filename
  );
  if(ctx == NULL)
    rb_raise(rb_eRuntimeError, "Could not create a parser context");

  ctx->userData = NOKOGIRI_SAX_TUPLE_NEW(ctx, self);

  ctx->sax2 = 1;
  DATA_PTR(self) = ctx;
  return self;
}
native_write(chunk, last_chunk) click to toggle source

Write chunk to PushParser. last_chunk triggers the end_document handle

static VALUE native_write(VALUE self, VALUE _chunk, VALUE _last_chunk)
{
  xmlParserCtxtPtr ctx;
  const char * chunk  = NULL;
  int size            = 0;


  Data_Get_Struct(self, xmlParserCtxt, ctx);

  if(Qnil != _chunk) {
    chunk = StringValuePtr(_chunk);
    size = (int)RSTRING_LEN(_chunk);
  }

  if(xmlParseChunk(ctx, chunk, size, Qtrue == _last_chunk ? 1 : 0)) {
    if (!(ctx->options & XML_PARSE_RECOVER)) {
      xmlErrorPtr e = xmlCtxtGetLastError(ctx);
      Nokogiri_error_raise(NULL, e);
    }
  }

  return self;
}