coupa-libxml-ruby 1.1.4

data/ext/libxml/ruby_xml_error.h

@@ -0,0 +1,14 @@
+/* $Id: rxml_ns.h 324 2008-07-08 23:00:02Z cfis $ */
+
+/* Please see the LICENSE file for copyright and distribution information */
+
+#ifndef __RXML_ERROR__
+#define __RXML_ERROR__
+
+extern VALUE eXMLError;
+
+void rxml_init_error();
+VALUE rxml_error_wrap(xmlErrorPtr xerror);
+void rxml_raise(xmlErrorPtr xerror);
+
+#endif

data/ext/libxml/ruby_xml_html_parser.c

@@ -0,0 +1,92 @@
+/* $Id: ruby_xml_html_parser.c 758 2009-01-25 20:36:03Z cfis $ */
+
+/* Please see the LICENSE file for copyright and distribution information */
+
+#include "ruby_libxml.h"
+
+/* Document-class: LibXML::XML::HTMLParser
+ *
+ * The HTML parser implements an HTML 4.0 non-verifying parser with an API
+ * compatible with the XML::Parser.  In contrast with the XML::Parser,
+ * it can parse "real world" HTML, even if it severely broken from a
+ * specification point of view.
+ *
+ * The HTML parser creates an in-memory document object
+ * that consist of any number of XML::Node instances.  This is simple
+ * and powerful model, but has the major limitation that the size of
+ * the document that can be processed is limited by the amount of
+ * memory available.
+ *
+ * Using the html parser is simple:
+ *
+ *   parser = XML::HTMLParser.file('my_file')
+ *   doc = parser.parse
+ *
+ * You can also parse documents (see XML::HTMLParser.document), 
+ * strings (see XML::HTMLParser.string) and io objects (see
+ * XML::HTMLParser.io).
+ */
+
+VALUE cXMLHtmlParser;
+static ID CONTEXT_ATTR;
+
+
+/* call-seq:
+ *    XML::HTMLParser.initialize -> parser
+ *
+ * Initializes a new parser instance with no pre-determined source.
+ */
+static VALUE rxml_html_parser_initialize(int argc, VALUE *argv, VALUE self)
+{
+  VALUE context = Qnil;
+
+  rb_scan_args(argc, argv, "01", &context);
+
+  if (context == Qnil)
+  {
+    rb_warn("Passing no parameters to XML::HTMLParser.new is deprecated.  Pass an instance of XML::Parser::Context instead.");
+    context = rb_class_new_instance(0, NULL, cXMLParserContext);
+  }
+
+  rb_ivar_set(self, CONTEXT_ATTR, context);
+  return self;
+}
+
+/*
+ * call-seq:
+ *    parser.parse -> XML::Document
+ *
+ * Parse the input XML and create an XML::Document with
+ * it's content. If an error occurs, XML::Parser::ParseError
+ * is thrown.
+ */
+static VALUE rxml_html_parser_parse(VALUE self)
+{
+  xmlParserCtxtPtr ctxt;
+  VALUE context = rb_ivar_get(self, CONTEXT_ATTR);
+  
+  Data_Get_Struct(context, xmlParserCtxt, ctxt);
+
+  if (htmlParseDocument(ctxt) == -1 && ! ctxt->recovery)
+  {
+    if (ctxt->myDoc)
+      xmlFreeDoc(ctxt->myDoc);
+    rxml_raise(&ctxt->lastError);
+  }
+
+  return rxml_document_wrap(ctxt->myDoc);
+}
+
+void rxml_init_html_parser(void)
+{
+  CONTEXT_ATTR = rb_intern("@context");
+
+  cXMLHtmlParser = rb_define_class_under(mXML, "HTMLParser", rb_cObject);
+
+  /* Atributes */
+  rb_define_attr(cXMLHtmlParser, "input", 1, 0);
+
+  /* Instance methods */
+  rb_define_method(cXMLHtmlParser, "initialize", rxml_html_parser_initialize, -1);
+  rb_define_method(cXMLHtmlParser, "parse", rxml_html_parser_parse, 0);
+}

data/ext/libxml/ruby_xml_html_parser.h

@@ -0,0 +1,12 @@
+/* $Id: ruby_xml_html_parser.h 758 2009-01-25 20:36:03Z cfis $ */
+
+/* Please see the LICENSE file for copyright and distribution information */
+
+#ifndef __RXML_HTML_PARSER__
+#define __RXML_HTML_PARSER__
+
+extern VALUE cXMLHtmlParser;
+
+void rxml_init_html_parser(void);
+
+#endif

data/ext/libxml/ruby_xml_html_parser_context.c

@@ -0,0 +1,308 @@
+/* $Id: ruby_xml_parser_context.c 710 2009-01-20 05:30:51Z cfis $ */
+
+/* Please see the LICENSE file for copyright and distribution information */
+
+#include "ruby_libxml.h"
+#include "ruby_xml_html_parser_context.h"
+
+
+/*
+ * Document-class: LibXML::XML::HTMLParser::Context
+ *
+ * The XML::HTMLParser::Context class provides in-depth control over how
+ * a document is parsed.
+ */
+
+VALUE cXMLHtmlParserContext;
+static ID IO_ATTR;
+
+/* OS X 10.5 ships with libxml2 version 2.6.16 which does not expose the
+   htmlNewParserCtxt (or htmlInitParserCtxt which it uses) method.  htmlNewParserCtxt
+   wasn't added to the libxml2 header files until 2.6.27.  So the next two 
+   methods are simply copied from a newer version of libxml2 (2.7.2). */
+#if LIBXML_VERSION < 20627
+#define XML_CTXT_FINISH_DTD_0 0xabcd1234
+static int htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
+{
+  htmlSAXHandler *sax;
+  if (ctxt == NULL) return(-1);
+  
+  memset(ctxt, 0, sizeof(htmlParserCtxt));
+  ctxt->dict = xmlDictCreate();
+  if (ctxt->dict == NULL) {
+    rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
+    return(-1);
+  }
+  sax = (htmlSAXHandler *) xmlMalloc(sizeof(htmlSAXHandler));
+  if (sax == NULL) {
+    rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
+    return(-1);
+  }
+  else
+    memset(sax, 0, sizeof(htmlSAXHandler));
+
+  ctxt->inputTab = (htmlParserInputPtr *) xmlMalloc(5 * sizeof(htmlParserInputPtr));
+  if (ctxt->inputTab == NULL) {
+    rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
+    ctxt->inputNr = 0;
+    ctxt->inputMax = 0;
+    ctxt->input = NULL;
+    return(-1);
+  }
+  ctxt->inputNr = 0;
+  ctxt->inputMax = 5;
+  ctxt->input = NULL;
+  ctxt->version = NULL;
+  ctxt->encoding = NULL;
+  ctxt->standalone = -1;
+  ctxt->instate = XML_PARSER_START;
+
+  ctxt->nodeTab = (htmlNodePtr *) xmlMalloc(10 * sizeof(htmlNodePtr));
+  if (ctxt->nodeTab == NULL) {
+    rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
+    ctxt->nodeNr = 0;
+    ctxt->nodeMax = 0;
+    ctxt->node = NULL;
+    ctxt->inputNr = 0;
+    ctxt->inputMax = 0;
+    ctxt->input = NULL;
+    return(-1);
+  }
+  ctxt->nodeNr = 0;
+  ctxt->nodeMax = 10;
+  ctxt->node = NULL;
+
+  ctxt->nameTab = (const xmlChar **) xmlMalloc(10 * sizeof(xmlChar *));
+  if (ctxt->nameTab == NULL) {
+    rb_raise(rb_eNoMemError, "htmlInitParserCtxt: out of memory\n");
+    ctxt->nameNr = 0;
+    ctxt->nameMax = 10;
+    ctxt->name = NULL;
+    ctxt->nodeNr = 0;
+    ctxt->nodeMax = 0;
+    ctxt->node = NULL;
+    ctxt->inputNr = 0;
+    ctxt->inputMax = 0;
+    ctxt->input = NULL;
+    return(-1);
+  }
+  ctxt->nameNr = 0;
+  ctxt->nameMax = 10;
+  ctxt->name = NULL;
+  	
+  if (sax == NULL) ctxt->sax = (xmlSAXHandlerPtr) &htmlDefaultSAXHandler;
+  else {
+    ctxt->sax = sax;
+    memcpy(sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
+  }
+  ctxt->userData = ctxt;
+  ctxt->myDoc = NULL;
+  ctxt->wellFormed = 1;
+  ctxt->replaceEntities = 0;
+  ctxt->linenumbers = xmlLineNumbersDefaultValue;
+  ctxt->html = 1;
+  ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
+  ctxt->vctxt.userData = ctxt;
+  ctxt->vctxt.error = xmlParserValidityError;
+  ctxt->vctxt.warning = xmlParserValidityWarning;
+  ctxt->record_info = 0;
+  ctxt->validate = 0;
+  ctxt->nbChars = 0;
+  ctxt->checkIndex = 0;
+  ctxt->catalogs = NULL;
+  xmlInitNodeInfoSeq(&ctxt->node_seq);
+  return(0);
+}
+
+static htmlParserCtxtPtr htmlNewParserCtxt(void)
+{
+  xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt));
+  if (ctxt == NULL) {
+    rb_raise(rb_eNoMemError, "NewParserCtxt: out of memory\n");
+    return(NULL);
+  }
+  memset(ctxt, 0, sizeof(xmlParserCtxt));
+  if (htmlInitParserCtxt(ctxt) < 0) {
+    htmlFreeParserCtxt(ctxt);
+   return(NULL);
+  }
+  return(ctxt);
+}
+#endif
+
+static void rxml_html_parser_context_free(htmlParserCtxtPtr ctxt)
+{
+  htmlFreeParserCtxt(ctxt);
+}
+
+static VALUE rxml_html_parser_context_wrap(htmlParserCtxtPtr ctxt)
+{
+  return Data_Wrap_Struct(cXMLHtmlParserContext, NULL, rxml_html_parser_context_free, ctxt);
+}
+
+/* call-seq:
+ *    XML::HTMLParser::Context.file(file) -> XML::HTMLParser::Context
+ *
+ * Creates a new parser context based on the specified file or uri.
+ *
+ * Parameters:
+ *
+ *  file - A filename or uri.
+*/
+static VALUE rxml_html_parser_context_file(VALUE klass, VALUE file)
+{
+  htmlParserCtxtPtr ctxt = htmlCreateFileParserCtxt(StringValuePtr(file), NULL);
+  if (!ctxt)
+    rxml_raise(&xmlLastError);
+
+  /* This is annoying, but xmlInitParserCtxt (called indirectly above) and 
+     xmlCtxtUseOptionsInternal (called below) initialize slightly different
+     context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
+     sets to 0 and xmlCtxtUseOptionsInternal sets to 1.  So we have to call both. */
+  htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
+
+  return rxml_html_parser_context_wrap(ctxt);
+}
+
+/* call-seq:
+ *    XML::HTMLParser::Context.io(io) -> XML::HTMLParser::Context
+ *
+ * Creates a new parser context based on the specified io object.
+ *
+ * Parameters:
+ *
+ *  io - A ruby IO object.
+*/
+static VALUE rxml_html_parser_context_io(VALUE klass, VALUE io)
+{
+  VALUE result;
+  htmlParserCtxtPtr ctxt;
+  xmlParserInputBufferPtr input;
+  xmlParserInputPtr stream;
+
+  if (NIL_P(io))
+    rb_raise(rb_eTypeError, "Must pass in an IO object");
+
+  input = xmlParserInputBufferCreateIO((xmlInputReadCallback) rxml_read_callback, NULL,
+                                     (void*)io, XML_CHAR_ENCODING_NONE);
+
+  ctxt = htmlNewParserCtxt();
+  if (!ctxt)
+  {
+    xmlFreeParserInputBuffer(input);
+    rxml_raise(&xmlLastError);
+  }
+
+  /* This is annoying, but xmlInitParserCtxt (called indirectly above) and 
+     xmlCtxtUseOptionsInternal (called below) initialize slightly different
+     context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
+     sets to 0 and xmlCtxtUseOptionsInternal sets to 1.  So we have to call both. */
+  htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
+
+  stream = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE);
+
+  if (!stream)
+  {
+    xmlFreeParserInputBuffer(input);
+    xmlFreeParserCtxt(ctxt);
+    rxml_raise(&xmlLastError);
+  }
+  inputPush(ctxt, stream);
+  result = rxml_html_parser_context_wrap(ctxt);
+
+  /* Attach io object to parser so it won't get freed.*/
+  rb_ivar_set(result, IO_ATTR, io);
+
+  return result;
+}
+
+/* call-seq:
+ *    XML::HTMLParser::Context.string(string) -> XML::HTMLParser::Context
+ *
+ * Creates a new parser context based on the specified string.
+ *
+ * Parameters:
+ *
+ *  string - A string that contains the data to parse.
+*/
+static VALUE rxml_html_parser_context_string(VALUE klass, VALUE string)
+{
+  htmlParserCtxtPtr ctxt;
+  Check_Type(string, T_STRING);
+
+  if (RSTRING_LEN(string) == 0)
+    rb_raise(rb_eArgError, "Must specify a string with one or more characters");
+
+  ctxt = xmlCreateMemoryParserCtxt(StringValuePtr(string),
+                                   RSTRING_LEN(string));
+  if (!ctxt)
+    rxml_raise(&xmlLastError);
+
+  /* This is annoying, but xmlInitParserCtxt (called indirectly above) and 
+     xmlCtxtUseOptionsInternal (called below) initialize slightly different
+     context options, in particular XML_PARSE_NODICT which xmlInitParserCtxt
+     sets to 0 and xmlCtxtUseOptionsInternal sets to 1.  So we have to call both. */
+  htmlCtxtUseOptions(ctxt, rxml_libxml_default_options());
+
+  htmlDefaultSAXHandlerInit();
+  if (ctxt->sax != NULL)
+    memcpy(ctxt->sax, &htmlDefaultSAXHandler, sizeof(xmlSAXHandlerV1));
+  
+  return rxml_html_parser_context_wrap(ctxt);
+}
+
+/*
+ * call-seq:
+ *    context.disable_cdata = (true|false)
+ *
+ * Control whether the CDATA nodes will be created in this context.
+ */
+static VALUE rxml_html_parser_context_disable_cdata_set(VALUE self, VALUE bool)
+{
+  htmlParserCtxtPtr ctxt;
+  Data_Get_Struct(self, htmlParserCtxt, ctxt);
+
+  if (ctxt->sax == NULL)
+    rb_raise(rb_eRuntimeError, "Sax handler is not yet set");
+
+  /* LibXML controls this internally with the default SAX handler. */ 
+  if (bool)
+    ctxt->sax->cdataBlock = NULL;
+  else
+    ctxt->sax->cdataBlock = htmlDefaultSAXHandler.cdataBlock;
+
+  return bool;
+}
+
+/*
+ * call-seq:
+ *    context.options = XML::Parser::Options::NOENT |
+                        XML::Parser::Options::NOCDATA
+ *
+ * Provides control over the execution of a parser.  Valid values 
+ * are the constants defined on XML::Parser::Options.  Multiple
+ * options can be combined by using Bitwise OR (|).
+ */
+static VALUE rxml_html_parser_context_options_set(VALUE self, VALUE options)
+{
+  int result;
+  htmlParserCtxtPtr ctxt;
+  Check_Type(options, T_FIXNUM);
+
+  Data_Get_Struct(self, htmlParserCtxt, ctxt);
+  result = htmlCtxtUseOptions(ctxt, NUM2INT(options));
+
+  return self;
+}
+
+void rxml_init_html_parser_context(void)
+{
+  IO_ATTR = ID2SYM(rb_intern("@io"));
+  cXMLHtmlParserContext = rb_define_class_under(cXMLHtmlParser, "Context", cXMLParserContext);
+
+  rb_define_singleton_method(cXMLHtmlParserContext, "file", rxml_html_parser_context_file, 1);
+  rb_define_singleton_method(cXMLHtmlParserContext, "io", rxml_html_parser_context_io, 1);
+  rb_define_singleton_method(cXMLHtmlParserContext, "string", rxml_html_parser_context_string, 1);
+  rb_define_method(cXMLHtmlParserContext, "disable_cdata=", rxml_html_parser_context_disable_cdata_set, 1);
+  rb_define_method(cXMLHtmlParserContext, "options=", rxml_html_parser_context_options_set, 1);
+}

data/ext/libxml/ruby_xml_html_parser_context.h

@@ -0,0 +1,12 @@
+/* $Id: ruby_xml_parser_context.h 666 2008-12-07 00:16:50Z cfis $ */
+
+/* Please see the LICENSE file for copyright and distribution information */
+
+#ifndef __RXML_HTML_PARSER_CONTEXT__
+#define __RXML_HTML_PARSER_CONTEXT__
+
+extern VALUE cXMLHtmlParserContext;
+
+void rxml_init_html_parser_context(void);
+
+#endif