parser.h

ACE自适配通信环境（ADAPTIVE Communication Environment）是可以自由使用、开放源码的面向对象（OO）框架（Framework）

  /**
   *  Parse the version string in an XML Prolog section.
   *
   *  @param str String containing the version number if successful.
   *  @return 0 if the string was read successfully, 0 otherwise.
   */
  int parse_version (ACEXML_Char*& str);

  /**
   *  Parse the version number in a VersionInfo declaration.
   */
  int parse_version_num (ACEXML_Char*& str);

  /**
   *  Parse the encoding name in an XML Prolog section.
   *
   *  @param str String containing the encoding name if successful.
   *  @return 0 if the string was read successfully, 0 otherwise.
   */
  int parse_encname (ACEXML_Char*& str);

  /**
   *  Parse a SDDecl string.
   *
   *  @param str String containing the encoding name if successful.
   *  @return 0 if the string was read successfully, -1 otherwise.
   */
  int parse_sddecl (ACEXML_Char*& str);

  /**
   *  Parse an attribute name.
   *
   *  @retval str String containing the value of the attribute name
   *             if successful.
   *  @retval 0 otherwise.
   */
  ACEXML_Char* parse_attname (ACEXML_ENV_SINGLE_ARG_DECL)
    ACE_THROW_SPEC ((ACEXML_SAXException));

  /**
   *  Parse an attribute value.
   *
   *  @param str String containing the value of the attribute if successful.
   *  @return 0 if attribute value was read successfully, -1 otherwise.
   */
  int parse_attvalue (ACEXML_Char*& str ACEXML_ENV_ARG_DECL)
    ACE_THROW_SPEC ((ACEXML_SAXException));

  /**
   *  Parse a tokenized type attribute.
   *
   *  @return 0 if attribute type was read successfully, -1 otherwise.
   */
  int parse_tokenized_type (ACEXML_ENV_SINGLE_ARG_DECL)
    ACE_THROW_SPEC ((ACEXML_SAXException));

  /**
   *  Parse a SystemLiteral.
   *
   *  @param str String containing the SystemLiteral if successful.
   *  @return 0 if the string was read successfully, 0 otherwise.
   */
  int parse_system_literal (ACEXML_Char*& str);

   /**
   *  Parse a PubidLiteral.
   *
   *  @param str String containing the PubidLiteral if successful.
   *  @return 0 if the string was read successfully, 0 otherwise.
   */
  int parse_pubid_literal (ACEXML_Char*& str);

  /**
   * Check if a character @a c is a whitespace.
   *
   * @retval 1 if @a c is a valid white space character. 0 otherwise.
   */
  int is_whitespace (const ACEXML_Char c) const;

  /**
   * Check if a character @a c is a valid Char.
   *
   * @retval 1 if @a c is a valid character. 0 otherwise.
   */
  int isChar (ACEXML_UCS4 c) const;

  /**
   * Check if a character @a c is a valid CharRef character.
   *
   * @retval 1 if @a c is a valid character reference character, 0 otherwise.
   */
  int isCharRef (const ACEXML_Char c) const;

  /**
   * Check if a character @a c is a BaseChar.
   *
   * @retval 1 if @a c is a valid BaseChar character, 0 otherwise.
   */
  int isBasechar (const ACEXML_Char c) const;

  /**
   * Check if a character @a c is a Ideographic.
   *
   * @retval 1 if @a c is a valid Ideographic character, 0 otherwise.
   */
  int isIdeographic (const ACEXML_Char c) const;

  /**
   * Check if a character @a c is a CombiningChar.
   *
   * @retval 1 if @a c is a valid CombiningChar character, 0 otherwise.
   */
  int isCombiningchar (const ACEXML_Char c) const;

  /**
   * Check if a character @a c is a Digit.
   *
   * @retval 1 if @a c is a valid Digit character, 0 otherwise.
   */
  int isDigit (const ACEXML_Char c) const;

  /**
   * Check if a character @a c is an Extender.
   *
   * @retval 1 if @a c is a valid Extender character, 0 otherwise.
   */
  int isExtender (const ACEXML_Char c) const;

  /**
   * Check if a character @a c is a Letter.
   *
   * @retval 1 if @a c is a valid Letter character, 0 otherwise.
   */
  int isLetter (const ACEXML_Char c) const;

  /**
   * Check if a character is an acceptable NameChar.
   *
   * @retval 1 if @a c is a valid NameChar character, 0 otherwise.
   */
  int isNameChar (const ACEXML_Char c) const;

  /**
   * Check if a character is a PubidChar.
   *
   * @retval 1 if @a c is a valid PubidChar character, 0 otherwise.
   */
  int isPubidChar (const ACEXML_Char c) const;

  /// Get a character.
  virtual ACEXML_Char get (void);

  /// Peek a character.
  virtual ACEXML_Char peek (void);

private:

  // *** Helper functions for parsing XML

  /**
   * Skip any whitespaces encountered until the first non-whitespace
   * character is encountered.
   *
   * @return The next non-whitespace character from the CharStream.
   *
   * @sa skip_whitespace_count
   */
  ACEXML_Char skip_whitespace (void);

  /**
   * Skip any whitespaces encountered until the first non-whitespace
   * character.  The first non-whitespace character is not consumed.
   * This method does peek into the input CharStream and therefore
   * is more expensive than @ref skip_whitespace.
   *
   * @param peek If non-null, @a peek points to a ACEXML_Char where
   *        skip_whitespace_count stores the first non-whitespace
   *        character it sees (character is not removed from the stream.)
   *
   * @return The number of whitespace characters consumed.
   *
   * @sa skip_whitespace
   */
  int skip_whitespace_count (ACEXML_Char *peek = 0);

  /**
   * Skip an equal sign.
   *
   * @retval 0 when succeeds, -1 if no equal sign is found.
   */
  int skip_equal (void);

  /**
   * Get a quoted string.  Quoted strings are used to specify
   * attribute values and this routine will replace character and
   * entity references on-the-fly.  Parameter entities are not allowed
   * (or replaced) in this function.  (But regular entities are.)
   *
   * @param str returns the un-quoted string.
   *
   * @retval 0 on success, -1 otherwise.
   */
  int get_quoted_string (ACEXML_Char *&str);

  /**
   * Check if a character @a c is a Digit.
   *
   * @retval 1 if @a c is a valid Digit character, 0 otherwise.
   */
  int isNormalDigit (const ACEXML_Char c) const;

  /**
   * Dispatch errors to ErrorHandler.
   *
   */
  void error (const ACEXML_Char* msg ACEXML_ENV_ARG_DECL)
    ACE_THROW_SPEC ((ACEXML_SAXException));

  /**
   * Dispatch warnings to ErrorHandler.
   *
   */
  void warning (const ACEXML_Char* msg ACEXML_ENV_ARG_DECL)
    ACE_THROW_SPEC ((ACEXML_SAXException));

  /**
   * Dispatch fatal errors to ErrorHandler.
   *
   */
  void fatal_error (const ACEXML_Char* msg ACEXML_ENV_ARG_DECL)
    ACE_THROW_SPEC ((ACEXML_SAXException));

  /**
   * Dispatch prefix mapping calls to the ContentHandler.
   *
   * @param prefix Namespace prefix
   * @param uri Namespace URI
   * @param name Local name
   * @param start 1 => startPrefixMapping 0 => endPrefixMapping
   */
  void prefix_mapping (const ACEXML_Char* prefix,
                       const ACEXML_Char* uri,
                       int start ACEXML_ENV_ARG_DECL)
    ACE_THROW_SPEC ((ACEXML_SAXException));
  /**
   *  Parse a keyword.
   */
  int parse_token (const ACEXML_Char* keyword);

  /**
   *  Push the current context on to the stack.
   *
   */
  int push_context (ACEXML_Parser_Context* context);

  /**
   *  Pop the top element in the stack and replace current context with that.
   */
  int pop_context (int GE_ref ACEXML_ENV_ARG_DECL);

  /**
   *  Create a new ACEXML_CharStream from @a systemId and @a publicId and
   *  replace the current input stream with the newly created stream.
   */
  virtual int switch_input (ACEXML_CharStream* cstream,
                            const ACEXML_Char* systemId,
                            const ACEXML_Char* publicId = 0);
  /**
   *  Create a new ACEXML_InputSource from @a systemId and @a publicId and
   *  replace the current input source with the newly created InputSource.
   */
  virtual int switch_input (ACEXML_InputSource* input,
                            const ACEXML_Char* systemId,
                            const ACEXML_Char* publicId = 0);

  /**
   * Check for a parameter entity reference. This is used to check for the
   * occurence of a PE Reference withing markupDecl. Additionally this
   * function consumes any leading or trailing whitespace around the PE
   * Reference.
   *
   * @retval Number of whitespace characters skipped.
   */
  int check_for_PE_reference (ACEXML_ENV_SINGLE_ARG_DECL);

  /**
   *  Reset the parser state.
   *
   */
  void reset (void);

  /**
   * Very trivial, non-conformant normalization of a systemid.
   *
   */
  ACEXML_Char* normalize_systemid (const ACEXML_Char* systemId);

  // Feature names:

  /**
   * \addtogroup acexml_parser_features
   * @{
   */

  /**
   * @var simple_parsing_feature_
   *
   * This constant string defines the name of "simple XML parsing"
   * feature.  When this feature is enabled, ACEXML parser is allowed
   * to parse a simple XML stream without mandated XML prolog
   * and no DTD defintion.
   */
  static const ACEXML_Char simple_parsing_feature_[];

  /**
   * @var namespaces_feature_
   *
   * This constant string defines the SAX XML Namespace feature. When this
   * feature is enabled, ACEXML parser allows access by namespace qualified
   * names.
   */
  static const ACEXML_Char namespaces_feature_[];

  /**
   *  @var namespace_prefixes_feature_
   *
   *  This constant string defines the SAX XML Namespace prefixes feature.
   *  Normally the list of attributes returned by the parser will not
   *  contain attributes used as namespace declarations (xmlns*). When this
   *  feature is enabled, the list of attributes contains the namespace
   *  declarations also.
   */
  static const ACEXML_Char namespace_prefixes_feature_[];

  /**
   *  @var validation_feature_
   *
   *  This constant string defines the SAX XML Validation feature. When
   *  this feature is enabled, the parser validates the document in
   *  addition to checking for well-formedness.
   */
  static const ACEXML_Char validation_feature_[];

  /* @} */

  /// Keeping track of the handlers. We do not manage the memory for
  /// handlers.
  ACEXML_DTDHandler *dtd_handler_;
  ACEXML_EntityResolver *entity_resolver_;
  ACEXML_ContentHandler *content_handler_;
  ACEXML_ErrorHandler *error_handler_;

  /// Document Type
  ACEXML_Char *doctype_;

  /// Current parser context
  ACEXML_Parser_Context* current_;

  /// Stack used to hold the Parser_Context
  ACE_Unbounded_Stack<ACEXML_Parser_Context*> ctx_stack_;

  /*
   * The following two are essentially chains of references and is used by
   * the parser to determine if there is any recursion. We keep two of
   * these one for general entities and one for parameter entities, as they
   * both fall under different namespaces.
   *
   */
  /// Set used to hold the general entity references that are active.
  ACE_Unbounded_Stack<ACEXML_Char*> GE_reference_;

  /// Set used to hold the parameter entity references that are active.
  ACE_Unbounded_Stack<ACEXML_Char*> PE_reference_;

  /// Obstack used by the parser to hold all the strings parsed
  ACE_Obstack_T<ACEXML_Char> obstack_;

  /// Alternative obstack used to hold any strings when the original is in use
  ACE_Obstack_T<ACEXML_Char> alt_stack_;

  /// Namespace stack used by the parser to implement support for Namespaces
  ACEXML_NamespaceSupport xml_namespace_;

  /// T => We are processing a nested namespace
  int nested_namespace_;

  /// Set of internal parsed general entities in the document
  ACEXML_Entity_Manager internal_GE_;

  /// Set of external parsed general entities in the document
  ACEXML_Entity_Manager external_GE_;

  /// Set of unparsed entities in the document
  ACEXML_Entity_Manager unparsed_entities_;

  /// Set of predefined entities used by the parser
  ACEXML_Entity_Manager predef_entities_;

  /// Set of internal parsed parameter entities in the document
  ACEXML_Entity_Manager internal_PE_;

  /// Set of external parsed parameter entities in the document
  ACEXML_Entity_Manager external_PE_;

  /// Set of notations declared in the document
  ACEXML_Entity_Manager notations_;

  /// State of the parser when it encounters a reference.
  ACEXML_ParserInt::ReferenceState ref_state_;

  /// T => We are parsing an external subset
  int external_subset_;

  /// T => We are parsing an external entity value
  int external_entity_;

  /// T => Internal DTD has parameter entity references
  int has_pe_refs_;

  /// If set, the document is a standalone XML document
  int standalone_;

  /// If set, the document has an external DTD subset
  int external_dtd_;

  /// If set, the document has an internal DTD
  int internal_dtd_;

  /// Feature flags
  /// If set, the parser should parse a document without a prolog
  int simple_parsing_;

  /// If set, the parser should also validate
  int validate_;

  /// If set, the parser should allow access by namespace qualified names.
  int namespaces_;

  /// If set, the parser should include namespace declarations in the list
  /// of attributes of an element.
  int namespace_prefixes_;

};

#if defined (__ACEXML_INLINE__)
# include "ACEXML/parser/parser/Parser.i"
#endif /* __ACEXML_INLINE__ */

#include /**/ "ace/post.h"

#endif /* _ACEXML_BASIC_PARSER_H_ */