http://git-wip-us.apache.org/repos/asf/hadoop/blob/90c7003c/hadoop-hdfs-project/hadoop-hdfs-native-client/src/main/native/libhdfspp/third_party/rapidxml-1.13/rapidxml/rapidxml.hpp ---------------------------------------------------------------------- diff --git a/hadoop-hdfs-project/hadoop-hdfs-native-client/src/main/native/libhdfspp/third_party/rapidxml-1.13/rapidxml/rapidxml.hpp b/hadoop-hdfs-project/hadoop-hdfs-native-client/src/main/native/libhdfspp/third_party/rapidxml-1.13/rapidxml/rapidxml.hpp new file mode 100644 index 0000000..460238b --- /dev/null +++ b/hadoop-hdfs-project/hadoop-hdfs-native-client/src/main/native/libhdfspp/third_party/rapidxml-1.13/rapidxml/rapidxml.hpp @@ -0,0 +1,2596 @@ +#ifndef RAPIDXML_HPP_INCLUDED +#define RAPIDXML_HPP_INCLUDED + +// Copyright (C) 2006, 2009 Marcin Kalicinski +// Version 1.13 +// Revision $DateTime: 2009/05/13 01:46:17 $ +//! \file rapidxml.hpp This file contains rapidxml parser and DOM implementation + +// If standard library is disabled, user must provide implementations of required functions and typedefs +#if !defined(RAPIDXML_NO_STDLIB) + #include <cstdlib> // For std::size_t + #include <cassert> // For assert + #include <new> // For placement new +#endif + +// On MSVC, disable "conditional expression is constant" warning (level 4). +// This warning is almost impossible to avoid with certain types of templated code +#ifdef _MSC_VER + #pragma warning(push) + #pragma warning(disable:4127) // Conditional expression is constant +#endif + +/////////////////////////////////////////////////////////////////////////// +// RAPIDXML_PARSE_ERROR + +#if defined(RAPIDXML_NO_EXCEPTIONS) + +#define RAPIDXML_PARSE_ERROR(what, where) { parse_error_handler(what, where); assert(0); } + +namespace rapidxml +{ + //! When exceptions are disabled by defining RAPIDXML_NO_EXCEPTIONS, + //! this function is called to notify user about the error. + //! It must be defined by the user. + //! <br><br> + //! This function cannot return. If it does, the results are undefined. + //! <br><br> + //! A very simple definition might look like that: + //! <pre> + //! void %rapidxml::%parse_error_handler(const char *what, void *where) + //! { + //! std::cout << "Parse error: " << what << "\n"; + //! std::abort(); + //! } + //! </pre> + //! \param what Human readable description of the error. + //! \param where Pointer to character data where error was detected. + void parse_error_handler(const char *what, void *where); +} + +#else + +#include <exception> // For std::exception + +#define RAPIDXML_PARSE_ERROR(what, where) throw parse_error(what, where) + +namespace rapidxml +{ + + //! Parse error exception. + //! This exception is thrown by the parser when an error occurs. + //! Use what() function to get human-readable error message. + //! Use where() function to get a pointer to position within source text where error was detected. + //! <br><br> + //! If throwing exceptions by the parser is undesirable, + //! it can be disabled by defining RAPIDXML_NO_EXCEPTIONS macro before rapidxml.hpp is included. + //! This will cause the parser to call rapidxml::parse_error_handler() function instead of throwing an exception. + //! This function must be defined by the user. + //! <br><br> + //! This class derives from <code>std::exception</code> class. + class parse_error: public std::exception + { + + public: + + //! Constructs parse error + parse_error(const char *what, void *where) + : m_what(what) + , m_where(where) + { + } + + //! Gets human readable description of error. + //! \return Pointer to null terminated description of the error. + virtual const char *what() const throw() + { + return m_what; + } + + //! Gets pointer to character data where error happened. + //! Ch should be the same as char type of xml_document that produced the error. + //! \return Pointer to location within the parsed string where error occured. + template<class Ch> + Ch *where() const + { + return reinterpret_cast<Ch *>(m_where); + } + + private: + + const char *m_what; + void *m_where; + + }; +} + +#endif + +/////////////////////////////////////////////////////////////////////////// +// Pool sizes + +#ifndef RAPIDXML_STATIC_POOL_SIZE + // Size of static memory block of memory_pool. + // Define RAPIDXML_STATIC_POOL_SIZE before including rapidxml.hpp if you want to override the default value. + // No dynamic memory allocations are performed by memory_pool until static memory is exhausted. + #define RAPIDXML_STATIC_POOL_SIZE (64 * 1024) +#endif + +#ifndef RAPIDXML_DYNAMIC_POOL_SIZE + // Size of dynamic memory block of memory_pool. + // Define RAPIDXML_DYNAMIC_POOL_SIZE before including rapidxml.hpp if you want to override the default value. + // After the static block is exhausted, dynamic blocks with approximately this size are allocated by memory_pool. + #define RAPIDXML_DYNAMIC_POOL_SIZE (64 * 1024) +#endif + +#ifndef RAPIDXML_ALIGNMENT + // Memory allocation alignment. + // Define RAPIDXML_ALIGNMENT before including rapidxml.hpp if you want to override the default value, which is the size of pointer. + // All memory allocations for nodes, attributes and strings will be aligned to this value. + // This must be a power of 2 and at least 1, otherwise memory_pool will not work. + #define RAPIDXML_ALIGNMENT sizeof(void *) +#endif + +namespace rapidxml +{ + // Forward declarations + template<class Ch> class xml_node; + template<class Ch> class xml_attribute; + template<class Ch> class xml_document; + + //! Enumeration listing all node types produced by the parser. + //! Use xml_node::type() function to query node type. + enum node_type + { + node_document, //!< A document node. Name and value are empty. + node_element, //!< An element node. Name contains element name. Value contains text of first data node. + node_data, //!< A data node. Name is empty. Value contains data text. + node_cdata, //!< A CDATA node. Name is empty. Value contains data text. + node_comment, //!< A comment node. Name is empty. Value contains comment text. + node_declaration, //!< A declaration node. Name and value are empty. Declaration parameters (version, encoding and standalone) are in node attributes. + node_doctype, //!< A DOCTYPE node. Name is empty. Value contains DOCTYPE text. + node_pi //!< A PI node. Name contains target. Value contains instructions. + }; + + /////////////////////////////////////////////////////////////////////// + // Parsing flags + + //! Parse flag instructing the parser to not create data nodes. + //! Text of first data node will still be placed in value of parent element, unless rapidxml::parse_no_element_values flag is also specified. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_no_data_nodes = 0x1; + + //! Parse flag instructing the parser to not use text of first data node as a value of parent element. + //! Can be combined with other flags by use of | operator. + //! Note that child data nodes of element node take precendence over its value when printing. + //! That is, if element has one or more child data nodes <em>and</em> a value, the value will be ignored. + //! Use rapidxml::parse_no_data_nodes flag to prevent creation of data nodes if you want to manipulate data using values of elements. + //! <br><br> + //! See xml_document::parse() function. + const int parse_no_element_values = 0x2; + + //! Parse flag instructing the parser to not place zero terminators after strings in the source text. + //! By default zero terminators are placed, modifying source text. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_no_string_terminators = 0x4; + + //! Parse flag instructing the parser to not translate entities in the source text. + //! By default entities are translated, modifying source text. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_no_entity_translation = 0x8; + + //! Parse flag instructing the parser to disable UTF-8 handling and assume plain 8 bit characters. + //! By default, UTF-8 handling is enabled. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_no_utf8 = 0x10; + + //! Parse flag instructing the parser to create XML declaration node. + //! By default, declaration node is not created. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_declaration_node = 0x20; + + //! Parse flag instructing the parser to create comments nodes. + //! By default, comment nodes are not created. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_comment_nodes = 0x40; + + //! Parse flag instructing the parser to create DOCTYPE node. + //! By default, doctype node is not created. + //! Although W3C specification allows at most one DOCTYPE node, RapidXml will silently accept documents with more than one. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_doctype_node = 0x80; + + //! Parse flag instructing the parser to create PI nodes. + //! By default, PI nodes are not created. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_pi_nodes = 0x100; + + //! Parse flag instructing the parser to validate closing tag names. + //! If not set, name inside closing tag is irrelevant to the parser. + //! By default, closing tags are not validated. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_validate_closing_tags = 0x200; + + //! Parse flag instructing the parser to trim all leading and trailing whitespace of data nodes. + //! By default, whitespace is not trimmed. + //! This flag does not cause the parser to modify source text. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_trim_whitespace = 0x400; + + //! Parse flag instructing the parser to condense all whitespace runs of data nodes to a single space character. + //! Trimming of leading and trailing whitespace of data is controlled by rapidxml::parse_trim_whitespace flag. + //! By default, whitespace is not normalized. + //! If this flag is specified, source text will be modified. + //! Can be combined with other flags by use of | operator. + //! <br><br> + //! See xml_document::parse() function. + const int parse_normalize_whitespace = 0x800; + + // Compound flags + + //! Parse flags which represent default behaviour of the parser. + //! This is always equal to 0, so that all other flags can be simply ored together. + //! Normally there is no need to inconveniently disable flags by anding with their negated (~) values. + //! This also means that meaning of each flag is a <i>negation</i> of the default setting. + //! For example, if flag name is rapidxml::parse_no_utf8, it means that utf-8 is <i>enabled</i> by default, + //! and using the flag will disable it. + //! <br><br> + //! See xml_document::parse() function. + const int parse_default = 0; + + //! A combination of parse flags that forbids any modifications of the source text. + //! This also results in faster parsing. However, note that the following will occur: + //! <ul> + //! <li>names and values of nodes will not be zero terminated, you have to use xml_base::name_size() and xml_base::value_size() functions to determine where name and value ends</li> + //! <li>entities will not be translated</li> + //! <li>whitespace will not be normalized</li> + //! </ul> + //! See xml_document::parse() function. + const int parse_non_destructive = parse_no_string_terminators | parse_no_entity_translation; + + //! A combination of parse flags resulting in fastest possible parsing, without sacrificing important data. + //! <br><br> + //! See xml_document::parse() function. + const int parse_fastest = parse_non_destructive | parse_no_data_nodes; + + //! A combination of parse flags resulting in largest amount of data being extracted. + //! This usually results in slowest parsing. + //! <br><br> + //! See xml_document::parse() function. + const int parse_full = parse_declaration_node | parse_comment_nodes | parse_doctype_node | parse_pi_nodes | parse_validate_closing_tags; + + /////////////////////////////////////////////////////////////////////// + // Internals + + //! \cond internal + namespace internal + { + + // Struct that contains lookup tables for the parser + // It must be a template to allow correct linking (because it has static data members, which are defined in a header file). + template<int Dummy> + struct lookup_tables + { + static const unsigned char lookup_whitespace[256]; // Whitespace table + static const unsigned char lookup_node_name[256]; // Node name table + static const unsigned char lookup_text[256]; // Text table + static const unsigned char lookup_text_pure_no_ws[256]; // Text table + static const unsigned char lookup_text_pure_with_ws[256]; // Text table + static const unsigned char lookup_attribute_name[256]; // Attribute name table + static const unsigned char lookup_attribute_data_1[256]; // Attribute data table with single quote + static const unsigned char lookup_attribute_data_1_pure[256]; // Attribute data table with single quote + static const unsigned char lookup_attribute_data_2[256]; // Attribute data table with double quotes + static const unsigned char lookup_attribute_data_2_pure[256]; // Attribute data table with double quotes + static const unsigned char lookup_digits[256]; // Digits + static const unsigned char lookup_upcase[256]; // To uppercase conversion table for ASCII characters + }; + + // Find length of the string + template<class Ch> + inline std::size_t measure(const Ch *p) + { + const Ch *tmp = p; + while (*tmp) + ++tmp; + return tmp - p; + } + + // Compare strings for equality + template<class Ch> + inline bool compare(const Ch *p1, std::size_t size1, const Ch *p2, std::size_t size2, bool case_sensitive) + { + if (size1 != size2) + return false; + if (case_sensitive) + { + for (const Ch *end = p1 + size1; p1 < end; ++p1, ++p2) + if (*p1 != *p2) + return false; + } + else + { + for (const Ch *end = p1 + size1; p1 < end; ++p1, ++p2) + if (lookup_tables<0>::lookup_upcase[static_cast<unsigned char>(*p1)] != lookup_tables<0>::lookup_upcase[static_cast<unsigned char>(*p2)]) + return false; + } + return true; + } + } + //! \endcond + + /////////////////////////////////////////////////////////////////////// + // Memory pool + + //! This class is used by the parser to create new nodes and attributes, without overheads of dynamic memory allocation. + //! In most cases, you will not need to use this class directly. + //! However, if you need to create nodes manually or modify names/values of nodes, + //! you are encouraged to use memory_pool of relevant xml_document to allocate the memory. + //! Not only is this faster than allocating them by using <code>new</code> operator, + //! but also their lifetime will be tied to the lifetime of document, + //! possibly simplyfing memory management. + //! <br><br> + //! Call allocate_node() or allocate_attribute() functions to obtain new nodes or attributes from the pool. + //! You can also call allocate_string() function to allocate strings. + //! Such strings can then be used as names or values of nodes without worrying about their lifetime. + //! Note that there is no <code>free()</code> function -- all allocations are freed at once when clear() function is called, + //! or when the pool is destroyed. + //! <br><br> + //! It is also possible to create a standalone memory_pool, and use it + //! to allocate nodes, whose lifetime will not be tied to any document. + //! <br><br> + //! Pool maintains <code>RAPIDXML_STATIC_POOL_SIZE</code> bytes of statically allocated memory. + //! Until static memory is exhausted, no dynamic memory allocations are done. + //! When static memory is exhausted, pool allocates additional blocks of memory of size <code>RAPIDXML_DYNAMIC_POOL_SIZE</code> each, + //! by using global <code>new[]</code> and <code>delete[]</code> operators. + //! This behaviour can be changed by setting custom allocation routines. + //! Use set_allocator() function to set them. + //! <br><br> + //! Allocations for nodes, attributes and strings are aligned at <code>RAPIDXML_ALIGNMENT</code> bytes. + //! This value defaults to the size of pointer on target architecture. + //! <br><br> + //! To obtain absolutely top performance from the parser, + //! it is important that all nodes are allocated from a single, contiguous block of memory. + //! Otherwise, cache misses when jumping between two (or more) disjoint blocks of memory can slow down parsing quite considerably. + //! If required, you can tweak <code>RAPIDXML_STATIC_POOL_SIZE</code>, <code>RAPIDXML_DYNAMIC_POOL_SIZE</code> and <code>RAPIDXML_ALIGNMENT</code> + //! to obtain best wasted memory to performance compromise. + //! To do it, define their values before rapidxml.hpp file is included. + //! \param Ch Character type of created nodes. + template<class Ch = char> + class memory_pool + { + + public: + + //! \cond internal + typedef void *(alloc_func)(std::size_t); // Type of user-defined function used to allocate memory + typedef void (free_func)(void *); // Type of user-defined function used to free memory + //! \endcond + + //! Constructs empty pool with default allocator functions. + memory_pool() + : m_alloc_func(0) + , m_free_func(0) + { + init(); + } + + //! Destroys pool and frees all the memory. + //! This causes memory occupied by nodes allocated by the pool to be freed. + //! Nodes allocated from the pool are no longer valid. + ~memory_pool() + { + clear(); + } + + //! Allocates a new node from the pool, and optionally assigns name and value to it. + //! If the allocation request cannot be accomodated, this function will throw <code>std::bad_alloc</code>. + //! If exceptions are disabled by defining RAPIDXML_NO_EXCEPTIONS, this function + //! will call rapidxml::parse_error_handler() function. + //! \param type Type of node to create. + //! \param name Name to assign to the node, or 0 to assign no name. + //! \param value Value to assign to the node, or 0 to assign no value. + //! \param name_size Size of name to assign, or 0 to automatically calculate size from name string. + //! \param value_size Size of value to assign, or 0 to automatically calculate size from value string. + //! \return Pointer to allocated node. This pointer will never be NULL. + xml_node<Ch> *allocate_node(node_type type, + const Ch *name = 0, const Ch *value = 0, + std::size_t name_size = 0, std::size_t value_size = 0) + { + void *memory = allocate_aligned(sizeof(xml_node<Ch>)); + xml_node<Ch> *node = new(memory) xml_node<Ch>(type); + if (name) + { + if (name_size > 0) + node->name(name, name_size); + else + node->name(name); + } + if (value) + { + if (value_size > 0) + node->value(value, value_size); + else + node->value(value); + } + return node; + } + + //! Allocates a new attribute from the pool, and optionally assigns name and value to it. + //! If the allocation request cannot be accomodated, this function will throw <code>std::bad_alloc</code>. + //! If exceptions are disabled by defining RAPIDXML_NO_EXCEPTIONS, this function + //! will call rapidxml::parse_error_handler() function. + //! \param name Name to assign to the attribute, or 0 to assign no name. + //! \param value Value to assign to the attribute, or 0 to assign no value. + //! \param name_size Size of name to assign, or 0 to automatically calculate size from name string. + //! \param value_size Size of value to assign, or 0 to automatically calculate size from value string. + //! \return Pointer to allocated attribute. This pointer will never be NULL. + xml_attribute<Ch> *allocate_attribute(const Ch *name = 0, const Ch *value = 0, + std::size_t name_size = 0, std::size_t value_size = 0) + { + void *memory = allocate_aligned(sizeof(xml_attribute<Ch>)); + xml_attribute<Ch> *attribute = new(memory) xml_attribute<Ch>; + if (name) + { + if (name_size > 0) + attribute->name(name, name_size); + else + attribute->name(name); + } + if (value) + { + if (value_size > 0) + attribute->value(value, value_size); + else + attribute->value(value); + } + return attribute; + } + + //! Allocates a char array of given size from the pool, and optionally copies a given string to it. + //! If the allocation request cannot be accomodated, this function will throw <code>std::bad_alloc</code>. + //! If exceptions are disabled by defining RAPIDXML_NO_EXCEPTIONS, this function + //! will call rapidxml::parse_error_handler() function. + //! \param source String to initialize the allocated memory with, or 0 to not initialize it. + //! \param size Number of characters to allocate, or zero to calculate it automatically from source string length; if size is 0, source string must be specified and null terminated. + //! \return Pointer to allocated char array. This pointer will never be NULL. + Ch *allocate_string(const Ch *source = 0, std::size_t size = 0) + { + assert(source || size); // Either source or size (or both) must be specified + if (size == 0) + size = internal::measure(source) + 1; + Ch *result = static_cast<Ch *>(allocate_aligned(size * sizeof(Ch))); + if (source) + for (std::size_t i = 0; i < size; ++i) + result[i] = source[i]; + return result; + } + + //! Clones an xml_node and its hierarchy of child nodes and attributes. + //! Nodes and attributes are allocated from this memory pool. + //! Names and values are not cloned, they are shared between the clone and the source. + //! Result node can be optionally specified as a second parameter, + //! in which case its contents will be replaced with cloned source node. + //! This is useful when you want to clone entire document. + //! \param source Node to clone. + //! \param result Node to put results in, or 0 to automatically allocate result node + //! \return Pointer to cloned node. This pointer will never be NULL. + xml_node<Ch> *clone_node(const xml_node<Ch> *source, xml_node<Ch> *result = 0) + { + // Prepare result node + if (result) + { + result->remove_all_attributes(); + result->remove_all_nodes(); + result->type(source->type()); + } + else + result = allocate_node(source->type()); + + // Clone name and value + result->name(source->name(), source->name_size()); + result->value(source->value(), source->value_size()); + + // Clone child nodes and attributes + for (xml_node<Ch> *child = source->first_node(); child; child = child->next_sibling()) + result->append_node(clone_node(child)); + for (xml_attribute<Ch> *attr = source->first_attribute(); attr; attr = attr->next_attribute()) + result->append_attribute(allocate_attribute(attr->name(), attr->value(), attr->name_size(), attr->value_size())); + + return result; + } + + //! Clears the pool. + //! This causes memory occupied by nodes allocated by the pool to be freed. + //! Any nodes or strings allocated from the pool will no longer be valid. + void clear() + { + while (m_begin != m_static_memory) + { + char *previous_begin = reinterpret_cast<header *>(align(m_begin))->previous_begin; + if (m_free_func) + m_free_func(m_begin); + else + delete[] m_begin; + m_begin = previous_begin; + } + init(); + } + + //! Sets or resets the user-defined memory allocation functions for the pool. + //! This can only be called when no memory is allocated from the pool yet, otherwise results are undefined. + //! Allocation function must not return invalid pointer on failure. It should either throw, + //! stop the program, or use <code>longjmp()</code> function to pass control to other place of program. + //! If it returns invalid pointer, results are undefined. + //! <br><br> + //! User defined allocation functions must have the following forms: + //! <br><code> + //! <br>void *allocate(std::size_t size); + //! <br>void free(void *pointer); + //! </code><br> + //! \param af Allocation function, or 0 to restore default function + //! \param ff Free function, or 0 to restore default function + void set_allocator(alloc_func *af, free_func *ff) + { + assert(m_begin == m_static_memory && m_ptr == align(m_begin)); // Verify that no memory is allocated yet + m_alloc_func = af; + m_free_func = ff; + } + + private: + + struct header + { + char *previous_begin; + }; + + void init() + { + m_begin = m_static_memory; + m_ptr = align(m_begin); + m_end = m_static_memory + sizeof(m_static_memory); + } + + char *align(char *ptr) + { + std::size_t alignment = ((RAPIDXML_ALIGNMENT - (std::size_t(ptr) & (RAPIDXML_ALIGNMENT - 1))) & (RAPIDXML_ALIGNMENT - 1)); + return ptr + alignment; + } + + char *allocate_raw(std::size_t size) + { + // Allocate + void *memory; + if (m_alloc_func) // Allocate memory using either user-specified allocation function or global operator new[] + { + memory = m_alloc_func(size); + assert(memory); // Allocator is not allowed to return 0, on failure it must either throw, stop the program or use longjmp + } + else + { + memory = new char[size]; +#ifdef RAPIDXML_NO_EXCEPTIONS + if (!memory) // If exceptions are disabled, verify memory allocation, because new will not be able to throw bad_alloc + RAPIDXML_PARSE_ERROR("out of memory", 0); +#endif + } + return static_cast<char *>(memory); + } + + void *allocate_aligned(std::size_t size) + { + // Calculate aligned pointer + char *result = align(m_ptr); + + // If not enough memory left in current pool, allocate a new pool + if (result + size > m_end) + { + // Calculate required pool size (may be bigger than RAPIDXML_DYNAMIC_POOL_SIZE) + std::size_t pool_size = RAPIDXML_DYNAMIC_POOL_SIZE; + if (pool_size < size) + pool_size = size; + + // Allocate + std::size_t alloc_size = sizeof(header) + (2 * RAPIDXML_ALIGNMENT - 2) + pool_size; // 2 alignments required in worst case: one for header, one for actual allocation + char *raw_memory = allocate_raw(alloc_size); + + // Setup new pool in allocated memory + char *pool = align(raw_memory); + header *new_header = reinterpret_cast<header *>(pool); + new_header->previous_begin = m_begin; + m_begin = raw_memory; + m_ptr = pool + sizeof(header); + m_end = raw_memory + alloc_size; + + // Calculate aligned pointer again using new pool + result = align(m_ptr); + } + + // Update pool and return aligned pointer + m_ptr = result + size; + return result; + } + + char *m_begin; // Start of raw memory making up current pool + char *m_ptr; // First free byte in current pool + char *m_end; // One past last available byte in current pool + char m_static_memory[RAPIDXML_STATIC_POOL_SIZE]; // Static raw memory + alloc_func *m_alloc_func; // Allocator function, or 0 if default is to be used + free_func *m_free_func; // Free function, or 0 if default is to be used + }; + + /////////////////////////////////////////////////////////////////////////// + // XML base + + //! Base class for xml_node and xml_attribute implementing common functions: + //! name(), name_size(), value(), value_size() and parent(). + //! \param Ch Character type to use + template<class Ch = char> + class xml_base + { + + public: + + /////////////////////////////////////////////////////////////////////////// + // Construction & destruction + + // Construct a base with empty name, value and parent + xml_base() + : m_name(0) + , m_value(0) + , m_parent(0) + { + } + + /////////////////////////////////////////////////////////////////////////// + // Node data access + + //! Gets name of the node. + //! Interpretation of name depends on type of node. + //! Note that name will not be zero-terminated if rapidxml::parse_no_string_terminators option was selected during parse. + //! <br><br> + //! Use name_size() function to determine length of the name. + //! \return Name of node, or empty string if node has no name. + Ch *name() const + { + return m_name ? m_name : nullstr(); + } + + //! Gets size of node name, not including terminator character. + //! This function works correctly irrespective of whether name is or is not zero terminated. + //! \return Size of node name, in characters. + std::size_t name_size() const + { + return m_name ? m_name_size : 0; + } + + //! Gets value of node. + //! Interpretation of value depends on type of node. + //! Note that value will not be zero-terminated if rapidxml::parse_no_string_terminators option was selected during parse. + //! <br><br> + //! Use value_size() function to determine length of the value. + //! \return Value of node, or empty string if node has no value. + Ch *value() const + { + return m_value ? m_value : nullstr(); + } + + //! Gets size of node value, not including terminator character. + //! This function works correctly irrespective of whether value is or is not zero terminated. + //! \return Size of node value, in characters. + std::size_t value_size() const + { + return m_value ? m_value_size : 0; + } + + /////////////////////////////////////////////////////////////////////////// + // Node modification + + //! Sets name of node to a non zero-terminated string. + //! See \ref ownership_of_strings. + //! <br><br> + //! Note that node does not own its name or value, it only stores a pointer to it. + //! It will not delete or otherwise free the pointer on destruction. + //! It is reponsibility of the user to properly manage lifetime of the string. + //! The easiest way to achieve it is to use memory_pool of the document to allocate the string - + //! on destruction of the document the string will be automatically freed. + //! <br><br> + //! Size of name must be specified separately, because name does not have to be zero terminated. + //! Use name(const Ch *) function to have the length automatically calculated (string must be zero terminated). + //! \param name Name of node to set. Does not have to be zero terminated. + //! \param size Size of name, in characters. This does not include zero terminator, if one is present. + void name(const Ch *name, std::size_t size) + { + m_name = const_cast<Ch *>(name); + m_name_size = size; + } + + //! Sets name of node to a zero-terminated string. + //! See also \ref ownership_of_strings and xml_node::name(const Ch *, std::size_t). + //! \param name Name of node to set. Must be zero terminated. + void name(const Ch *name) + { + this->name(name, internal::measure(name)); + } + + //! Sets value of node to a non zero-terminated string. + //! See \ref ownership_of_strings. + //! <br><br> + //! Note that node does not own its name or value, it only stores a pointer to it. + //! It will not delete or otherwise free the pointer on destruction. + //! It is reponsibility of the user to properly manage lifetime of the string. + //! The easiest way to achieve it is to use memory_pool of the document to allocate the string - + //! on destruction of the document the string will be automatically freed. + //! <br><br> + //! Size of value must be specified separately, because it does not have to be zero terminated. + //! Use value(const Ch *) function to have the length automatically calculated (string must be zero terminated). + //! <br><br> + //! If an element has a child node of type node_data, it will take precedence over element value when printing. + //! If you want to manipulate data of elements using values, use parser flag rapidxml::parse_no_data_nodes to prevent creation of data nodes by the parser. + //! \param value value of node to set. Does not have to be zero terminated. + //! \param size Size of value, in characters. This does not include zero terminator, if one is present. + void value(const Ch *value, std::size_t size) + { + m_value = const_cast<Ch *>(value); + m_value_size = size; + } + + //! Sets value of node to a zero-terminated string. + //! See also \ref ownership_of_strings and xml_node::value(const Ch *, std::size_t). + //! \param value Vame of node to set. Must be zero terminated. + void value(const Ch *value) + { + this->value(value, internal::measure(value)); + } + + /////////////////////////////////////////////////////////////////////////// + // Related nodes access + + //! Gets node parent. + //! \return Pointer to parent node, or 0 if there is no parent. + xml_node<Ch> *parent() const + { + return m_parent; + } + + protected: + + // Return empty string + static Ch *nullstr() + { + static Ch zero = Ch('\0'); + return &zero; + } + + Ch *m_name; // Name of node, or 0 if no name + Ch *m_value; // Value of node, or 0 if no value + std::size_t m_name_size; // Length of node name, or undefined of no name + std::size_t m_value_size; // Length of node value, or undefined if no value + xml_node<Ch> *m_parent; // Pointer to parent node, or 0 if none + + }; + + //! Class representing attribute node of XML document. + //! Each attribute has name and value strings, which are available through name() and value() functions (inherited from xml_base). + //! Note that after parse, both name and value of attribute will point to interior of source text used for parsing. + //! Thus, this text must persist in memory for the lifetime of attribute. + //! \param Ch Character type to use. + template<class Ch = char> + class xml_attribute: public xml_base<Ch> + { + + friend class xml_node<Ch>; + + public: + + /////////////////////////////////////////////////////////////////////////// + // Construction & destruction + + //! Constructs an empty attribute with the specified type. + //! Consider using memory_pool of appropriate xml_document if allocating attributes manually. + xml_attribute() + { + } + + /////////////////////////////////////////////////////////////////////////// + // Related nodes access + + //! Gets document of which attribute is a child. + //! \return Pointer to document that contains this attribute, or 0 if there is no parent document. + xml_document<Ch> *document() const + { + if (xml_node<Ch> *node = this->parent()) + { + while (node->parent()) + node = node->parent(); + return node->type() == node_document ? static_cast<xml_document<Ch> *>(node) : 0; + } + else + return 0; + } + + //! Gets previous attribute, optionally matching attribute name. + //! \param name Name of attribute to find, or 0 to return previous attribute regardless of its name; this string doesn't have to be zero-terminated if name_size is non-zero + //! \param name_size Size of name, in characters, or 0 to have size calculated automatically from string + //! \param case_sensitive Should name comparison be case-sensitive; non case-sensitive comparison works properly only for ASCII characters + //! \return Pointer to found attribute, or 0 if not found. + xml_attribute<Ch> *previous_attribute(const Ch *name = 0, std::size_t name_size = 0, bool case_sensitive = true) const + { + if (name) + { + if (name_size == 0) + name_size = internal::measure(name); + for (xml_attribute<Ch> *attribute = m_prev_attribute; attribute; attribute = attribute->m_prev_attribute) + if (internal::compare(attribute->name(), attribute->name_size(), name, name_size, case_sensitive)) + return attribute; + return 0; + } + else + return this->m_parent ? m_prev_attribute : 0; + } + + //! Gets next attribute, optionally matching attribute name. + //! \param name Name of attribute to find, or 0 to return next attribute regardless of its name; this string doesn't have to be zero-terminated if name_size is non-zero + //! \param name_size Size of name, in characters, or 0 to have size calculated automatically from string + //! \param case_sensitive Should name comparison be case-sensitive; non case-sensitive comparison works properly only for ASCII characters + //! \return Pointer to found attribute, or 0 if not found. + xml_attribute<Ch> *next_attribute(const Ch *name = 0, std::size_t name_size = 0, bool case_sensitive = true) const + { + if (name) + { + if (name_size == 0) + name_size = internal::measure(name); + for (xml_attribute<Ch> *attribute = m_next_attribute; attribute; attribute = attribute->m_next_attribute) + if (internal::compare(attribute->name(), attribute->name_size(), name, name_size, case_sensitive)) + return attribute; + return 0; + } + else + return this->m_parent ? m_next_attribute : 0; + } + + private: + + xml_attribute<Ch> *m_prev_attribute; // Pointer to previous sibling of attribute, or 0 if none; only valid if parent is non-zero + xml_attribute<Ch> *m_next_attribute; // Pointer to next sibling of attribute, or 0 if none; only valid if parent is non-zero + + }; + + /////////////////////////////////////////////////////////////////////////// + // XML node + + //! Class representing a node of XML document. + //! Each node may have associated name and value strings, which are available through name() and value() functions. + //! Interpretation of name and value depends on type of the node. + //! Type of node can be determined by using type() function. + //! <br><br> + //! Note that after parse, both name and value of node, if any, will point interior of source text used for parsing. + //! Thus, this text must persist in the memory for the lifetime of node. + //! \param Ch Character type to use. + template<class Ch = char> + class xml_node: public xml_base<Ch> + { + + public: + + /////////////////////////////////////////////////////////////////////////// + // Construction & destruction + + //! Constructs an empty node with the specified type. + //! Consider using memory_pool of appropriate document to allocate nodes manually. + //! \param type Type of node to construct. + xml_node(node_type type) + : m_type(type) + , m_first_node(0) + , m_first_attribute(0) + { + } + + /////////////////////////////////////////////////////////////////////////// + // Node data access + + //! Gets type of node. + //! \return Type of node. + node_type type() const + { + return m_type; + } + + /////////////////////////////////////////////////////////////////////////// + // Related nodes access + + //! Gets document of which node is a child. + //! \return Pointer to document that contains this node, or 0 if there is no parent document. + xml_document<Ch> *document() const + { + xml_node<Ch> *node = const_cast<xml_node<Ch> *>(this); + while (node->parent()) + node = node->parent(); + return node->type() == node_document ? static_cast<xml_document<Ch> *>(node) : 0; + } + + //! Gets first child node, optionally matching node name. + //! \param name Name of child to find, or 0 to return first child regardless of its name; this string doesn't have to be zero-terminated if name_size is non-zero + //! \param name_size Size of name, in characters, or 0 to have size calculated automatically from string + //! \param case_sensitive Should name comparison be case-sensitive; non case-sensitive comparison works properly only for ASCII characters + //! \return Pointer to found child, or 0 if not found. + xml_node<Ch> *first_node(const Ch *name = 0, std::size_t name_size = 0, bool case_sensitive = true) const + { + if (name) + { + if (name_size == 0) + name_size = internal::measure(name); + for (xml_node<Ch> *child = m_first_node; child; child = child->next_sibling()) + if (internal::compare(child->name(), child->name_size(), name, name_size, case_sensitive)) + return child; + return 0; + } + else + return m_first_node; + } + + //! Gets last child node, optionally matching node name. + //! Behaviour is undefined if node has no children. + //! Use first_node() to test if node has children. + //! \param name Name of child to find, or 0 to return last child regardless of its name; this string doesn't have to be zero-terminated if name_size is non-zero + //! \param name_size Size of name, in characters, or 0 to have size calculated automatically from string + //! \param case_sensitive Should name comparison be case-sensitive; non case-sensitive comparison works properly only for ASCII characters + //! \return Pointer to found child, or 0 if not found. + xml_node<Ch> *last_node(const Ch *name = 0, std::size_t name_size = 0, bool case_sensitive = true) const + { + assert(m_first_node); // Cannot query for last child if node has no children + if (name) + { + if (name_size == 0) + name_size = internal::measure(name); + for (xml_node<Ch> *child = m_last_node; child; child = child->previous_sibling()) + if (internal::compare(child->name(), child->name_size(), name, name_size, case_sensitive)) + return child; + return 0; + } + else + return m_last_node; + } + + //! Gets previous sibling node, optionally matching node name. + //! Behaviour is undefined if node has no parent. + //! Use parent() to test if node has a parent. + //! \param name Name of sibling to find, or 0 to return previous sibling regardless of its name; this string doesn't have to be zero-terminated if name_size is non-zero + //! \param name_size Size of name, in characters, or 0 to have size calculated automatically from string + //! \param case_sensitive Should name comparison be case-sensitive; non case-sensitive comparison works properly only for ASCII characters + //! \return Pointer to found sibling, or 0 if not found. + xml_node<Ch> *previous_sibling(const Ch *name = 0, std::size_t name_size = 0, bool case_sensitive = true) const + { + assert(this->m_parent); // Cannot query for siblings if node has no parent + if (name) + { + if (name_size == 0) + name_size = internal::measure(name); + for (xml_node<Ch> *sibling = m_prev_sibling; sibling; sibling = sibling->m_prev_sibling) + if (internal::compare(sibling->name(), sibling->name_size(), name, name_size, case_sensitive)) + return sibling; + return 0; + } + else + return m_prev_sibling; + } + + //! Gets next sibling node, optionally matching node name. + //! Behaviour is undefined if node has no parent. + //! Use parent() to test if node has a parent. + //! \param name Name of sibling to find, or 0 to return next sibling regardless of its name; this string doesn't have to be zero-terminated if name_size is non-zero + //! \param name_size Size of name, in characters, or 0 to have size calculated automatically from string + //! \param case_sensitive Should name comparison be case-sensitive; non case-sensitive comparison works properly only for ASCII characters + //! \return Pointer to found sibling, or 0 if not found. + xml_node<Ch> *next_sibling(const Ch *name = 0, std::size_t name_size = 0, bool case_sensitive = true) const + { + assert(this->m_parent); // Cannot query for siblings if node has no parent + if (name) + { + if (name_size == 0) + name_size = internal::measure(name); + for (xml_node<Ch> *sibling = m_next_sibling; sibling; sibling = sibling->m_next_sibling) + if (internal::compare(sibling->name(), sibling->name_size(), name, name_size, case_sensitive)) + return sibling; + return 0; + } + else + return m_next_sibling; + } + + //! Gets first attribute of node, optionally matching attribute name. + //! \param name Name of attribute to find, or 0 to return first attribute regardless of its name; this string doesn't have to be zero-terminated if name_size is non-zero + //! \param name_size Size of name, in characters, or 0 to have size calculated automatically from string + //! \param case_sensitive Should name comparison be case-sensitive; non case-sensitive comparison works properly only for ASCII characters + //! \return Pointer to found attribute, or 0 if not found. + xml_attribute<Ch> *first_attribute(const Ch *name = 0, std::size_t name_size = 0, bool case_sensitive = true) const + { + if (name) + { + if (name_size == 0) + name_size = internal::measure(name); + for (xml_attribute<Ch> *attribute = m_first_attribute; attribute; attribute = attribute->m_next_attribute) + if (internal::compare(attribute->name(), attribute->name_size(), name, name_size, case_sensitive)) + return attribute; + return 0; + } + else + return m_first_attribute; + } + + //! Gets last attribute of node, optionally matching attribute name. + //! \param name Name of attribute to find, or 0 to return last attribute regardless of its name; this string doesn't have to be zero-terminated if name_size is non-zero + //! \param name_size Size of name, in characters, or 0 to have size calculated automatically from string + //! \param case_sensitive Should name comparison be case-sensitive; non case-sensitive comparison works properly only for ASCII characters + //! \return Pointer to found attribute, or 0 if not found. + xml_attribute<Ch> *last_attribute(const Ch *name = 0, std::size_t name_size = 0, bool case_sensitive = true) const + { + if (name) + { + if (name_size == 0) + name_size = internal::measure(name); + for (xml_attribute<Ch> *attribute = m_last_attribute; attribute; attribute = attribute->m_prev_attribute) + if (internal::compare(attribute->name(), attribute->name_size(), name, name_size, case_sensitive)) + return attribute; + return 0; + } + else + return m_first_attribute ? m_last_attribute : 0; + } + + /////////////////////////////////////////////////////////////////////////// + // Node modification + + //! Sets type of node. + //! \param type Type of node to set. + void type(node_type type) + { + m_type = type; + } + + /////////////////////////////////////////////////////////////////////////// + // Node manipulation + + //! Prepends a new child node. + //! The prepended child becomes the first child, and all existing children are moved one position back. + //! \param child Node to prepend. + void prepend_node(xml_node<Ch> *child) + { + assert(child && !child->parent() && child->type() != node_document); + if (first_node()) + { + child->m_next_sibling = m_first_node; + m_first_node->m_prev_sibling = child; + } + else + { + child->m_next_sibling = 0; + m_last_node = child; + } + m_first_node = child; + child->m_parent = this; + child->m_prev_sibling = 0; + } + + //! Appends a new child node. + //! The appended child becomes the last child. + //! \param child Node to append. + void append_node(xml_node<Ch> *child) + { + assert(child && !child->parent() && child->type() != node_document); + if (first_node()) + { + child->m_prev_sibling = m_last_node; + m_last_node->m_next_sibling = child; + } + else + { + child->m_prev_sibling = 0; + m_first_node = child; + } + m_last_node = child; + child->m_parent = this; + child->m_next_sibling = 0; + } + + //! Inserts a new child node at specified place inside the node. + //! All children after and including the specified node are moved one position back. + //! \param where Place where to insert the child, or 0 to insert at the back. + //! \param child Node to insert. + void insert_node(xml_node<Ch> *where, xml_node<Ch> *child) + { + assert(!where || where->parent() == this); + assert(child && !child->parent() && child->type() != node_document); + if (where == m_first_node) + prepend_node(child); + else if (where == 0) + append_node(child); + else + { + child->m_prev_sibling = where->m_prev_sibling; + child->m_next_sibling = where; + where->m_prev_sibling->m_next_sibling = child; + where->m_prev_sibling = child; + child->m_parent = this; + } + } + + //! Removes first child node. + //! If node has no children, behaviour is undefined. + //! Use first_node() to test if node has children. + void remove_first_node() + { + assert(first_node()); + xml_node<Ch> *child = m_first_node; + m_first_node = child->m_next_sibling; + if (child->m_next_sibling) + child->m_next_sibling->m_prev_sibling = 0; + else + m_last_node = 0; + child->m_parent = 0; + } + + //! Removes last child of the node. + //! If node has no children, behaviour is undefined. + //! Use first_node() to test if node has children. + void remove_last_node() + { + assert(first_node()); + xml_node<Ch> *child = m_last_node; + if (child->m_prev_sibling) + { + m_last_node = child->m_prev_sibling; + child->m_prev_sibling->m_next_sibling = 0; + } + else + m_first_node = 0; + child->m_parent = 0; + } + + //! Removes specified child from the node + // \param where Pointer to child to be removed. + void remove_node(xml_node<Ch> *where) + { + assert(where && where->parent() == this); + assert(first_node()); + if (where == m_first_node) + remove_first_node(); + else if (where == m_last_node) + remove_last_node(); + else + { + where->m_prev_sibling->m_next_sibling = where->m_next_sibling; + where->m_next_sibling->m_prev_sibling = where->m_prev_sibling; + where->m_parent = 0; + } + } + + //! Removes all child nodes (but not attributes). + void remove_all_nodes() + { + for (xml_node<Ch> *node = first_node(); node; node = node->m_next_sibling) + node->m_parent = 0; + m_first_node = 0; + } + + //! Prepends a new attribute to the node. + //! \param attribute Attribute to prepend. + void prepend_attribute(xml_attribute<Ch> *attribute) + { + assert(attribute && !attribute->parent()); + if (first_attribute()) + { + attribute->m_next_attribute = m_first_attribute; + m_first_attribute->m_prev_attribute = attribute; + } + else + { + attribute->m_next_attribute = 0; + m_last_attribute = attribute; + } + m_first_attribute = attribute; + attribute->m_parent = this; + attribute->m_prev_attribute = 0; + } + + //! Appends a new attribute to the node. + //! \param attribute Attribute to append. + void append_attribute(xml_attribute<Ch> *attribute) + { + assert(attribute && !attribute->parent()); + if (first_attribute()) + { + attribute->m_prev_attribute = m_last_attribute; + m_last_attribute->m_next_attribute = attribute; + } + else + { + attribute->m_prev_attribute = 0; + m_first_attribute = attribute; + } + m_last_attribute = attribute; + attribute->m_parent = this; + attribute->m_next_attribute = 0; + } + + //! Inserts a new attribute at specified place inside the node. + //! All attributes after and including the specified attribute are moved one position back. + //! \param where Place where to insert the attribute, or 0 to insert at the back. + //! \param attribute Attribute to insert. + void insert_attribute(xml_attribute<Ch> *where, xml_attribute<Ch> *attribute) + { + assert(!where || where->parent() == this); + assert(attribute && !attribute->parent()); + if (where == m_first_attribute) + prepend_attribute(attribute); + else if (where == 0) + append_attribute(attribute); + else + { + attribute->m_prev_attribute = where->m_prev_attribute; + attribute->m_next_attribute = where; + where->m_prev_attribute->m_next_attribute = attribute; + where->m_prev_attribute = attribute; + attribute->m_parent = this; + } + } + + //! Removes first attribute of the node. + //! If node has no attributes, behaviour is undefined. + //! Use first_attribute() to test if node has attributes. + void remove_first_attribute() + { + assert(first_attribute()); + xml_attribute<Ch> *attribute = m_first_attribute; + if (attribute->m_next_attribute) + { + attribute->m_next_attribute->m_prev_attribute = 0; + } + else + m_last_attribute = 0; + attribute->m_parent = 0; + m_first_attribute = attribute->m_next_attribute; + } + + //! Removes last attribute of the node. + //! If node has no attributes, behaviour is undefined. + //! Use first_attribute() to test if node has attributes. + void remove_last_attribute() + { + assert(first_attribute()); + xml_attribute<Ch> *attribute = m_last_attribute; + if (attribute->m_prev_attribute) + { + attribute->m_prev_attribute->m_next_attribute = 0; + m_last_attribute = attribute->m_prev_attribute; + } + else + m_first_attribute = 0; + attribute->m_parent = 0; + } + + //! Removes specified attribute from node. + //! \param where Pointer to attribute to be removed. + void remove_attribute(xml_attribute<Ch> *where) + { + assert(first_attribute() && where->parent() == this); + if (where == m_first_attribute) + remove_first_attribute(); + else if (where == m_last_attribute) + remove_last_attribute(); + else + { + where->m_prev_attribute->m_next_attribute = where->m_next_attribute; + where->m_next_attribute->m_prev_attribute = where->m_prev_attribute; + where->m_parent = 0; + } + } + + //! Removes all attributes of node. + void remove_all_attributes() + { + for (xml_attribute<Ch> *attribute = first_attribute(); attribute; attribute = attribute->m_next_attribute) + attribute->m_parent = 0; + m_first_attribute = 0; + } + + private: + + /////////////////////////////////////////////////////////////////////////// + // Restrictions + + // No copying + xml_node(const xml_node &); + void operator =(const xml_node &); + + /////////////////////////////////////////////////////////////////////////// + // Data members + + // Note that some of the pointers below have UNDEFINED values if certain other pointers are 0. + // This is required for maximum performance, as it allows the parser to omit initialization of + // unneded/redundant values. + // + // The rules are as follows: + // 1. first_node and first_attribute contain valid pointers, or 0 if node has no children/attributes respectively + // 2. last_node and last_attribute are valid only if node has at least one child/attribute respectively, otherwise they contain garbage + // 3. prev_sibling and next_sibling are valid only if node has a parent, otherwise they contain garbage + + node_type m_type; // Type of node; always valid + xml_node<Ch> *m_first_node; // Pointer to first child node, or 0 if none; always valid + xml_node<Ch> *m_last_node; // Pointer to last child node, or 0 if none; this value is only valid if m_first_node is non-zero + xml_attribute<Ch> *m_first_attribute; // Pointer to first attribute of node, or 0 if none; always valid + xml_attribute<Ch> *m_last_attribute; // Pointer to last attribute of node, or 0 if none; this value is only valid if m_first_attribute is non-zero + xml_node<Ch> *m_prev_sibling; // Pointer to previous sibling of node, or 0 if none; this value is only valid if m_parent is non-zero + xml_node<Ch> *m_next_sibling; // Pointer to next sibling of node, or 0 if none; this value is only valid if m_parent is non-zero + + }; + + /////////////////////////////////////////////////////////////////////////// + // XML document + + //! This class represents root of the DOM hierarchy. + //! It is also an xml_node and a memory_pool through public inheritance. + //! Use parse() function to build a DOM tree from a zero-terminated XML text string. + //! parse() function allocates memory for nodes and attributes by using functions of xml_document, + //! which are inherited from memory_pool. + //! To access root node of the document, use the document itself, as if it was an xml_node. + //! \param Ch Character type to use. + template<class Ch = char> + class xml_document: public xml_node<Ch>, public memory_pool<Ch> + { + + public: + + //! Constructs empty XML document + xml_document() + : xml_node<Ch>(node_document) + { + } + + //! Parses zero-terminated XML string according to given flags. + //! Passed string will be modified by the parser, unless rapidxml::parse_non_destructive flag is used. + //! The string must persist for the lifetime of the document. + //! In case of error, rapidxml::parse_error exception will be thrown. + //! <br><br> + //! If you want to parse contents of a file, you must first load the file into the memory, and pass pointer to its beginning. + //! Make sure that data is zero-terminated. + //! <br><br> + //! Document can be parsed into multiple times. + //! Each new call to parse removes previous nodes and attributes (if any), but does not clear memory pool. + //! \param text XML data to parse; pointer is non-const to denote fact that this data may be modified by the parser. + template<int Flags> + void parse(Ch *text) + { + assert(text); + + // Remove current contents + this->remove_all_nodes(); + this->remove_all_attributes(); + + // Parse BOM, if any + parse_bom<Flags>(text); + + // Parse children + while (1) + { + // Skip whitespace before node + skip<whitespace_pred, Flags>(text); + if (*text == 0) + break; + + // Parse and append new child + if (*text == Ch('<')) + { + ++text; // Skip '<' + if (xml_node<Ch> *node = parse_node<Flags>(text)) + this->append_node(node); + } + else + RAPIDXML_PARSE_ERROR("expected <", text); + } + + } + + //! Clears the document by deleting all nodes and clearing the memory pool. + //! All nodes owned by document pool are destroyed. + void clear() + { + this->remove_all_nodes(); + this->remove_all_attributes(); + memory_pool<Ch>::clear(); + } + + private: + + /////////////////////////////////////////////////////////////////////// + // Internal character utility functions + + // Detect whitespace character + struct whitespace_pred + { + static unsigned char test(Ch ch) + { + return internal::lookup_tables<0>::lookup_whitespace[static_cast<unsigned char>(ch)]; + } + }; + + // Detect node name character + struct node_name_pred + { + static unsigned char test(Ch ch) + { + return internal::lookup_tables<0>::lookup_node_name[static_cast<unsigned char>(ch)]; + } + }; + + // Detect attribute name character + struct attribute_name_pred + { + static unsigned char test(Ch ch) + { + return internal::lookup_tables<0>::lookup_attribute_name[static_cast<unsigned char>(ch)]; + } + }; + + // Detect text character (PCDATA) + struct text_pred + { + static unsigned char test(Ch ch) + { + return internal::lookup_tables<0>::lookup_text[static_cast<unsigned char>(ch)]; + } + }; + + // Detect text character (PCDATA) that does not require processing + struct text_pure_no_ws_pred + { + static unsigned char test(Ch ch) + { + return internal::lookup_tables<0>::lookup_text_pure_no_ws[static_cast<unsigned char>(ch)]; + } + }; + + // Detect text character (PCDATA) that does not require processing + struct text_pure_with_ws_pred + { + static unsigned char test(Ch ch) + { + return internal::lookup_tables<0>::lookup_text_pure_with_ws[static_cast<unsigned char>(ch)]; + } + }; + + // Detect attribute value character + template<Ch Quote> + struct attribute_value_pred + { + static unsigned char test(Ch ch) + { + if (Quote == Ch('\'')) + return internal::lookup_tables<0>::lookup_attribute_data_1[static_cast<unsigned char>(ch)]; + if (Quote == Ch('\"')) + return internal::lookup_tables<0>::lookup_attribute_data_2[static_cast<unsigned char>(ch)]; + return 0; // Should never be executed, to avoid warnings on Comeau + } + }; + + // Detect attribute value character + template<Ch Quote> + struct attribute_value_pure_pred + { + static unsigned char test(Ch ch) + { + if (Quote == Ch('\'')) + return internal::lookup_tables<0>::lookup_attribute_data_1_pure[static_cast<unsigned char>(ch)]; + if (Quote == Ch('\"')) + return internal::lookup_tables<0>::lookup_attribute_data_2_pure[static_cast<unsigned char>(ch)]; + return 0; // Should never be executed, to avoid warnings on Comeau + } + }; + + // Insert coded character, using UTF8 or 8-bit ASCII + template<int Flags> + static void insert_coded_character(Ch *&text, unsigned long code) + { + if (Flags & parse_no_utf8) + { + // Insert 8-bit ASCII character + // Todo: possibly verify that code is less than 256 and use replacement char otherwise? + text[0] = static_cast<unsigned char>(code); + text += 1; + } + else + { + // Insert UTF8 sequence + if (code < 0x80) // 1 byte sequence + { + text[0] = static_cast<unsigned char>(code); + text += 1; + } + else if (code < 0x800) // 2 byte sequence + { + text[1] = static_cast<unsigned char>((code | 0x80) & 0xBF); code >>= 6; + text[0] = static_cast<unsigned char>(code | 0xC0); + text += 2; + } + else if (code < 0x10000) // 3 byte sequence + { + text[2] = static_cast<unsigned char>((code | 0x80) & 0xBF); code >>= 6; + text[1] = static_cast<unsigned char>((code | 0x80) & 0xBF); code >>= 6; + text[0] = static_cast<unsigned char>(code | 0xE0); + text += 3; + } + else if (code < 0x110000) // 4 byte sequence + { + text[3] = static_cast<unsigned char>((code | 0x80) & 0xBF); code >>= 6; + text[2] = static_cast<unsigned char>((code | 0x80) & 0xBF); code >>= 6; + text[1] = static_cast<unsigned char>((code | 0x80) & 0xBF); code >>= 6; + text[0] = static_cast<unsigned char>(code | 0xF0); + text += 4; + } + else // Invalid, only codes up to 0x10FFFF are allowed in Unicode + { + RAPIDXML_PARSE_ERROR("invalid numeric character entity", text); + } + } + } + + // Skip characters until predicate evaluates to true + template<class StopPred, int Flags> + static void skip(Ch *&text) + { + Ch *tmp = text; + while (StopPred::test(*tmp)) + ++tmp; + text = tmp; + } + + // Skip characters until predicate evaluates to true while doing the following: + // - replacing XML character entity references with proper characters (' & " < > &#...;) + // - condensing whitespace sequences to single space character + template<class StopPred, class StopPredPure, int Flags> + static Ch *skip_and_expand_character_refs(Ch *&text) + { + // If entity translation, whitespace condense and whitespace trimming is disabled, use plain skip + if (Flags & parse_no_entity_translation && + !(Flags & parse_normalize_whitespace) && + !(Flags & parse_trim_whitespace)) + { + skip<StopPred, Flags>(text); + return text; + } + + // Use simple skip until first modification is detected + skip<StopPredPure, Flags>(text); + + // Use translation skip + Ch *src = text; + Ch *dest = src; + while (StopPred::test(*src)) + { + // If entity translation is enabled + if (!(Flags & parse_no_entity_translation)) + { + // Test if replacement is needed + if (src[0] == Ch('&')) + { + switch (src[1]) + { + + // & ' + case Ch('a'): + if (src[2] == Ch('m') && src[3] == Ch('p') && src[4] == Ch(';')) + { + *dest = Ch('&'); + ++dest; + src += 5; + continue; + } + if (src[2] == Ch('p') && src[3] == Ch('o') && src[4] == Ch('s') && src[5] == Ch(';')) + { + *dest = Ch('\''); + ++dest; + src += 6; + continue; + } + break; + + // " + case Ch('q'): + if (src[2] == Ch('u') && src[3] == Ch('o') && src[4] == Ch('t') && src[5] == Ch(';')) + { + *dest = Ch('"'); + ++dest; + src += 6; + continue; + } + break; + + // > + case Ch('g'): + if (src[2] == Ch('t') && src[3] == Ch(';')) + { + *dest = Ch('>'); + ++dest; + src += 4; + continue; + } + break; + + // < + case Ch('l'): + if (src[2] == Ch('t') && src[3] == Ch(';')) + { + *dest = Ch('<'); + ++dest; + src += 4; + continue; + } + break; + + // &#...; - assumes ASCII + case Ch('#'): + if (src[2] == Ch('x')) + { + unsigned long code = 0; + src += 3; // Skip &#x + while (1) + { + unsigned char digit = internal::lookup_tables<0>::lookup_digits[static_cast<unsigned char>(*src)]; + if (digit == 0xFF) + break; + code = code * 16 + digit; + ++src; + } + insert_coded_character<Flags>(dest, code); // Put character in output + } + else + { + unsigned long code = 0; + src += 2; // Skip &# + while (1) + { + unsigned char digit = internal::lookup_tables<0>::lookup_digits[static_cast<unsigned char>(*src)]; + if (digit == 0xFF) + break; + code = code * 10 + digit; + ++src; + } + insert_coded_character<Flags>(dest, code); // Put character in output + } + if (*src == Ch(';')) + ++src; + else + RAPIDXML_PARSE_ERROR("expected ;", src); + continue; + + // Something else + default: + // Ignore, just copy '&' verbatim + break; + + } + } + } + + // If whitespace condensing is enabled + if (Flags & parse_normalize_whitespace) + { + // Test if condensing is needed + if (whitespace_pred::test(*src)) + { + *dest = Ch(' '); ++dest; // Put single space in dest + ++src; // Skip first whitespace char + // Skip remaining whitespace chars + while (whitespace_pred::test(*src)) + ++src; + continue; + } + } + + // No replacement, only copy character + *dest++ = *src++; + + } + + // Return new end + text = src; + return dest; + + } + + /////////////////////////////////////////////////////////////////////// + // Internal parsing functions + + // Parse BOM, if any + template<int Flags> + void parse_bom(Ch *&text) + { + // UTF-8? + if (static_cast<unsigned char>(text[0]) == 0xEF && + static_cast<unsigned char>(text[1]) == 0xBB && + static_cast<unsigned char>(text[2]) == 0xBF) + { + text += 3; // Skup utf-8 bom + } + } + + // Parse XML declaration (<?xml...) + template<int Flags> + xml_node<Ch> *parse_xml_declaration(Ch *&text) + { + // If parsing of declaration is disabled + if (!(Flags & parse_declaration_node)) + { + // Skip until end of declaration + while (text[0] != Ch('?') || text[1] != Ch('>')) + { + if (!text[0]) + RAPIDXML_PARSE_ERROR("unexpected end of data", text); + ++text; + } + text += 2; // Skip '?>' + return 0; + } + + // Create declaration + xml_node<Ch> *declaration = this->allocate_node(node_declaration); + + // Skip whitespace before attributes or ?> + skip<whitespace_pred, Flags>(text); + + // Parse declaration attributes + parse_node_attributes<Flags>(text, declaration); + + // Skip ?> + if (text[0] != Ch('?') || text[1] != Ch('>')) + RAPIDXML_PARSE_ERROR("expected ?>", text); + text += 2; + + return declaration; + } + + // Parse XML comment (<!--...) + template<int Flags> + xml_node<Ch> *parse_comment(Ch *&text) + { + // If parsing of comments is disabled + if (!(Flags & parse_comment_nodes)) + { + // Skip until end of comment + while (text[0] != Ch('-') || text[1] != Ch('-') || text[2] != Ch('>')) + { + if (!text[0]) + RAPIDXML_PARSE_ERROR("unexpected end of data", text); + ++text; + } + text += 3; // Skip '-->' + return 0; // Do not produce comment node + } + + // Remember value start + Ch *value = text; + + // Skip until end of comment + while (text[0] != Ch('-') || text[1] != Ch('-') || text[2] != Ch('>')) + { + if (!text[0]) + RAPIDXML_PARSE_ERROR("unexpected end of data", text); + ++text; + } + + // Create comment node + xml_node<Ch> *comment = this->allocate_node(node_comment); + comment->value(value, text - value); + + // Place zero terminator after comment value + if (!(Flags & parse_no_string_terminators)) + *text = Ch('\0'); + + text += 3; // Skip '-->' + return comment; + } + + // Parse DOCTYPE + template<int Flags> + xml_node<Ch> *parse_doctype(Ch *&text) + { + // Remember value start + Ch *value = text; + + // Skip to > + while (*text != Ch('>')) + { + // Determine character type + switch (*text) + { + + // If '[' encountered, scan for matching ending ']' using naive algorithm with depth + // This works for all W3C test files except for 2 most wicked + case Ch('['): + { + ++text; // Skip '[' + int depth = 1; + while (depth > 0) + { + switch (*text) + { + case Ch('['): ++depth; break; + case Ch(']'): --depth; break; + case 0: RAPIDXML_PARSE_ERROR("unexpected end of data", text); + } + ++text; + } + break; + } + + // Error on end of text + case Ch('\0'): + RAPIDXML_PARSE_ERROR("unexpected end of data", text); + + // Other character, skip it + default: + ++text; + + } + } + + // If DOCTYPE nodes enabled + if (Flags & parse_doctype_node) + { + // Create a new doctype node + xml_node<Ch> *doctype = this->allocate_node(node_doctype); + doctype->value(value, text - value); + + // Place zero terminator after value + if (!(Flags & parse_no_string_terminators)) + *text = Ch('\0'); + + text += 1; // skip '>' + return doctype; + } + else + { + text += 1; // skip '>' + return 0; + } + + } + + // Parse PI + template<int Flags> + xml_node<Ch> *parse_pi(Ch *&text) + { + // If creation of PI nodes is enabled + if (Flags & parse_pi_nodes) + { + // Create pi node + xml_node<Ch> *pi = this->allocate_node(node_pi); + + // Extract PI target name + Ch *name = text; + skip<node_name_pred, Flags>(text); + if (text == name) + RAPIDXML_PARSE_ERROR("expected PI target", text); + pi->name(name, text - name); + + // Skip whitespace between pi target and pi + skip<whitespace_pred, Flags>(text); + + // Remember start of pi + Ch *value = text; + + // Skip to '?>' + while (text[0] != Ch('?') || text[1] != Ch('>')) + { + if (*text == Ch('\0')) + RAPIDXML_PARSE_ERROR("unexpected end of data", text); + ++text; + } + + // Set pi value (verbatim, no entity expansion or whitespace normalization) + pi->value(value, text - value); + + // Place zero terminator after name and value + if (!(Flags & parse_no_string_terminators)) + { + pi->name()[pi->name_size()] = Ch('\0'); + pi->value()[pi->value_size()] = Ch('\0'); + } + + text += 2; // Skip '?>' + return pi; + } + else + { + // Skip to '?>' + while (text[0] != Ch('?') || text[1] != Ch('>')) + { + if (*text == Ch('\0')) + RAPIDXML_PARSE_ERROR("unexpected end of data", text); + ++text; + } + text += 2; // Skip '?>' + return 0; + } + } + + // Parse and append data + // Return character that ends data. + // This is necessary because this character might have been overwritten by a terminating 0 + template<int Flags> + Ch parse_and_append_data(xml_node<Ch> *node, Ch *&text, Ch *contents_start) + { + // Backup to contents start if whitespace trimming is disabled + if (!(Flags & parse_trim_whitespace)) + text = contents_start; + + // Skip until end of data + Ch *value = text, *end; + if (Flags & parse_normalize_whitespace) + end = skip_and_expand_character_refs<text_pred, text_pure_with_ws_pred, Flags>(text); + else + end = skip_and_expand_character_refs<text_pred, text_pure_no_ws_pred, Flags>(text); + + // Trim trailing whitespace if flag is set; leading was already trimmed by whitespace skip after > + if (Flags & parse_trim_whitespace) + { + if (Flags & parse_normalize_whitespace) + { + // Whitespace is already condensed to single space characters by skipping function, so just trim 1 char off the end + if (*(end - 1) == Ch(' ')) + --end; + } + else + { + // Backup until non-whitespace character is found + while (whitespace_pred::test(*(end - 1))) + --end; + } + } + + // If characters are still left between end and value (this test is only necessary if normalization is enabled) + // Create new data node + if (!(Flags & parse_no_data_nodes)) + { + xml_node<Ch> *data = this->allocate_node(node_data); + data->value(value, end - value); + node->append_node(data); + } + + // Add data to parent node if no data exists yet + if (!(Flags & parse_no_element_values)) + if (*node->value() == Ch('\0')) + node->value(value, end - value); + + // Place zero terminator after value + if (!(Flags & parse_no_string_terminators)) + { + Ch ch = *text; + *end = Ch('\0'); + return ch; // Return character that ends data; this is required because zero terminator overwritten it + } + + // Return character that ends data + return *text; + } + + // Parse CDATA + template<int Flags> + xml_node<Ch> *parse_cdata(Ch *
<TRUNCATED>
