mxml-file.c

适用于嵌入式系统的XML解析库, 规模比libxml2小得多.

/*
 * "$Id: mxml-file.c 329 2008-01-13 00:42:35Z mike $"
 *
 * File loading code for Mini-XML, a small XML-like file parsing library.
 *
 * Copyright 2003-2008 by Michael Sweet.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Library General Public
 * License as published by the Free Software Foundation; either
 * version 2, or (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * Contents:
 *
 *   mxmlLoadFd()            - Load a file descriptor into an XML node tree.
 *   mxmlLoadFile()          - Load a file into an XML node tree.
 *   mxmlLoadString()        - Load a string into an XML node tree.
 *   mxmlSaveAllocString()   - Save an XML node tree to an allocated string.
 *   mxmlSaveFd()            - Save an XML tree to a file descriptor.
 *   mxmlSaveFile()          - Save an XML tree to a file.
 *   mxmlSaveString()        - Save an XML node tree to a string.
 *   mxmlSAXLoadFd()         - Load a file descriptor into an XML node tree
 *                             using a SAX callback.
 *   mxmlSAXLoadFile()       - Load a file into an XML node tree
 *                             using a SAX callback.
 *   mxmlSAXLoadString()     - Load a string into an XML node tree
 *                             using a SAX callback.
 *   mxmlSetCustomHandlers() - Set the handling functions for custom data.
 *   mxmlSetErrorCallback()  - Set the error message callback.
 *   mxmlSetWrapMargin()     - Set the the wrap margin when saving XML data.
 *   mxml_add_char()         - Add a character to a buffer, expanding as needed.
 *   mxml_fd_getc()          - Read a character from a file descriptor.
 *   mxml_fd_putc()          - Write a character to a file descriptor.
 *   mxml_fd_read()          - Read a buffer of data from a file descriptor.
 *   mxml_fd_write()         - Write a buffer of data to a file descriptor.
 *   mxml_file_getc()        - Get a character from a file.
 *   mxml_file_putc()        - Write a character to a file.
 *   mxml_get_entity()       - Get the character corresponding to an entity...
 *   mxml_load_data()        - Load data into an XML node tree.
 *   mxml_parse_element()    - Parse an element for any attributes...
 *   mxml_string_getc()      - Get a character from a string.
 *   mxml_string_putc()      - Write a character to a string.
 *   mxml_write_name()       - Write a name string.
 *   mxml_write_node()       - Save an XML node to a file.
 *   mxml_write_string()     - Write a string, escaping & and < as needed.
 *   mxml_write_ws()         - Do whitespace callback...
 */

/*
 * Include necessary headers...
 */

#include "mxml-private.h"
#ifdef WIN32
#  include <io.h>
#else
#  include <unistd.h>
#endif /* WIN32 */


/*
 * Character encoding...
 */

#define ENCODE_UTF8	0		/* UTF-8 */
#define ENCODE_UTF16BE	1		/* UTF-16 Big-Endian */
#define ENCODE_UTF16LE	2		/* UTF-16 Little-Endian */


/*
 * Macro to test for a bad XML character...
 */

#define mxml_bad_char(ch) ((ch) < ' ' && (ch) != '\n' && (ch) != '\r' && (ch) != '\t')


/*
 * Types and structures...
 */

typedef int (*_mxml_getc_cb_t)(void *, int *);
typedef int (*_mxml_putc_cb_t)(int, void *);

typedef struct _mxml_fdbuf_s		/**** File descriptor buffer ****/
{
  int		fd;			/* File descriptor */
  unsigned char	*current,		/* Current position in buffer */
		*end,			/* End of buffer */
		buffer[8192];		/* Character buffer */
} _mxml_fdbuf_t;


/*
 * Local functions...
 */

static int		mxml_add_char(int ch, char **ptr, char **buffer,
			              int *bufsize);
static int		mxml_fd_getc(void *p, int *encoding);
static int		mxml_fd_putc(int ch, void *p);
static int		mxml_fd_read(_mxml_fdbuf_t *buf);
static int		mxml_fd_write(_mxml_fdbuf_t *buf);
static int		mxml_file_getc(void *p, int *encoding);
static int		mxml_file_putc(int ch, void *p);
static int		mxml_get_entity(mxml_node_t *parent, void *p,
			                int *encoding,
					_mxml_getc_cb_t getc_cb);
static inline int	mxml_isspace(int ch)
			{
			  return (ch == ' ' || ch == '\t' || ch == '\r' ||
			          ch == '\n');
			}
static mxml_node_t	*mxml_load_data(mxml_node_t *top, void *p,
			                mxml_load_cb_t cb,
			                _mxml_getc_cb_t getc_cb,
                                        mxml_sax_cb_t sax_cb, void *sax_data);
static int		mxml_parse_element(mxml_node_t *node, void *p,
			                   int *encoding,
					   _mxml_getc_cb_t getc_cb);
static int		mxml_string_getc(void *p, int *encoding);
static int		mxml_string_putc(int ch, void *p);
static int		mxml_write_name(const char *s, void *p,
					_mxml_putc_cb_t putc_cb);
static int		mxml_write_node(mxml_node_t *node, void *p,
			                mxml_save_cb_t cb, int col,
					_mxml_putc_cb_t putc_cb,
					_mxml_global_t *global);
static int		mxml_write_string(const char *s, void *p,
					  _mxml_putc_cb_t putc_cb);
static int		mxml_write_ws(mxml_node_t *node, void *p, 
			              mxml_save_cb_t cb, int ws,
				      int col, _mxml_putc_cb_t putc_cb);


/*
 * 'mxmlLoadFd()' - Load a file descriptor into an XML node tree.
 *
 * The nodes in the specified file are added to the specified top node.
 * If no top node is provided, the XML file MUST be well-formed with a
 * single parent node like <?xml> for the entire file. The callback
 * function returns the value type that should be used for child nodes.
 * If MXML_NO_CALLBACK is specified then all child nodes will be either
 * MXML_ELEMENT or MXML_TEXT nodes.
 *
 * The constants MXML_INTEGER_CALLBACK, MXML_OPAQUE_CALLBACK,
 * MXML_REAL_CALLBACK, and MXML_TEXT_CALLBACK are defined for loading
 * child nodes of the specified type.
 */

mxml_node_t *				/* O - First node or NULL if the file could not be read. */
mxmlLoadFd(mxml_node_t    *top,		/* I - Top node */
           int            fd,		/* I - File descriptor to read from */
           mxml_load_cb_t cb)		/* I - Callback function or MXML_NO_CALLBACK */
{
  _mxml_fdbuf_t	buf;			/* File descriptor buffer */


 /*
  * Initialize the file descriptor buffer...
  */

  buf.fd      = fd;
  buf.current = buf.buffer;
  buf.end     = buf.buffer;

 /*
  * Read the XML data...
  */

  return (mxml_load_data(top, &buf, cb, mxml_fd_getc, MXML_NO_CALLBACK, NULL));
}


/*
 * 'mxmlLoadFile()' - Load a file into an XML node tree.
 *
 * The nodes in the specified file are added to the specified top node.
 * If no top node is provided, the XML file MUST be well-formed with a
 * single parent node like <?xml> for the entire file. The callback
 * function returns the value type that should be used for child nodes.
 * If MXML_NO_CALLBACK is specified then all child nodes will be either
 * MXML_ELEMENT or MXML_TEXT nodes.
 *
 * The constants MXML_INTEGER_CALLBACK, MXML_OPAQUE_CALLBACK,
 * MXML_REAL_CALLBACK, and MXML_TEXT_CALLBACK are defined for loading
 * child nodes of the specified type.
 */

mxml_node_t *				/* O - First node or NULL if the file could not be read. */
mxmlLoadFile(mxml_node_t    *top,	/* I - Top node */
             FILE           *fp,	/* I - File to read from */
             mxml_load_cb_t cb)		/* I - Callback function or MXML_NO_CALLBACK */
{
 /*
  * Read the XML data...
  */

  return (mxml_load_data(top, fp, cb, mxml_file_getc, MXML_NO_CALLBACK, NULL));
}


/*
 * 'mxmlLoadString()' - Load a string into an XML node tree.
 *
 * The nodes in the specified string are added to the specified top node.
 * If no top node is provided, the XML string MUST be well-formed with a
 * single parent node like <?xml> for the entire string. The callback
 * function returns the value type that should be used for child nodes.
 * If MXML_NO_CALLBACK is specified then all child nodes will be either
 * MXML_ELEMENT or MXML_TEXT nodes.
 *
 * The constants MXML_INTEGER_CALLBACK, MXML_OPAQUE_CALLBACK,
 * MXML_REAL_CALLBACK, and MXML_TEXT_CALLBACK are defined for loading
 * child nodes of the specified type.
 */

mxml_node_t *				/* O - First node or NULL if the string has errors. */
mxmlLoadString(mxml_node_t    *top,	/* I - Top node */
               const char     *s,	/* I - String to load */
               mxml_load_cb_t cb)	/* I - Callback function or MXML_NO_CALLBACK */
{
 /*
  * Read the XML data...
  */

  return (mxml_load_data(top, (void *)&s, cb, mxml_string_getc, MXML_NO_CALLBACK,
                         NULL));
}


/*
 * 'mxmlSaveAllocString()' - Save an XML node tree to an allocated string.
 *
 * This function returns a pointer to a string containing the textual
 * representation of the XML node tree.  The string should be freed
 * using the free() function when you are done with it.  NULL is returned
 * if the node would produce an empty string or if the string cannot be
 * allocated.
 *
 * The callback argument specifies a function that returns a whitespace
 * string or NULL before and after each element. If MXML_NO_CALLBACK
 * is specified, whitespace will only be added before MXML_TEXT nodes
 * with leading whitespace and before attribute names inside opening
 * element tags.
 */

char *					/* O - Allocated string or NULL */
mxmlSaveAllocString(
    mxml_node_t    *node,		/* I - Node to write */
    mxml_save_cb_t cb)			/* I - Whitespace callback or MXML_NO_CALLBACK */
{
  int	bytes;				/* Required bytes */
  char	buffer[8192];			/* Temporary buffer */
  char	*s;				/* Allocated string */


 /*
  * Write the node to the temporary buffer...
  */

  bytes = mxmlSaveString(node, buffer, sizeof(buffer), cb);

  if (bytes <= 0)
    return (NULL);

  if (bytes < (int)(sizeof(buffer) - 1))
  {
   /*
    * Node fit inside the buffer, so just duplicate that string and
    * return...
    */

    return (strdup(buffer));
  }

 /*
  * Allocate a buffer of the required size and save the node to the
  * new buffer...
  */

  if ((s = malloc(bytes + 1)) == NULL)
    return (NULL);

  mxmlSaveString(node, s, bytes + 1, cb);

 /*
  * Return the allocated string...
  */

  return (s);
}


/*
 * 'mxmlSaveFd()' - Save an XML tree to a file descriptor.
 *
 * The callback argument specifies a function that returns a whitespace
 * string or NULL before and after each element. If MXML_NO_CALLBACK
 * is specified, whitespace will only be added before MXML_TEXT nodes
 * with leading whitespace and before attribute names inside opening
 * element tags.
 */

int					/* O - 0 on success, -1 on error. */
mxmlSaveFd(mxml_node_t    *node,	/* I - Node to write */
           int            fd,		/* I - File descriptor to write to */
	   mxml_save_cb_t cb)		/* I - Whitespace callback or MXML_NO_CALLBACK */
{
  int		col;			/* Final column */
  _mxml_fdbuf_t	buf;			/* File descriptor buffer */
  _mxml_global_t *global = _mxml_global();
					/* Global data */


 /*
  * Initialize the file descriptor buffer...
  */

  buf.fd      = fd;
  buf.current = buf.buffer;
  buf.end     = buf.buffer + sizeof(buf.buffer) - 4;

 /*
  * Write the node...
  */

  if ((col = mxml_write_node(node, &buf, cb, 0, mxml_fd_putc, global)) < 0)
    return (-1);

  if (col > 0)
    if (mxml_fd_putc('\n', &buf) < 0)
      return (-1);

 /*
  * Flush and return...
  */

  return (mxml_fd_write(&buf));
}


/*
 * 'mxmlSaveFile()' - Save an XML tree to a file.
 *
 * The callback argument specifies a function that returns a whitespace
 * string or NULL before and after each element. If MXML_NO_CALLBACK
 * is specified, whitespace will only be added before MXML_TEXT nodes
 * with leading whitespace and before attribute names inside opening
 * element tags.
 */

int					/* O - 0 on success, -1 on error. */
mxmlSaveFile(mxml_node_t    *node,	/* I - Node to write */
             FILE           *fp,	/* I - File to write to */
	     mxml_save_cb_t cb)		/* I - Whitespace callback or MXML_NO_CALLBACK */
{
  int	col;				/* Final column */
  _mxml_global_t *global = _mxml_global();
					/* Global data */


 /*
  * Write the node...
  */

  if ((col = mxml_write_node(node, fp, cb, 0, mxml_file_putc, global)) < 0)
    return (-1);

  if (col > 0)
    if (putc('\n', fp) < 0)
      return (-1);

 /*
  * Return 0 (success)...
  */

  return (0);
}


/*
 * 'mxmlSaveString()' - Save an XML node tree to a string.
 *
 * This function returns the total number of bytes that would be
 * required for the string but only copies (bufsize - 1) characters
 * into the specified buffer.
 *
 * The callback argument specifies a function that returns a whitespace
 * string or NULL before and after each element. If MXML_NO_CALLBACK
 * is specified, whitespace will only be added before MXML_TEXT nodes
 * with leading whitespace and before attribute names inside opening
 * element tags.
 */

int					/* O - Size of string */
mxmlSaveString(mxml_node_t    *node,	/* I - Node to write */
               char           *buffer,	/* I - String buffer */
               int            bufsize,	/* I - Size of string buffer */
               mxml_save_cb_t cb)	/* I - Whitespace callback or MXML_NO_CALLBACK */
{
  int	col;				/* Final column */
  char	*ptr[2];			/* Pointers for putc_cb */
  _mxml_global_t *global = _mxml_global();
					/* Global data */


 /*
  * Write the node...
  */

  ptr[0] = buffer;
  ptr[1] = buffer + bufsize;

  if ((col = mxml_write_node(node, ptr, cb, 0, mxml_string_putc, global)) < 0)
    return (-1);

  if (col > 0)
    mxml_string_putc('\n', ptr);

 /*
  * Nul-terminate the buffer...
  */

  if (ptr[0] >= ptr[1])
    buffer[bufsize - 1] = '\0';
  else
    ptr[0][0] = '\0';

 /*
  * Return the number of characters...
  */

  return (ptr[0] - buffer);
}


/*
 * 'mxmlSAXLoadFd()' - Load a file descriptor into an XML node tree
 *                     using a SAX callback.
 *
 * The nodes in the specified file are added to the specified top node.
 * If no top node is provided, the XML file MUST be well-formed with a
 * single parent node like <?xml> for the entire file. The callback
 * function returns the value type that should be used for child nodes.
 * If MXML_NO_CALLBACK is specified then all child nodes will be either
 * MXML_ELEMENT or MXML_TEXT nodes.
 *
 * The constants MXML_INTEGER_CALLBACK, MXML_OPAQUE_CALLBACK,
 * MXML_REAL_CALLBACK, and MXML_TEXT_CALLBACK are defined for loading
 * child nodes of the specified type.
 *
 * The SAX callback must call mxmlRetain() for any nodes that need to
 * be kept for later use. Otherwise, nodes are deleted when the parent
 * node is closed or after each data, comment, CDATA, or directive node.
 *
 * @since Mini-XML 2.3@
 */

mxml_node_t *				/* O - First node or NULL if the file could not be read. */
mxmlSAXLoadFd(mxml_node_t    *top,	/* I - Top node */
              int            fd,	/* I - File descriptor to read from */
              mxml_load_cb_t cb,	/* I - Callback function or MXML_NO_CALLBACK */
              mxml_sax_cb_t  sax_cb,	/* I - SAX callback or MXML_NO_CALLBACK */
              void           *sax_data)	/* I - SAX user data */
{
  _mxml_fdbuf_t	buf;			/* File descriptor buffer */


 /*
  * Initialize the file descriptor buffer...
  */

  buf.fd      = fd;
  buf.current = buf.buffer;
  buf.end     = buf.buffer;

 /*
  * Read the XML data...
  */

  return (mxml_load_data(top, &buf, cb, mxml_fd_getc, sax_cb, sax_data));
}


/*
 * 'mxmlSAXLoadFile()' - Load a file into an XML node tree
 *                       using a SAX callback.
 *
 * The nodes in the specified file are added to the specified top node.
 * If no top node is provided, the XML file MUST be well-formed with a
 * single parent node like <?xml> for the entire file. The callback
 * function returns the value type that should be used for child nodes.
 * If MXML_NO_CALLBACK is specified then all child nodes will be either
 * MXML_ELEMENT or MXML_TEXT nodes.
 *
 * The constants MXML_INTEGER_CALLBACK, MXML_OPAQUE_CALLBACK,
 * MXML_REAL_CALLBACK, and MXML_TEXT_CALLBACK are defined for loading
 * child nodes of the specified type.
 *
 * The SAX callback must call mxmlRetain() for any nodes that need to
 * be kept for later use. Otherwise, nodes are deleted when the parent
 * node is closed or after each data, comment, CDATA, or directive node.
 *
 * @since Mini-XML 2.3@
 */

mxml_node_t *				/* O - First node or NULL if the file could not be read. */
mxmlSAXLoadFile(
    mxml_node_t    *top,		/* I - Top node */
    FILE           *fp,			/* I - File to read from */
    mxml_load_cb_t cb,			/* I - Callback function or MXML_NO_CALLBACK */
    mxml_sax_cb_t  sax_cb,		/* I - SAX callback or MXML_NO_CALLBACK */
    void           *sax_data)		/* I - SAX user data */
{
 /*
  * Read the XML data...
  */

  return (mxml_load_data(top, fp, cb, mxml_file_getc, sax_cb, sax_data));
}


/*
 * 'mxmlSAXLoadString()' - Load a string into an XML node tree
 *                         using a SAX callback.
 *
 * The nodes in the specified string are added to the specified top node.
 * If no top node is provided, the XML string MUST be well-formed with a
 * single parent node like <?xml> for the entire string. The callback
 * function returns the value type that should be used for child nodes.
 * If MXML_NO_CALLBACK is specified then all child nodes will be either
 * MXML_ELEMENT or MXML_TEXT nodes.
 *
 * The constants MXML_INTEGER_CALLBACK, MXML_OPAQUE_CALLBACK,
 * MXML_REAL_CALLBACK, and MXML_TEXT_CALLBACK are defined for loading
 * child nodes of the specified type.
 *
 * The SAX callback must call mxmlRetain() for any nodes that need to
 * be kept for later use. Otherwise, nodes are deleted when the parent
 * node is closed or after each data, comment, CDATA, or directive node.
 *
 * @since Mini-XML 2.3@
 */

mxml_node_t *				/* O - First node or NULL if the string has errors. */
mxmlSAXLoadString(
    mxml_node_t    *top,		/* I - Top node */
    const char     *s,			/* I - String to load */
    mxml_load_cb_t cb,			/* I - Callback function or MXML_NO_CALLBACK */
    mxml_sax_cb_t  sax_cb,		/* I - SAX callback or MXML_NO_CALLBACK */
    void           *sax_data)		/* I - SAX user data */
{
 /*
  * Read the XML data...
  */

  return (mxml_load_data(top, (void *)&s, cb, mxml_string_getc, sax_cb, sax_data));
}


/*
 * 'mxmlSetCustomHandlers()' - Set the handling functions for custom data.
 *
 * The load function accepts a node pointer and a data string and must
 * return 0 on success and non-zero on error.
 *
 * The save function accepts a node pointer and must return a malloc'd
 * string on success and NULL on error.
 * 
 */

void
mxmlSetCustomHandlers(
    mxml_custom_load_cb_t load,		/* I - Load function */
    mxml_custom_save_cb_t save)		/* I - Save function */
{
  _mxml_global_t *global = _mxml_global();
					/* Global data */


  global->custom_load_cb = load;
  global->custom_save_cb = save;
}


/*
 * 'mxmlSetErrorCallback()' - Set the error message callback.
 */

void
mxmlSetErrorCallback(mxml_error_cb_t cb)/* I - Error callback function */
{