mxml-file.c

asterisk 是一个很有知名度开源软件

/*
 * "$Id: mxml-file.c 22267 2006-04-24 17:11:45Z kpfleming $"
 *
 * File loading code for Mini-XML, a small XML-like file parsing library.
 *
 * Copyright 2003-2005 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.
 *   mxmlSetCustomHandlers() - Set the handling functions for custom data.
 *   mxmlSetErrorCallback()  - Set the error message callback.
 *   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 "config.h"
#include "mxml.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')


/*
 * Structures...
 */

typedef struct mxml_fdbuf_s		/**** File descriptor buffer (@private) ****/
{
  int		fd;			/* File descriptor */
  unsigned char	*current,		/* Current position in buffer */
		*end,			/* End of buffer */
		buffer[8192];		/* Character buffer */
} mxml_fdbuf_t;


/*
 * Global error handler...
 */

extern void	(*mxml_error_cb)(const char *);


/*
 * Custom data handlers...
 */

static mxml_custom_load_cb_t	mxml_custom_load_cb = NULL;
static mxml_custom_save_cb_t	mxml_custom_save_cb = NULL;


/*
 * 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,
					int (*getc_cb)(void *, int *));
static mxml_node_t	*mxml_load_data(mxml_node_t *top, void *p,
			                mxml_type_t (*cb)(mxml_node_t *),
			                int (*getc_cb)(void *, int *));
static int		mxml_parse_element(mxml_node_t *node, void *p,
			                   int *encoding,
					   int (*getc_cb)(void *, int *));
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,
					int (*putc_cb)(int, void *));
static int		mxml_write_node(mxml_node_t *node, void *p,
			                const char *(*cb)(mxml_node_t *, int),
					int col,
					int (*putc_cb)(int, void *));
static int		mxml_write_string(const char *s, void *p,
					  int (*putc_cb)(int, void *));
static int		mxml_write_ws(mxml_node_t *node, void *p, 
			              const char *(*cb)(mxml_node_t *, int), int ws,
				      int col, int (*putc_cb)(int, void *));


/*
 * '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_type_t (*cb)(mxml_node_t *node))
					/* 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));
}


/*
 * '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_type_t (*cb)(mxml_node_t *node))
					/* I - Callback function or MXML_NO_CALLBACK */
{
 /*
  * Read the XML data...
  */

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


/*
 * '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_type_t (*cb)(mxml_node_t *node))
					/* I - Callback function or MXML_NO_CALLBACK */
{
 /*
  * Read the XML data...
  */

  return (mxml_load_data(top, &s, cb, mxml_string_getc));
}


/*
 * '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 */
                    const char  *(*cb)(mxml_node_t *node, int ws))
					/* 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 */
	   const char  *(*cb)(mxml_node_t *node, int ws))
					/* I - Whitespace callback or MXML_NO_CALLBACK */
{
  int		col;			/* Final column */
  mxml_fdbuf_t	buf;			/* File descriptor buffer */


 /*
  * 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)) < 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 */
	     const char  *(*cb)(mxml_node_t *node, int ws))
					/* I - Whitespace callback or MXML_NO_CALLBACK */
{
  int	col;				/* Final column */


 /*
  * Write the node...
  */

  if ((col = mxml_write_node(node, fp, cb, 0, mxml_file_putc)) < 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 */
               const char  *(*cb)(mxml_node_t *node, int ws))
					/* I - Whitespace callback or MXML_NO_CALLBACK */
{
  int	col;				/* Final column */
  char	*ptr[2];			/* Pointers for putc_cb */


 /*
  * Write the node...
  */

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

  if ((col = mxml_write_node(node, ptr, cb, 0, mxml_string_putc)) < 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);
}


/*
 * '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_custom_load_cb = load;
  mxml_custom_save_cb = save;
}


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

void
mxmlSetErrorCallback(void (*cb)(const char *))
					/* I - Error callback function */
{
  mxml_error_cb = cb;
}


/*
 * 'mxml_add_char()' - Add a character to a buffer, expanding as needed.
 */

static int				/* O  - 0 on success, -1 on error */
mxml_add_char(int  ch,			/* I  - Character to add */
              char **bufptr,		/* IO - Current position in buffer */
	      char **buffer,		/* IO - Current buffer */
	      int  *bufsize)		/* IO - Current buffer size */
{
  char	*newbuffer;			/* New buffer value */


  if (*bufptr >= (*buffer + *bufsize - 4))
  {
   /*
    * Increase the size of the buffer...
    */

    if (*bufsize < 1024)
      (*bufsize) *= 2;
    else
      (*bufsize) += 1024;

    if ((newbuffer = realloc(*buffer, *bufsize)) == NULL)
    {
      free(*buffer);

      mxml_error("Unable to expand string buffer to %d bytes!", *bufsize);

      return (-1);
    }

    *bufptr = newbuffer + (*bufptr - *buffer);
    *buffer = newbuffer;
  }

  if (ch < 0x80)
  {
   /*
    * Single byte ASCII...
    */

    *(*bufptr)++ = ch;
  }
  else if (ch < 0x800)
  {
   /*
    * Two-byte UTF-8...
    */

    *(*bufptr)++ = 0xc0 | (ch >> 6);
    *(*bufptr)++ = 0x80 | (ch & 0x3f);
  }
  else if (ch < 0x10000)
  {
   /*
    * Three-byte UTF-8...
    */

    *(*bufptr)++ = 0xe0 | (ch >> 12);
    *(*bufptr)++ = 0x80 | ((ch >> 6) & 0x3f);
    *(*bufptr)++ = 0x80 | (ch & 0x3f);
  }
  else
  {
   /*
    * Four-byte UTF-8...
    */

    *(*bufptr)++ = 0xf0 | (ch >> 18);
    *(*bufptr)++ = 0x80 | ((ch >> 12) & 0x3f);
    *(*bufptr)++ = 0x80 | ((ch >> 6) & 0x3f);
    *(*bufptr)++ = 0x80 | (ch & 0x3f);
  }

  return (0);
}


/*
 * 'mxml_fd_getc()' - Read a character from a file descriptor.
 */

static int				/* O  - Character or EOF */
mxml_fd_getc(void *p,			/* I  - File descriptor buffer */
             int  *encoding)		/* IO - Encoding */
{
  mxml_fdbuf_t	*buf;			/* File descriptor buffer */
  int		ch,			/* Current character */
		temp;			/* Temporary character */


 /*
  * Grab the next character in the buffer...
  */

  buf = (mxml_fdbuf_t *)p;

  if (buf->current >= buf->end)
    if (mxml_fd_read(buf) < 0)
      return (EOF);

  ch = *(buf->current)++;

  switch (*encoding)
  {
    case ENCODE_UTF8 :
       /*
	* Got a UTF-8 character; convert UTF-8 to Unicode and return...
	*/

	if (!(ch & 0x80))
	{
#if DEBUG > 1
          printf("mxml_fd_getc: %c (0x%04x)\n", ch < ' ' ? '.' : ch, ch);
#endif /* DEBUG > 1 */

	  if (mxml_bad_char(ch))
	  {
	    mxml_error("Bad control character 0x%02x not allowed by XML standard!",
        	       ch);
	    return (EOF);
	  }

	  return (ch);
        }
	else if (ch == 0xfe)
	{
	 /*
	  * UTF-16 big-endian BOM?
	  */

	  if (buf->current >= buf->end)
	    if (mxml_fd_read(buf) < 0)
	      return (EOF);

	  ch = *(buf->current)++;
          
	  if (ch != 0xff)
	    return (EOF);

	  *encoding = ENCODE_UTF16BE;

	  return (mxml_fd_getc(p, encoding));
	}
	else if (ch == 0xff)
	{
	 /*
	  * UTF-16 little-endian BOM?
	  */

	  if (buf->current >= buf->end)
	    if (mxml_fd_read(buf) < 0)
	      return (EOF);

	  ch = *(buf->current)++;
          
	  if (ch != 0xfe)
	    return (EOF);

	  *encoding = ENCODE_UTF16LE;

	  return (mxml_fd_getc(p, encoding));
	}
	else if ((ch & 0xe0) == 0xc0)
	{
	 /*
	  * Two-byte value...
	  */

	  if (buf->current >= buf->end)
	    if (mxml_fd_read(buf) < 0)
	      return (EOF);

	  temp = *(buf->current)++;

	  if ((temp & 0xc0) != 0x80)
	    return (EOF);

	  ch = ((ch & 0x1f) << 6) | (temp & 0x3f);

	  if (ch < 0x80)
	    return (EOF);
	}
	else if ((ch & 0xf0) == 0xe0)
	{
	 /*
	  * Three-byte value...
	  */

	  if (buf->current >= buf->end)
	    if (mxml_fd_read(buf) < 0)
	      return (EOF);

	  temp = *(buf->current)++;

	  if ((temp & 0xc0) != 0x80)
	    return (EOF);

	  ch = ((ch & 0x0f) << 6) | (temp & 0x3f);

	  if (buf->current >= buf->end)
	    if (mxml_fd_read(buf) < 0)
	      return (EOF);

	  temp = *(buf->current)++;

	  if ((temp & 0xc0) != 0x80)
	    return (EOF);

	  ch = (ch << 6) | (temp & 0x3f);

	  if (ch < 0x800)
	    return (EOF);
	}
	else if ((ch & 0xf8) == 0xf0)
	{
	 /*
	  * Four-byte value...
	  */

	  if (buf->current >= buf->end)
	    if (mxml_fd_read(buf) < 0)
	      return (EOF);

	  temp = *(buf->current)++;

	  if ((temp & 0xc0) != 0x80)
	    return (EOF);

	  ch = ((ch & 0x07) << 6) | (temp & 0x3f);

	  if (buf->current >= buf->end)
	    if (mxml_fd_read(buf) < 0)
	      return (EOF);