xmltok.c

一个用visual studio C++开发的XML文件解析和生成程序.

/*
Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
See the file copying.txt for copying permission.
*/

#include "xmldef.h"
#include "xmltok.h"
#include "nametab.h"

#ifdef XML_DTD
#define IGNORE_SECTION_TOK_VTABLE , PREFIX(ignoreSectionTok)
#else
#define IGNORE_SECTION_TOK_VTABLE /* as nothing */
#endif

#define VTABLE1 \
  { PREFIX(prologTok), PREFIX(contentTok), \
    PREFIX(cdataSectionTok) IGNORE_SECTION_TOK_VTABLE }, \
  { PREFIX(attributeValueTok), PREFIX(entityValueTok) }, \
  PREFIX(sameName), \
  PREFIX(nameMatchesAscii), \
  PREFIX(nameLength), \
  PREFIX(skipS), \
  PREFIX(getAtts), \
  PREFIX(charRefNumber), \
  PREFIX(predefinedEntityName), \
  PREFIX(updatePosition), \
  PREFIX(isPublicId)

#define VTABLE VTABLE1, PREFIX(toUtf8), PREFIX(toUtf16)

#define UCS2_GET_NAMING(pages, hi, lo) \
   (namingBitmap[(pages[hi] << 3) + ((lo) >> 5)] & (1 << ((lo) & 0x1F)))

/* A 2 byte UTF-8 representation splits the characters 11 bits
between the bottom 5 and 6 bits of the bytes.
We need 8 bits to index into pages, 3 bits to add to that index and
5 bits to generate the mask. */
#define UTF8_GET_NAMING2(pages, byte) \
    (namingBitmap[((pages)[(((byte)[0]) >> 2) & 7] << 3) \
                      + ((((byte)[0]) & 3) << 1) \
                      + ((((byte)[1]) >> 5) & 1)] \
         & (1 << (((byte)[1]) & 0x1F)))

/* A 3 byte UTF-8 representation splits the characters 16 bits
between the bottom 4, 6 and 6 bits of the bytes.
We need 8 bits to index into pages, 3 bits to add to that index and
5 bits to generate the mask. */
#define UTF8_GET_NAMING3(pages, byte) \
  (namingBitmap[((pages)[((((byte)[0]) & 0xF) << 4) \
                             + ((((byte)[1]) >> 2) & 0xF)] \
		       << 3) \
                      + ((((byte)[1]) & 3) << 1) \
                      + ((((byte)[2]) >> 5) & 1)] \
         & (1 << (((byte)[2]) & 0x1F)))

#define UTF8_GET_NAMING(pages, p, n) \
  ((n) == 2 \
  ? UTF8_GET_NAMING2(pages, (const unsigned char *)(p)) \
  : ((n) == 3 \
     ? UTF8_GET_NAMING3(pages, (const unsigned char *)(p)) \
     : 0))

#define UTF8_INVALID3(p) \
  ((*p) == 0xED \
  ? (((p)[1] & 0x20) != 0) \
  : ((*p) == 0xEF \
     ? ((p)[1] == 0xBF && ((p)[2] == 0xBF || (p)[2] == 0xBE)) \
     : 0))

#define UTF8_INVALID4(p) ((*p) == 0xF4 && ((p)[1] & 0x30) != 0)

static
int isNever(const ENCODING *enc, const char *p)
{
  return 0;
}

static
int utf8_isName2(const ENCODING *enc, const char *p)
{
  return UTF8_GET_NAMING2(namePages, (const unsigned char *)p);
}

static
int utf8_isName3(const ENCODING *enc, const char *p)
{
  return UTF8_GET_NAMING3(namePages, (const unsigned char *)p);
}

#define utf8_isName4 isNever

static
int utf8_isNmstrt2(const ENCODING *enc, const char *p)
{
  return UTF8_GET_NAMING2(nmstrtPages, (const unsigned char *)p);
}

static
int utf8_isNmstrt3(const ENCODING *enc, const char *p)
{
  return UTF8_GET_NAMING3(nmstrtPages, (const unsigned char *)p);
}

#define utf8_isNmstrt4 isNever

#define utf8_isInvalid2 isNever

static
int utf8_isInvalid3(const ENCODING *enc, const char *p)
{
  return UTF8_INVALID3((const unsigned char *)p);
}

static
int utf8_isInvalid4(const ENCODING *enc, const char *p)
{
  return UTF8_INVALID4((const unsigned char *)p);
}

struct normal_encoding {
  ENCODING enc;
  unsigned char type[256];
#ifdef XML_MIN_SIZE
  int (*byteType)(const ENCODING *, const char *);
  int (*isNameMin)(const ENCODING *, const char *);
  int (*isNmstrtMin)(const ENCODING *, const char *);
  int (*byteToAscii)(const ENCODING *, const char *);
  int (*charMatches)(const ENCODING *, const char *, int);
#endif /* XML_MIN_SIZE */
  int (*isName2)(const ENCODING *, const char *);
  int (*isName3)(const ENCODING *, const char *);
  int (*isName4)(const ENCODING *, const char *);
  int (*isNmstrt2)(const ENCODING *, const char *);
  int (*isNmstrt3)(const ENCODING *, const char *);
  int (*isNmstrt4)(const ENCODING *, const char *);
  int (*isInvalid2)(const ENCODING *, const char *);
  int (*isInvalid3)(const ENCODING *, const char *);
  int (*isInvalid4)(const ENCODING *, const char *);
};

#ifdef XML_MIN_SIZE

#define STANDARD_VTABLE(E) \
 E ## byteType, \
 E ## isNameMin, \
 E ## isNmstrtMin, \
 E ## byteToAscii, \
 E ## charMatches,

#else

#define STANDARD_VTABLE(E) /* as nothing */

#endif

#define NORMAL_VTABLE(E) \
 E ## isName2, \
 E ## isName3, \
 E ## isName4, \
 E ## isNmstrt2, \
 E ## isNmstrt3, \
 E ## isNmstrt4, \
 E ## isInvalid2, \
 E ## isInvalid3, \
 E ## isInvalid4

static int checkCharRefNumber(int);

#include "xmltok_impl.h"
#include "ascii.h"

#ifdef XML_MIN_SIZE
#define sb_isNameMin isNever
#define sb_isNmstrtMin isNever
#endif

#ifdef XML_MIN_SIZE
#define MINBPC(enc) ((enc)->minBytesPerChar)
#else
/* minimum bytes per character */
#define MINBPC(enc) 1
#endif

#define SB_BYTE_TYPE(enc, p) \
  (((struct normal_encoding *)(enc))->type[(unsigned char)*(p)])

#ifdef XML_MIN_SIZE
static
int sb_byteType(const ENCODING *enc, const char *p)
{
  return SB_BYTE_TYPE(enc, p);
}
#define BYTE_TYPE(enc, p) \
 (((const struct normal_encoding *)(enc))->byteType(enc, p))
#else
#define BYTE_TYPE(enc, p) SB_BYTE_TYPE(enc, p)
#endif

#ifdef XML_MIN_SIZE
#define BYTE_TO_ASCII(enc, p) \
 (((const struct normal_encoding *)(enc))->byteToAscii(enc, p))
static
int sb_byteToAscii(const ENCODING *enc, const char *p)
{
  return *p;
}
#else
#define BYTE_TO_ASCII(enc, p) (*(p))
#endif

#define IS_NAME_CHAR(enc, p, n) \
 (((const struct normal_encoding *)(enc))->isName ## n(enc, p))
#define IS_NMSTRT_CHAR(enc, p, n) \
 (((const struct normal_encoding *)(enc))->isNmstrt ## n(enc, p))
#define IS_INVALID_CHAR(enc, p, n) \
 (((const struct normal_encoding *)(enc))->isInvalid ## n(enc, p))

#ifdef XML_MIN_SIZE
#define IS_NAME_CHAR_MINBPC(enc, p) \
 (((const struct normal_encoding *)(enc))->isNameMin(enc, p))
#define IS_NMSTRT_CHAR_MINBPC(enc, p) \
 (((const struct normal_encoding *)(enc))->isNmstrtMin(enc, p))
#else
#define IS_NAME_CHAR_MINBPC(enc, p) (0)
#define IS_NMSTRT_CHAR_MINBPC(enc, p) (0)
#endif

#ifdef XML_MIN_SIZE
#define CHAR_MATCHES(enc, p, c) \
 (((const struct normal_encoding *)(enc))->charMatches(enc, p, c))
static
int sb_charMatches(const ENCODING *enc, const char *p, int c)
{
  return *p == c;
}
#else
/* c is an ASCII character */
#define CHAR_MATCHES(enc, p, c) (*(p) == c)
#endif

#define PREFIX(ident) normal_ ## ident
#include "xmltok_impl.c"

#undef MINBPC
#undef BYTE_TYPE
#undef BYTE_TO_ASCII
#undef CHAR_MATCHES
#undef IS_NAME_CHAR
#undef IS_NAME_CHAR_MINBPC
#undef IS_NMSTRT_CHAR
#undef IS_NMSTRT_CHAR_MINBPC
#undef IS_INVALID_CHAR

enum {  /* UTF8_cvalN is value of masked first byte of N byte sequence */
  UTF8_cval1 = 0x00,
  UTF8_cval2 = 0xc0,
  UTF8_cval3 = 0xe0,
  UTF8_cval4 = 0xf0
};

static
void utf8_toUtf8(const ENCODING *enc,
		 const char **fromP, const char *fromLim,
		 char **toP, const char *toLim)
{
  char *to;
  const char *from;
  if (fromLim - *fromP > toLim - *toP) {
    /* Avoid copying partial characters. */
    for (fromLim = *fromP + (toLim - *toP); fromLim > *fromP; fromLim--)
      if (((unsigned char)fromLim[-1] & 0xc0) != 0x80)
	break;
  }
  for (to = *toP, from = *fromP; from != fromLim; from++, to++)
    *to = *from;
  *fromP = from;
  *toP = to;
}

static
void utf8_toUtf16(const ENCODING *enc,
		  const char **fromP, const char *fromLim,
		  unsigned short **toP, const unsigned short *toLim)
{
  unsigned short *to = *toP;
  const char *from = *fromP;
  while (from != fromLim && to != toLim) {
    switch (((struct normal_encoding *)enc)->type[(unsigned char)*from]) {
    case BT_LEAD2:
      *to++ = ((from[0] & 0x1f) << 6) | (from[1] & 0x3f);
      from += 2;
      break;
    case BT_LEAD3:
      *to++ = ((from[0] & 0xf) << 12) | ((from[1] & 0x3f) << 6) | (from[2] & 0x3f);
      from += 3;
      break;
    case BT_LEAD4:
      {
	unsigned long n;
	if (to + 1 == toLim)
	  break;
	n = ((from[0] & 0x7) << 18) | ((from[1] & 0x3f) << 12) | ((from[2] & 0x3f) << 6) | (from[3] & 0x3f);
	n -= 0x10000;
	to[0] = (unsigned short)((n >> 10) | 0xD800);
	to[1] = (unsigned short)((n & 0x3FF) | 0xDC00);
	to += 2;
	from += 4;
      }
      break;
    default:
      *to++ = *from++;
      break;
    }
  }
  *fromP = from;
  *toP = to;
}

#ifdef XML_NS
static const struct normal_encoding utf8_encoding_ns = {
  { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
  {
#include "asciitab.h"
#include "utf8tab.h"
  },
  STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
};
#endif

static const struct normal_encoding utf8_encoding = {
  { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
  {
#define BT_COLON BT_NMSTRT
#include "asciitab.h"
#undef BT_COLON
#include "utf8tab.h"
  },
  STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
};

#ifdef XML_NS

static const struct normal_encoding internal_utf8_encoding_ns = {
  { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
  {
#include "iasciitab.h"
#include "utf8tab.h"
  },
  STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
};

#endif

static const struct normal_encoding internal_utf8_encoding = {
  { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
  {
#define BT_COLON BT_NMSTRT
#include "iasciitab.h"
#undef BT_COLON
#include "utf8tab.h"
  },
  STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
};

static
void latin1_toUtf8(const ENCODING *enc,
		   const char **fromP, const char *fromLim,
		   char **toP, const char *toLim)
{
  for (;;) {
    unsigned char c;
    if (*fromP == fromLim)
      break;
    c = (unsigned char)**fromP;
    if (c & 0x80) {
      if (toLim - *toP < 2)
	break;
      *(*toP)++ = ((c >> 6) | UTF8_cval2);
      *(*toP)++ = ((c & 0x3f) | 0x80);
      (*fromP)++;
    }
    else {
      if (*toP == toLim)
	break;
      *(*toP)++ = *(*fromP)++;
    }
  }
}

static
void latin1_toUtf16(const ENCODING *enc,
		    const char **fromP, const char *fromLim,
		    unsigned short **toP, const unsigned short *toLim)
{
  while (*fromP != fromLim && *toP != toLim)
    *(*toP)++ = (unsigned char)*(*fromP)++;
}

#ifdef XML_NS

static const struct normal_encoding latin1_encoding_ns = {
  { VTABLE1, latin1_toUtf8, latin1_toUtf16, 1, 0, 0 },
  {
#include "asciitab.h"
#include "latin1tab.h"
  },
  STANDARD_VTABLE(sb_)
};

#endif

static const struct normal_encoding latin1_encoding = {
  { VTABLE1, latin1_toUtf8, latin1_toUtf16, 1, 0, 0 },
  {
#define BT_COLON BT_NMSTRT
#include "asciitab.h"
#undef BT_COLON
#include "latin1tab.h"
  },
  STANDARD_VTABLE(sb_)
};

static
void ascii_toUtf8(const ENCODING *enc,
		  const char **fromP, const char *fromLim,
		  char **toP, const char *toLim)
{
  while (*fromP != fromLim && *toP != toLim)
    *(*toP)++ = *(*fromP)++;
}

#ifdef XML_NS

static const struct normal_encoding ascii_encoding_ns = {
  { VTABLE1, ascii_toUtf8, latin1_toUtf16, 1, 1, 0 },
  {
#include "asciitab.h"
/* BT_NONXML == 0 */
  },
  STANDARD_VTABLE(sb_)
};

#endif

static const struct normal_encoding ascii_encoding = {
  { VTABLE1, ascii_toUtf8, latin1_toUtf16, 1, 1, 0 },
  {
#define BT_COLON BT_NMSTRT
#include "asciitab.h"
#undef BT_COLON
/* BT_NONXML == 0 */
  },
  STANDARD_VTABLE(sb_)
};

static int unicode_byte_type(char hi, char lo)
{
  switch ((unsigned char)hi) {
  case 0xD8: case 0xD9: case 0xDA: case 0xDB:
    return BT_LEAD4;
  case 0xDC: case 0xDD: case 0xDE: case 0xDF:
    return BT_TRAIL;
  case 0xFF:
    switch ((unsigned char)lo) {
    case 0xFF:
    case 0xFE:
      return BT_NONXML;
    }
    break;
  }
  return BT_NONASCII;
}

#define DEFINE_UTF16_TO_UTF8(E) \
static \
void E ## toUtf8(const ENCODING *enc, \
		 const char **fromP, const char *fromLim, \
		 char **toP, const char *toLim) \
{ \
  const char *from; \
  for (from = *fromP; from != fromLim; from += 2) { \
    int plane; \
    unsigned char lo2; \
    unsigned char lo = GET_LO(from); \
    unsigned char hi = GET_HI(from); \
    switch (hi) { \
    case 0: \
      if (lo < 0x80) { \
        if (*toP == toLim) { \
          *fromP = from; \
	  return; \
        } \
        *(*toP)++ = lo; \
        break; \
      } \
      /* fall through */ \
    case 0x1: case 0x2: case 0x3: \
    case 0x4: case 0x5: case 0x6: case 0x7: \
      if (toLim -  *toP < 2) { \
        *fromP = from; \
	return; \
      } \
      *(*toP)++ = ((lo >> 6) | (hi << 2) |  UTF8_cval2); \
      *(*toP)++ = ((lo & 0x3f) | 0x80); \
      break; \
    default: \
      if (toLim -  *toP < 3)  { \
        *fromP = from; \
	return; \
      } \
      /* 16 bits divided 4, 6, 6 amongst 3 bytes */ \
      *(*toP)++ = ((hi >> 4) | UTF8_cval3); \
      *(*toP)++ = (((hi & 0xf) << 2) | (lo >> 6) | 0x80); \
      *(*toP)++ = ((lo & 0x3f) | 0x80); \
      break; \
    case 0xD8: case 0xD9: case 0xDA: case 0xDB: \
      if (toLim -  *toP < 4) { \
	*fromP = from; \
	return; \