You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@apr.apache.org by rj...@apache.org on 2010/10/08 22:01:11 UTC
svn commit: r1005995 [4/5] - in /apr/apr-util/branches/0.9.x: build/
xml/expat/ xml/expat/conftools/ xml/expat/lib/ xml/expat/win32/
Modified: apr/apr-util/branches/0.9.x/xml/expat/lib/xmltok.c
URL: http://svn.apache.org/viewvc/apr/apr-util/branches/0.9.x/xml/expat/lib/xmltok.c?rev=1005995&r1=1005994&r2=1005995&view=diff
==============================================================================
--- apr/apr-util/branches/0.9.x/xml/expat/lib/xmltok.c (original)
+++ apr/apr-util/branches/0.9.x/xml/expat/lib/xmltok.c Fri Oct 8 20:01:10 2010
@@ -1,17 +1,18 @@
-/*
-Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
-See the file COPYING for copying permission.
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
*/
-static char RCSId[]
- = "$Header: /home/cvs/apr-util/xml/expat/lib/xmltok.c,v 1.1 2001/02/28 14:41:26 gstein Exp $";
-
#ifdef COMPILED_FROM_DSP
-# include "winconfig.h"
+#include "winconfig.h"
+#elif defined(MACOS_CLASSIC)
+#include "macconfig.h"
#else
-# include <config.h>
+#ifdef HAVE_EXPAT_CONFIG_H
+#include <expat_config.h>
+#endif
#endif /* ndef COMPILED_FROM_DSP */
+#include "internal.h"
#include "xmltok.h"
#include "nametab.h"
@@ -40,24 +41,25 @@ static char RCSId[]
#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. */
+/* 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. */
+/* 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) \
+ << 3) \
+ ((((byte)[1]) & 3) << 1) \
+ ((((byte)[2]) >> 5) & 1)] \
& (1 << (((byte)[2]) & 0x1F)))
@@ -69,59 +71,97 @@ We need 8 bits to index into pages, 3 bi
? 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))
+/* Detection of invalid UTF-8 sequences is based on Table 3.1B
+ of Unicode 3.2: http://www.unicode.org/unicode/reports/tr28/
+ with the additional restriction of not allowing the Unicode
+ code points 0xFFFF and 0xFFFE (sequences EF,BF,BF and EF,BF,BE).
+ Implementation details:
+ (A & 0x80) == 0 means A < 0x80
+ and
+ (A & 0xC0) == 0xC0 means A > 0xBF
+*/
+
+#define UTF8_INVALID2(p) \
+ ((*p) < 0xC2 || ((p)[1] & 0x80) == 0 || ((p)[1] & 0xC0) == 0xC0)
-#define UTF8_INVALID4(p) ((*p) == 0xF4 && ((p)[1] & 0x30) != 0)
+#define UTF8_INVALID3(p) \
+ (((p)[2] & 0x80) == 0 \
+ || \
+ ((*p) == 0xEF && (p)[1] == 0xBF \
+ ? \
+ (p)[2] > 0xBD \
+ : \
+ ((p)[2] & 0xC0) == 0xC0) \
+ || \
+ ((*p) == 0xE0 \
+ ? \
+ (p)[1] < 0xA0 || ((p)[1] & 0xC0) == 0xC0 \
+ : \
+ ((p)[1] & 0x80) == 0 \
+ || \
+ ((*p) == 0xED ? (p)[1] > 0x9F : ((p)[1] & 0xC0) == 0xC0)))
+
+#define UTF8_INVALID4(p) \
+ (((p)[3] & 0x80) == 0 || ((p)[3] & 0xC0) == 0xC0 \
+ || \
+ ((p)[2] & 0x80) == 0 || ((p)[2] & 0xC0) == 0xC0 \
+ || \
+ ((*p) == 0xF0 \
+ ? \
+ (p)[1] < 0x90 || ((p)[1] & 0xC0) == 0xC0 \
+ : \
+ ((p)[1] & 0x80) == 0 \
+ || \
+ ((*p) == 0xF4 ? (p)[1] > 0x8F : ((p)[1] & 0xC0) == 0xC0)))
-static
-int isNever(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+isNever(const ENCODING *enc, const char *p)
{
return 0;
}
-static
-int utf8_isName2(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+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)
+static int PTRFASTCALL
+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)
+static int PTRFASTCALL
+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)
+static int PTRFASTCALL
+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 PTRFASTCALL
+utf8_isInvalid2(const ENCODING *enc, const char *p)
+{
+ return UTF8_INVALID2((const unsigned char *)p);
+}
-static
-int utf8_isInvalid3(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+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)
+static int PTRFASTCALL
+utf8_isInvalid4(const ENCODING *enc, const char *p)
{
return UTF8_INVALID4((const unsigned char *)p);
}
@@ -130,23 +170,25 @@ 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);
+ int (PTRFASTCALL *byteType)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNameMin)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNmstrtMin)(const ENCODING *, const char *);
+ int (PTRFASTCALL *byteToAscii)(const ENCODING *, const char *);
+ int (PTRCALL *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 *);
+ int (PTRFASTCALL *isName2)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isName3)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isName4)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNmstrt2)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNmstrt3)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNmstrt4)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isInvalid2)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isInvalid3)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isInvalid4)(const ENCODING *, const char *);
};
+#define AS_NORMAL_ENCODING(enc) ((const struct normal_encoding *) (enc))
+
#ifdef XML_MIN_SIZE
#define STANDARD_VTABLE(E) \
@@ -173,7 +215,7 @@ struct normal_encoding {
E ## isInvalid3, \
E ## isInvalid4
-static int checkCharRefNumber(int);
+static int FASTCALL checkCharRefNumber(int);
#include "xmltok_impl.h"
#include "ascii.h"
@@ -194,22 +236,22 @@ static int checkCharRefNumber(int);
(((struct normal_encoding *)(enc))->type[(unsigned char)*(p)])
#ifdef XML_MIN_SIZE
-static
-int sb_byteType(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+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))
+ (AS_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)
+ (AS_NORMAL_ENCODING(enc)->byteToAscii(enc, p))
+static int PTRFASTCALL
+sb_byteToAscii(const ENCODING *enc, const char *p)
{
return *p;
}
@@ -218,17 +260,17 @@ int sb_byteToAscii(const ENCODING *enc,
#endif
#define IS_NAME_CHAR(enc, p, n) \
- (((const struct normal_encoding *)(enc))->isName ## n(enc, p))
+ (AS_NORMAL_ENCODING(enc)->isName ## n(enc, p))
#define IS_NMSTRT_CHAR(enc, p, n) \
- (((const struct normal_encoding *)(enc))->isNmstrt ## n(enc, p))
+ (AS_NORMAL_ENCODING(enc)->isNmstrt ## n(enc, p))
#define IS_INVALID_CHAR(enc, p, n) \
- (((const struct normal_encoding *)(enc))->isInvalid ## n(enc, p))
+ (AS_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))
+ (AS_NORMAL_ENCODING(enc)->isNameMin(enc, p))
#define IS_NMSTRT_CHAR_MINBPC(enc, p) \
- (((const struct normal_encoding *)(enc))->isNmstrtMin(enc, p))
+ (AS_NORMAL_ENCODING(enc)->isNmstrtMin(enc, p))
#else
#define IS_NAME_CHAR_MINBPC(enc, p) (0)
#define IS_NMSTRT_CHAR_MINBPC(enc, p) (0)
@@ -236,9 +278,9 @@ int sb_byteToAscii(const ENCODING *enc,
#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)
+ (AS_NORMAL_ENCODING(enc)->charMatches(enc, p, c))
+static int PTRCALL
+sb_charMatches(const ENCODING *enc, const char *p, int c)
{
return *p == c;
}
@@ -267,10 +309,10 @@ enum { /* UTF8_cvalN is value of masked
UTF8_cval4 = 0xf0
};
-static
-void utf8_toUtf8(const ENCODING *enc,
- const char **fromP, const char *fromLim,
- char **toP, const char *toLim)
+static void PTRCALL
+utf8_toUtf8(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ char **toP, const char *toLim)
{
char *to;
const char *from;
@@ -278,7 +320,7 @@ void utf8_toUtf8(const ENCODING *enc,
/* Avoid copying partial characters. */
for (fromLim = *fromP + (toLim - *toP); fromLim > *fromP; fromLim--)
if (((unsigned char)fromLim[-1] & 0xc0) != 0x80)
- break;
+ break;
}
for (to = *toP, from = *fromP; from != fromLim; from++, to++)
*to = *from;
@@ -286,34 +328,36 @@ void utf8_toUtf8(const ENCODING *enc,
*toP = to;
}
-static
-void utf8_toUtf16(const ENCODING *enc,
- const char **fromP, const char *fromLim,
- unsigned short **toP, const unsigned short *toLim)
+static void PTRCALL
+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);
+ *to++ = (unsigned short)(((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);
+ *to++ = (unsigned short)(((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;
+ unsigned long n;
+ if (to + 1 == toLim)
+ goto after;
+ 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:
@@ -321,6 +365,7 @@ void utf8_toUtf16(const ENCODING *enc,
break;
}
}
+after:
*fromP = from;
*toP = to;
}
@@ -371,10 +416,10 @@ static const struct normal_encoding inte
STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
};
-static
-void latin1_toUtf8(const ENCODING *enc,
- const char **fromP, const char *fromLim,
- char **toP, const char *toLim)
+static void PTRCALL
+latin1_toUtf8(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ char **toP, const char *toLim)
{
for (;;) {
unsigned char c;
@@ -383,23 +428,23 @@ void latin1_toUtf8(const ENCODING *enc,
c = (unsigned char)**fromP;
if (c & 0x80) {
if (toLim - *toP < 2)
- break;
- *(*toP)++ = ((c >> 6) | UTF8_cval2);
- *(*toP)++ = ((c & 0x3f) | 0x80);
+ break;
+ *(*toP)++ = (char)((c >> 6) | UTF8_cval2);
+ *(*toP)++ = (char)((c & 0x3f) | 0x80);
(*fromP)++;
}
else {
if (*toP == toLim)
- break;
+ break;
*(*toP)++ = *(*fromP)++;
}
}
}
-static
-void latin1_toUtf16(const ENCODING *enc,
- const char **fromP, const char *fromLim,
- unsigned short **toP, const unsigned short *toLim)
+static void PTRCALL
+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)++;
@@ -429,10 +474,10 @@ static const struct normal_encoding lati
STANDARD_VTABLE(sb_)
};
-static
-void ascii_toUtf8(const ENCODING *enc,
- const char **fromP, const char *fromLim,
- char **toP, const char *toLim)
+static void PTRCALL
+ascii_toUtf8(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ char **toP, const char *toLim)
{
while (*fromP != fromLim && *toP != toLim)
*(*toP)++ = *(*fromP)++;
@@ -462,7 +507,8 @@ static const struct normal_encoding asci
STANDARD_VTABLE(sb_)
};
-static int unicode_byte_type(char hi, char lo)
+static int PTRFASTCALL
+unicode_byte_type(char hi, char lo)
{
switch ((unsigned char)hi) {
case 0xD8: case 0xD9: case 0xDA: case 0xDB:
@@ -481,10 +527,10 @@ static int unicode_byte_type(char hi, ch
}
#define DEFINE_UTF16_TO_UTF8(E) \
-static \
-void E ## toUtf8(const ENCODING *enc, \
- const char **fromP, const char *fromLim, \
- char **toP, const char *toLim) \
+static void PTRCALL \
+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) { \
@@ -497,7 +543,7 @@ void E ## toUtf8(const ENCODING *enc, \
if (lo < 0x80) { \
if (*toP == toLim) { \
*fromP = from; \
- return; \
+ return; \
} \
*(*toP)++ = lo; \
break; \
@@ -507,7 +553,7 @@ void E ## toUtf8(const ENCODING *enc, \
case 0x4: case 0x5: case 0x6: case 0x7: \
if (toLim - *toP < 2) { \
*fromP = from; \
- return; \
+ return; \
} \
*(*toP)++ = ((lo >> 6) | (hi << 2) | UTF8_cval2); \
*(*toP)++ = ((lo & 0x3f) | 0x80); \
@@ -515,7 +561,7 @@ void E ## toUtf8(const ENCODING *enc, \
default: \
if (toLim - *toP < 3) { \
*fromP = from; \
- return; \
+ return; \
} \
/* 16 bits divided 4, 6, 6 amongst 3 bytes */ \
*(*toP)++ = ((hi >> 4) | UTF8_cval3); \
@@ -524,8 +570,8 @@ void E ## toUtf8(const ENCODING *enc, \
break; \
case 0xD8: case 0xD9: case 0xDA: case 0xDB: \
if (toLim - *toP < 4) { \
- *fromP = from; \
- return; \
+ *fromP = from; \
+ return; \
} \
plane = (((hi & 0x3) << 2) | ((lo >> 6) & 0x3)) + 1; \
*(*toP)++ = ((plane >> 2) | UTF8_cval4); \
@@ -533,9 +579,9 @@ void E ## toUtf8(const ENCODING *enc, \
from += 2; \
lo2 = GET_LO(from); \
*(*toP)++ = (((lo & 0x3) << 4) \
- | ((GET_HI(from) & 0x3) << 2) \
- | (lo2 >> 6) \
- | 0x80); \
+ | ((GET_HI(from) & 0x3) << 2) \
+ | (lo2 >> 6) \
+ | 0x80); \
*(*toP)++ = ((lo2 & 0x3f) | 0x80); \
break; \
} \
@@ -544,10 +590,10 @@ void E ## toUtf8(const ENCODING *enc, \
}
#define DEFINE_UTF16_TO_UTF16(E) \
-static \
-void E ## toUtf16(const ENCODING *enc, \
- const char **fromP, const char *fromLim, \
- unsigned short **toP, const unsigned short *toLim) \
+static void PTRCALL \
+E ## toUtf16(const ENCODING *enc, \
+ const char **fromP, const char *fromLim, \
+ unsigned short **toP, const unsigned short *toLim) \
{ \
/* Avoid copying first half only of surrogate */ \
if (fromLim - *fromP > ((toLim - *toP) << 1) \
@@ -594,32 +640,32 @@ DEFINE_UTF16_TO_UTF16(big2_)
#ifdef XML_MIN_SIZE
-static
-int little2_byteType(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+little2_byteType(const ENCODING *enc, const char *p)
{
return LITTLE2_BYTE_TYPE(enc, p);
}
-static
-int little2_byteToAscii(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+little2_byteToAscii(const ENCODING *enc, const char *p)
{
return LITTLE2_BYTE_TO_ASCII(enc, p);
}
-static
-int little2_charMatches(const ENCODING *enc, const char *p, int c)
+static int PTRCALL
+little2_charMatches(const ENCODING *enc, const char *p, int c)
{
return LITTLE2_CHAR_MATCHES(enc, p, c);
}
-static
-int little2_isNameMin(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+little2_isNameMin(const ENCODING *enc, const char *p)
{
return LITTLE2_IS_NAME_CHAR_MINBPC(enc, p);
}
-static
-int little2_isNmstrtMin(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+little2_isNmstrtMin(const ENCODING *enc, const char *p)
{
return LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p);
}
@@ -634,7 +680,7 @@ int little2_isNmstrtMin(const ENCODING *
#define MINBPC(enc) 2
/* CHAR_MATCHES is guaranteed to have MINBPC bytes available. */
#define BYTE_TYPE(enc, p) LITTLE2_BYTE_TYPE(enc, p)
-#define BYTE_TO_ASCII(enc, p) LITTLE2_BYTE_TO_ASCII(enc, p)
+#define BYTE_TO_ASCII(enc, p) LITTLE2_BYTE_TO_ASCII(enc, p)
#define CHAR_MATCHES(enc, p, c) LITTLE2_CHAR_MATCHES(enc, p, c)
#define IS_NAME_CHAR(enc, p, n) 0
#define IS_NAME_CHAR_MINBPC(enc, p) LITTLE2_IS_NAME_CHAR_MINBPC(enc, p)
@@ -657,9 +703,9 @@ int little2_isNmstrtMin(const ENCODING *
#ifdef XML_NS
-static const struct normal_encoding little2_encoding_ns = {
+static const struct normal_encoding little2_encoding_ns = {
{ VTABLE, 2, 0,
-#if XML_BYTE_ORDER == 12
+#if BYTEORDER == 1234
1
#else
0
@@ -674,9 +720,9 @@ static const struct normal_encoding litt
#endif
-static const struct normal_encoding little2_encoding = {
+static const struct normal_encoding little2_encoding = {
{ VTABLE, 2, 0,
-#if XML_BYTE_ORDER == 12
+#if BYTEORDER == 1234
1
#else
0
@@ -691,11 +737,11 @@ static const struct normal_encoding litt
STANDARD_VTABLE(little2_)
};
-#if XML_BYTE_ORDER != 21
+#if BYTEORDER != 4321
#ifdef XML_NS
-static const struct normal_encoding internal_little2_encoding_ns = {
+static const struct normal_encoding internal_little2_encoding_ns = {
{ VTABLE, 2, 0, 1 },
{
#include "iasciitab.h"
@@ -706,7 +752,7 @@ static const struct normal_encoding inte
#endif
-static const struct normal_encoding internal_little2_encoding = {
+static const struct normal_encoding internal_little2_encoding = {
{ VTABLE, 2, 0, 1 },
{
#define BT_COLON BT_NMSTRT
@@ -733,32 +779,32 @@ static const struct normal_encoding inte
#ifdef XML_MIN_SIZE
-static
-int big2_byteType(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+big2_byteType(const ENCODING *enc, const char *p)
{
return BIG2_BYTE_TYPE(enc, p);
}
-static
-int big2_byteToAscii(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+big2_byteToAscii(const ENCODING *enc, const char *p)
{
return BIG2_BYTE_TO_ASCII(enc, p);
}
-static
-int big2_charMatches(const ENCODING *enc, const char *p, int c)
+static int PTRCALL
+big2_charMatches(const ENCODING *enc, const char *p, int c)
{
return BIG2_CHAR_MATCHES(enc, p, c);
}
-static
-int big2_isNameMin(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+big2_isNameMin(const ENCODING *enc, const char *p)
{
return BIG2_IS_NAME_CHAR_MINBPC(enc, p);
}
-static
-int big2_isNmstrtMin(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+big2_isNmstrtMin(const ENCODING *enc, const char *p)
{
return BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p);
}
@@ -773,7 +819,7 @@ int big2_isNmstrtMin(const ENCODING *enc
#define MINBPC(enc) 2
/* CHAR_MATCHES is guaranteed to have MINBPC bytes available. */
#define BYTE_TYPE(enc, p) BIG2_BYTE_TYPE(enc, p)
-#define BYTE_TO_ASCII(enc, p) BIG2_BYTE_TO_ASCII(enc, p)
+#define BYTE_TO_ASCII(enc, p) BIG2_BYTE_TO_ASCII(enc, p)
#define CHAR_MATCHES(enc, p, c) BIG2_CHAR_MATCHES(enc, p, c)
#define IS_NAME_CHAR(enc, p, n) 0
#define IS_NAME_CHAR_MINBPC(enc, p) BIG2_IS_NAME_CHAR_MINBPC(enc, p)
@@ -798,7 +844,7 @@ int big2_isNmstrtMin(const ENCODING *enc
static const struct normal_encoding big2_encoding_ns = {
{ VTABLE, 2, 0,
-#if XML_BYTE_ORDER == 21
+#if BYTEORDER == 4321
1
#else
0
@@ -815,7 +861,7 @@ static const struct normal_encoding big2
static const struct normal_encoding big2_encoding = {
{ VTABLE, 2, 0,
-#if XML_BYTE_ORDER == 21
+#if BYTEORDER == 4321
1
#else
0
@@ -830,7 +876,7 @@ static const struct normal_encoding big2
STANDARD_VTABLE(big2_)
};
-#if XML_BYTE_ORDER != 12
+#if BYTEORDER != 1234
#ifdef XML_NS
@@ -860,8 +906,8 @@ static const struct normal_encoding inte
#undef PREFIX
-static
-int streqci(const char *s1, const char *s2)
+static int FASTCALL
+streqci(const char *s1, const char *s2)
{
for (;;) {
char c1 = *s1++;
@@ -878,15 +924,15 @@ int streqci(const char *s1, const char *
return 1;
}
-static
-void initUpdatePosition(const ENCODING *enc, const char *ptr,
- const char *end, POSITION *pos)
+static void PTRCALL
+initUpdatePosition(const ENCODING *enc, const char *ptr,
+ const char *end, POSITION *pos)
{
normal_updatePosition(&utf8_encoding.enc, ptr, end, pos);
}
-static
-int toAscii(const ENCODING *enc, const char *ptr, const char *end)
+static int
+toAscii(const ENCODING *enc, const char *ptr, const char *end)
{
char buf[1];
char *p = buf;
@@ -897,34 +943,35 @@ int toAscii(const ENCODING *enc, const c
return buf[0];
}
-static
-int isSpace(int c)
+static int FASTCALL
+isSpace(int c)
{
switch (c) {
case 0x20:
case 0xD:
case 0xA:
- case 0x9:
+ case 0x9:
return 1;
}
return 0;
}
-/* Return 1 if there's just optional white space
-or there's an S followed by name=val. */
-static
-int parsePseudoAttribute(const ENCODING *enc,
- const char *ptr,
- const char *end,
- const char **namePtr,
- const char **nameEndPtr,
- const char **valPtr,
- const char **nextTokPtr)
+/* Return 1 if there's just optional white space or there's an S
+ followed by name=val.
+*/
+static int
+parsePseudoAttribute(const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **namePtr,
+ const char **nameEndPtr,
+ const char **valPtr,
+ const char **nextTokPtr)
{
int c;
char open;
if (ptr == end) {
- *namePtr = 0;
+ *namePtr = NULL;
return 1;
}
if (!isSpace(toAscii(enc, ptr, end))) {
@@ -935,7 +982,7 @@ int parsePseudoAttribute(const ENCODING
ptr += enc->minBytesPerChar;
} while (isSpace(toAscii(enc, ptr, end)));
if (ptr == end) {
- *namePtr = 0;
+ *namePtr = NULL;
return 1;
}
*namePtr = ptr;
@@ -952,11 +999,11 @@ int parsePseudoAttribute(const ENCODING
if (isSpace(c)) {
*nameEndPtr = ptr;
do {
- ptr += enc->minBytesPerChar;
+ ptr += enc->minBytesPerChar;
} while (isSpace(c = toAscii(enc, ptr, end)));
if (c != ASCII_EQUALS) {
- *nextTokPtr = ptr;
- return 0;
+ *nextTokPtr = ptr;
+ return 0;
}
break;
}
@@ -976,7 +1023,7 @@ int parsePseudoAttribute(const ENCODING
*nextTokPtr = ptr;
return 0;
}
- open = c;
+ open = (char)c;
ptr += enc->minBytesPerChar;
*valPtr = ptr;
for (;; ptr += enc->minBytesPerChar) {
@@ -984,11 +1031,11 @@ int parsePseudoAttribute(const ENCODING
if (c == open)
break;
if (!(ASCII_a <= c && c <= ASCII_z)
- && !(ASCII_A <= c && c <= ASCII_Z)
- && !(ASCII_0 <= c && c <= ASCII_9)
- && c != ASCII_PERIOD
- && c != ASCII_MINUS
- && c != ASCII_UNDERSCORE) {
+ && !(ASCII_A <= c && c <= ASCII_Z)
+ && !(ASCII_0 <= c && c <= ASCII_9)
+ && c != ASCII_PERIOD
+ && c != ASCII_MINUS
+ && c != ASCII_UNDERSCORE) {
*nextTokPtr = ptr;
return 0;
}
@@ -1006,7 +1053,8 @@ static const char KW_encoding[] = {
};
static const char KW_standalone[] = {
- ASCII_s, ASCII_t, ASCII_a, ASCII_n, ASCII_d, ASCII_a, ASCII_l, ASCII_o, ASCII_n, ASCII_e, '\0'
+ ASCII_s, ASCII_t, ASCII_a, ASCII_n, ASCII_d, ASCII_a, ASCII_l, ASCII_o,
+ ASCII_n, ASCII_e, '\0'
};
static const char KW_yes[] = {
@@ -1017,27 +1065,28 @@ static const char KW_no[] = {
ASCII_n, ASCII_o, '\0'
};
-static
-int doParseXmlDecl(const ENCODING *(*encodingFinder)(const ENCODING *,
- const char *,
- const char *),
- int isGeneralTextEntity,
- const ENCODING *enc,
- const char *ptr,
- const char *end,
- const char **badPtr,
- const char **versionPtr,
- const char **versionEndPtr,
- const char **encodingName,
- const ENCODING **encoding,
- int *standalone)
-{
- const char *val = 0;
- const char *name = 0;
- const char *nameEnd = 0;
+static int
+doParseXmlDecl(const ENCODING *(*encodingFinder)(const ENCODING *,
+ const char *,
+ const char *),
+ int isGeneralTextEntity,
+ const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr,
+ const char **versionPtr,
+ const char **versionEndPtr,
+ const char **encodingName,
+ const ENCODING **encoding,
+ int *standalone)
+{
+ const char *val = NULL;
+ const char *name = NULL;
+ const char *nameEnd = NULL;
ptr += 5 * enc->minBytesPerChar;
end -= 2 * enc->minBytesPerChar;
- if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr) || !name) {
+ if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr)
+ || !name) {
*badPtr = ptr;
return 0;
}
@@ -1058,9 +1107,9 @@ int doParseXmlDecl(const ENCODING *(*enc
}
if (!name) {
if (isGeneralTextEntity) {
- /* a TextDecl must have an EncodingDecl */
- *badPtr = ptr;
- return 0;
+ /* a TextDecl must have an EncodingDecl */
+ *badPtr = ptr;
+ return 0;
}
return 1;
}
@@ -1082,7 +1131,8 @@ int doParseXmlDecl(const ENCODING *(*enc
if (!name)
return 1;
}
- if (!XmlNameMatchesAscii(enc, name, nameEnd, KW_standalone) || isGeneralTextEntity) {
+ if (!XmlNameMatchesAscii(enc, name, nameEnd, KW_standalone)
+ || isGeneralTextEntity) {
*badPtr = name;
return 0;
}
@@ -1107,8 +1157,8 @@ int doParseXmlDecl(const ENCODING *(*enc
return 1;
}
-static
-int checkCharRefNumber(int result)
+static int FASTCALL
+checkCharRefNumber(int result)
{
switch (result >> 8) {
case 0xD8: case 0xD9: case 0xDA: case 0xDB:
@@ -1126,7 +1176,8 @@ int checkCharRefNumber(int result)
return result;
}
-int XmlUtf8Encode(int c, char *buf)
+int FASTCALL
+XmlUtf8Encode(int c, char *buf)
{
enum {
/* minN is minimum legal resulting value for N byte sequence */
@@ -1138,42 +1189,43 @@ int XmlUtf8Encode(int c, char *buf)
if (c < 0)
return 0;
if (c < min2) {
- buf[0] = (c | UTF8_cval1);
+ buf[0] = (char)(c | UTF8_cval1);
return 1;
}
if (c < min3) {
- buf[0] = ((c >> 6) | UTF8_cval2);
- buf[1] = ((c & 0x3f) | 0x80);
+ buf[0] = (char)((c >> 6) | UTF8_cval2);
+ buf[1] = (char)((c & 0x3f) | 0x80);
return 2;
}
if (c < min4) {
- buf[0] = ((c >> 12) | UTF8_cval3);
- buf[1] = (((c >> 6) & 0x3f) | 0x80);
- buf[2] = ((c & 0x3f) | 0x80);
+ buf[0] = (char)((c >> 12) | UTF8_cval3);
+ buf[1] = (char)(((c >> 6) & 0x3f) | 0x80);
+ buf[2] = (char)((c & 0x3f) | 0x80);
return 3;
}
if (c < 0x110000) {
- buf[0] = ((c >> 18) | UTF8_cval4);
- buf[1] = (((c >> 12) & 0x3f) | 0x80);
- buf[2] = (((c >> 6) & 0x3f) | 0x80);
- buf[3] = ((c & 0x3f) | 0x80);
+ buf[0] = (char)((c >> 18) | UTF8_cval4);
+ buf[1] = (char)(((c >> 12) & 0x3f) | 0x80);
+ buf[2] = (char)(((c >> 6) & 0x3f) | 0x80);
+ buf[3] = (char)((c & 0x3f) | 0x80);
return 4;
}
return 0;
}
-int XmlUtf16Encode(int charNum, unsigned short *buf)
+int FASTCALL
+XmlUtf16Encode(int charNum, unsigned short *buf)
{
if (charNum < 0)
return 0;
if (charNum < 0x10000) {
- buf[0] = charNum;
+ buf[0] = (unsigned short)charNum;
return 1;
}
if (charNum < 0x110000) {
charNum -= 0x10000;
- buf[0] = (charNum >> 10) + 0xD800;
- buf[1] = (charNum & 0x3FF) + 0xDC00;
+ buf[0] = (unsigned short)((charNum >> 10) + 0xD800);
+ buf[1] = (unsigned short)((charNum & 0x3FF) + 0xDC00);
return 2;
}
return 0;
@@ -1187,65 +1239,68 @@ struct unknown_encoding {
char utf8[256][4];
};
-int XmlSizeOfUnknownEncoding(void)
+#define AS_UNKNOWN_ENCODING(enc) ((const struct unknown_encoding *) (enc))
+
+int
+XmlSizeOfUnknownEncoding(void)
{
return sizeof(struct unknown_encoding);
}
-static
-int unknown_isName(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+unknown_isName(const ENCODING *enc, const char *p)
{
- int c = ((const struct unknown_encoding *)enc)
- ->convert(((const struct unknown_encoding *)enc)->userData, p);
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+ int c = uenc->convert(uenc->userData, p);
if (c & ~0xFFFF)
return 0;
return UCS2_GET_NAMING(namePages, c >> 8, c & 0xFF);
}
-static
-int unknown_isNmstrt(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+unknown_isNmstrt(const ENCODING *enc, const char *p)
{
- int c = ((const struct unknown_encoding *)enc)
- ->convert(((const struct unknown_encoding *)enc)->userData, p);
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+ int c = uenc->convert(uenc->userData, p);
if (c & ~0xFFFF)
return 0;
return UCS2_GET_NAMING(nmstrtPages, c >> 8, c & 0xFF);
}
-static
-int unknown_isInvalid(const ENCODING *enc, const char *p)
+static int PTRFASTCALL
+unknown_isInvalid(const ENCODING *enc, const char *p)
{
- int c = ((const struct unknown_encoding *)enc)
- ->convert(((const struct unknown_encoding *)enc)->userData, p);
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+ int c = uenc->convert(uenc->userData, p);
return (c & ~0xFFFF) || checkCharRefNumber(c) < 0;
}
-static
-void unknown_toUtf8(const ENCODING *enc,
- const char **fromP, const char *fromLim,
- char **toP, const char *toLim)
+static void PTRCALL
+unknown_toUtf8(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ char **toP, const char *toLim)
{
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
char buf[XML_UTF8_ENCODE_MAX];
for (;;) {
const char *utf8;
int n;
if (*fromP == fromLim)
break;
- utf8 = ((const struct unknown_encoding *)enc)->utf8[(unsigned char)**fromP];
+ utf8 = uenc->utf8[(unsigned char)**fromP];
n = *utf8++;
if (n == 0) {
- int c = ((const struct unknown_encoding *)enc)
- ->convert(((const struct unknown_encoding *)enc)->userData, *fromP);
+ int c = uenc->convert(uenc->userData, *fromP);
n = XmlUtf8Encode(c, buf);
if (n > toLim - *toP)
- break;
+ break;
utf8 = buf;
- *fromP += ((const struct normal_encoding *)enc)->type[(unsigned char)**fromP]
- - (BT_LEAD2 - 2);
+ *fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP]
+ - (BT_LEAD2 - 2));
}
else {
if (n > toLim - *toP)
- break;
+ break;
(*fromP)++;
}
do {
@@ -1254,19 +1309,19 @@ void unknown_toUtf8(const ENCODING *enc,
}
}
-static
-void unknown_toUtf16(const ENCODING *enc,
- const char **fromP, const char *fromLim,
- unsigned short **toP, const unsigned short *toLim)
+static void PTRCALL
+unknown_toUtf16(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ unsigned short **toP, const unsigned short *toLim)
{
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
while (*fromP != fromLim && *toP != toLim) {
- unsigned short c
- = ((const struct unknown_encoding *)enc)->utf16[(unsigned char)**fromP];
+ unsigned short c = uenc->utf16[(unsigned char)**fromP];
if (c == 0) {
- c = (unsigned short)((const struct unknown_encoding *)enc)
- ->convert(((const struct unknown_encoding *)enc)->userData, *fromP);
- *fromP += ((const struct normal_encoding *)enc)->type[(unsigned char)**fromP]
- - (BT_LEAD2 - 2);
+ c = (unsigned short)
+ uenc->convert(uenc->userData, *fromP);
+ *fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP]
+ - (BT_LEAD2 - 2));
}
else
(*fromP)++;
@@ -1276,18 +1331,18 @@ void unknown_toUtf16(const ENCODING *enc
ENCODING *
XmlInitUnknownEncoding(void *mem,
- int *table,
- int (*convert)(void *userData, const char *p),
- void *userData)
+ int *table,
+ CONVERTER convert,
+ void *userData)
{
int i;
- struct unknown_encoding *e = mem;
+ struct unknown_encoding *e = (struct unknown_encoding *)mem;
for (i = 0; i < (int)sizeof(struct normal_encoding); i++)
((char *)mem)[i] = ((char *)&latin1_encoding)[i];
for (i = 0; i < 128; i++)
if (latin1_encoding.type[i] != BT_OTHER
&& latin1_encoding.type[i] != BT_NONXML
- && table[i] != i)
+ && table[i] != i)
return 0;
for (i = 0; i < 256; i++) {
int c = table[i];
@@ -1300,20 +1355,20 @@ XmlInitUnknownEncoding(void *mem,
}
else if (c < 0) {
if (c < -4)
- return 0;
- e->normal.type[i] = BT_LEAD2 - (c + 2);
+ return 0;
+ e->normal.type[i] = (unsigned char)(BT_LEAD2 - (c + 2));
e->utf8[i][0] = 0;
e->utf16[i] = 0;
}
else if (c < 0x80) {
if (latin1_encoding.type[c] != BT_OTHER
- && latin1_encoding.type[c] != BT_NONXML
- && c != i)
- return 0;
+ && latin1_encoding.type[c] != BT_NONXML
+ && c != i)
+ return 0;
e->normal.type[i] = latin1_encoding.type[c];
e->utf8[i][0] = 1;
e->utf8[i][1] = (char)c;
- e->utf16[i] = c == 0 ? 0xFFFF : c;
+ e->utf16[i] = (unsigned short)(c == 0 ? 0xFFFF : c);
}
else if (checkCharRefNumber(c) < 0) {
e->normal.type[i] = BT_NONXML;
@@ -1324,15 +1379,15 @@ XmlInitUnknownEncoding(void *mem,
}
else {
if (c > 0xFFFF)
- return 0;
+ return 0;
if (UCS2_GET_NAMING(nmstrtPages, c >> 8, c & 0xff))
- e->normal.type[i] = BT_NMSTRT;
+ e->normal.type[i] = BT_NMSTRT;
else if (UCS2_GET_NAMING(namePages, c >> 8, c & 0xff))
- e->normal.type[i] = BT_NAME;
+ e->normal.type[i] = BT_NAME;
else
- e->normal.type[i] = BT_OTHER;
+ e->normal.type[i] = BT_OTHER;
e->utf8[i][0] = (char)XmlUtf8Encode(c, e->utf8[i] + 1);
- e->utf16[i] = c;
+ e->utf16[i] = (unsigned short)c;
}
}
e->userData = userData;
@@ -1368,26 +1423,30 @@ enum {
};
static const char KW_ISO_8859_1[] = {
- ASCII_I, ASCII_S, ASCII_O, ASCII_MINUS, ASCII_8, ASCII_8, ASCII_5, ASCII_9, ASCII_MINUS, ASCII_1, '\0'
+ ASCII_I, ASCII_S, ASCII_O, ASCII_MINUS, ASCII_8, ASCII_8, ASCII_5, ASCII_9,
+ ASCII_MINUS, ASCII_1, '\0'
};
static const char KW_US_ASCII[] = {
- ASCII_U, ASCII_S, ASCII_MINUS, ASCII_A, ASCII_S, ASCII_C, ASCII_I, ASCII_I, '\0'
+ ASCII_U, ASCII_S, ASCII_MINUS, ASCII_A, ASCII_S, ASCII_C, ASCII_I, ASCII_I,
+ '\0'
};
-static const char KW_UTF_8[] = {
+static const char KW_UTF_8[] = {
ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_8, '\0'
};
-static const char KW_UTF_16[] = {
+static const char KW_UTF_16[] = {
ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, '\0'
};
static const char KW_UTF_16BE[] = {
- ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_B, ASCII_E, '\0'
+ ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_B, ASCII_E,
+ '\0'
};
static const char KW_UTF_16LE[] = {
- ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_L, ASCII_E, '\0'
+ ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_L, ASCII_E,
+ '\0'
};
-static
-int getEncodingIndex(const char *name)
+static int FASTCALL
+getEncodingIndex(const char *name)
{
static const char *encodingNames[] = {
KW_ISO_8859_1,
@@ -1398,7 +1457,7 @@ int getEncodingIndex(const char *name)
KW_UTF_16LE,
};
int i;
- if (name == 0)
+ if (name == NULL)
return NO_ENC;
for (i = 0; i < (int)(sizeof(encodingNames)/sizeof(encodingNames[0])); i++)
if (streqci(name, encodingNames[i]))
@@ -1406,27 +1465,28 @@ int getEncodingIndex(const char *name)
return UNKNOWN_ENC;
}
-/* For binary compatibility, we store the index of the encoding specified
-at initialization in the isUtf16 member. */
+/* For binary compatibility, we store the index of the encoding
+ specified at initialization in the isUtf16 member.
+*/
#define INIT_ENC_INDEX(enc) ((int)(enc)->initEnc.isUtf16)
#define SET_INIT_ENC_INDEX(enc, i) ((enc)->initEnc.isUtf16 = (char)i)
-/* This is what detects the encoding.
-encodingTable maps from encoding indices to encodings;
-INIT_ENC_INDEX(enc) is the index of the external (protocol) specified encoding;
-state is XML_CONTENT_STATE if we're parsing an external text entity,
-and XML_PROLOG_STATE otherwise.
+/* This is what detects the encoding. encodingTable maps from
+ encoding indices to encodings; INIT_ENC_INDEX(enc) is the index of
+ the external (protocol) specified encoding; state is
+ XML_CONTENT_STATE if we're parsing an external text entity, and
+ XML_PROLOG_STATE otherwise.
*/
-static
-int initScan(const ENCODING **encodingTable,
- const INIT_ENCODING *enc,
- int state,
- const char *ptr,
- const char *end,
- const char **nextTokPtr)
+static int
+initScan(const ENCODING **encodingTable,
+ const INIT_ENCODING *enc,
+ int state,
+ const char *ptr,
+ const char *end,
+ const char **nextTokPtr)
{
const ENCODING **encPtr;
@@ -1453,8 +1513,8 @@ int initScan(const ENCODING **encodingTa
case 0xFF:
case 0xEF: /* possibly first byte of UTF-8 BOM */
if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
- && state == XML_CONTENT_STATE)
- break;
+ && state == XML_CONTENT_STATE)
+ break;
/* fall through */
case 0x00:
case 0x3C:
@@ -1465,23 +1525,23 @@ int initScan(const ENCODING **encodingTa
switch (((unsigned char)ptr[0] << 8) | (unsigned char)ptr[1]) {
case 0xFEFF:
if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
- && state == XML_CONTENT_STATE)
- break;
+ && state == XML_CONTENT_STATE)
+ break;
*nextTokPtr = ptr + 2;
*encPtr = encodingTable[UTF_16BE_ENC];
return XML_TOK_BOM;
/* 00 3C is handled in the default case */
case 0x3C00:
if ((INIT_ENC_INDEX(enc) == UTF_16BE_ENC
- || INIT_ENC_INDEX(enc) == UTF_16_ENC)
- && state == XML_CONTENT_STATE)
- break;
+ || INIT_ENC_INDEX(enc) == UTF_16_ENC)
+ && state == XML_CONTENT_STATE)
+ break;
*encPtr = encodingTable[UTF_16LE_ENC];
return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
case 0xFFFE:
if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
- && state == XML_CONTENT_STATE)
- break;
+ && state == XML_CONTENT_STATE)
+ break;
*nextTokPtr = ptr + 2;
*encPtr = encodingTable[UTF_16LE_ENC];
return XML_TOK_BOM;
@@ -1490,45 +1550,50 @@ int initScan(const ENCODING **encodingTa
/* If there's an explicitly specified (external) encoding
of ISO-8859-1 or some flavour of UTF-16
and this is an external text entity,
- don't look for the BOM,
- because it might be a legal data. */
+ don't look for the BOM,
+ because it might be a legal data.
+ */
if (state == XML_CONTENT_STATE) {
- int e = INIT_ENC_INDEX(enc);
- if (e == ISO_8859_1_ENC || e == UTF_16BE_ENC || e == UTF_16LE_ENC || e == UTF_16_ENC)
- break;
+ int e = INIT_ENC_INDEX(enc);
+ if (e == ISO_8859_1_ENC || e == UTF_16BE_ENC
+ || e == UTF_16LE_ENC || e == UTF_16_ENC)
+ break;
}
if (ptr + 2 == end)
- return XML_TOK_PARTIAL;
+ return XML_TOK_PARTIAL;
if ((unsigned char)ptr[2] == 0xBF) {
- *nextTokPtr = ptr + 3;
- *encPtr = encodingTable[UTF_8_ENC];
- return XML_TOK_BOM;
+ *nextTokPtr = ptr + 3;
+ *encPtr = encodingTable[UTF_8_ENC];
+ return XML_TOK_BOM;
}
break;
default:
if (ptr[0] == '\0') {
- /* 0 isn't a legal data character. Furthermore a document entity can only
- start with ASCII characters. So the only way this can fail to be big-endian
- UTF-16 if it it's an external parsed general entity that's labelled as
- UTF-16LE. */
- if (state == XML_CONTENT_STATE && INIT_ENC_INDEX(enc) == UTF_16LE_ENC)
- break;
- *encPtr = encodingTable[UTF_16BE_ENC];
- return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+ /* 0 isn't a legal data character. Furthermore a document
+ entity can only start with ASCII characters. So the only
+ way this can fail to be big-endian UTF-16 if it it's an
+ external parsed general entity that's labelled as
+ UTF-16LE.
+ */
+ if (state == XML_CONTENT_STATE && INIT_ENC_INDEX(enc) == UTF_16LE_ENC)
+ break;
+ *encPtr = encodingTable[UTF_16BE_ENC];
+ return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
}
else if (ptr[1] == '\0') {
- /* We could recover here in the case:
- - parsing an external entity
- - second byte is 0
- - no externally specified encoding
- - no encoding declaration
- by assuming UTF-16LE. But we don't, because this would mean when
- presented just with a single byte, we couldn't reliably determine
- whether we needed further bytes. */
- if (state == XML_CONTENT_STATE)
- break;
- *encPtr = encodingTable[UTF_16LE_ENC];
- return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+ /* We could recover here in the case:
+ - parsing an external entity
+ - second byte is 0
+ - no externally specified encoding
+ - no encoding declaration
+ by assuming UTF-16LE. But we don't, because this would mean when
+ presented just with a single byte, we couldn't reliably determine
+ whether we needed further bytes.
+ */
+ if (state == XML_CONTENT_STATE)
+ break;
+ *encPtr = encodingTable[UTF_16LE_ENC];
+ return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
}
break;
}
@@ -1556,9 +1621,9 @@ int initScan(const ENCODING **encodingTa
ENCODING *
XmlInitUnknownEncodingNS(void *mem,
- int *table,
- int (*convert)(void *userData, const char *p),
- void *userData)
+ int *table,
+ CONVERTER convert,
+ void *userData)
{
ENCODING *enc = XmlInitUnknownEncoding(mem, table, convert, userData);
if (enc)
Modified: apr/apr-util/branches/0.9.x/xml/expat/lib/xmltok.h
URL: http://svn.apache.org/viewvc/apr/apr-util/branches/0.9.x/xml/expat/lib/xmltok.h?rev=1005995&r1=1005994&r2=1005995&view=diff
==============================================================================
--- apr/apr-util/branches/0.9.x/xml/expat/lib/xmltok.h (original)
+++ apr/apr-util/branches/0.9.x/xml/expat/lib/xmltok.h Fri Oct 8 20:01:10 2010
@@ -1,6 +1,5 @@
-/*
-Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
-See the file COPYING for copying permission.
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
*/
#ifndef XmlTok_INCLUDED
@@ -11,19 +10,21 @@ extern "C" {
#endif
/* The following token may be returned by XmlContentTok */
-#define XML_TOK_TRAILING_RSQB -5 /* ] or ]] at the end of the scan; might be start of
- illegal ]]> sequence */
-/* The following tokens may be returned by both XmlPrologTok and XmlContentTok */
-#define XML_TOK_NONE -4 /* The string to be scanned is empty */
-#define XML_TOK_TRAILING_CR -3 /* A CR at the end of the scan;
- might be part of CRLF sequence */
-#define XML_TOK_PARTIAL_CHAR -2 /* only part of a multibyte sequence */
-#define XML_TOK_PARTIAL -1 /* only part of a token */
+#define XML_TOK_TRAILING_RSQB -5 /* ] or ]] at the end of the scan; might be
+ start of illegal ]]> sequence */
+/* The following tokens may be returned by both XmlPrologTok and
+ XmlContentTok.
+*/
+#define XML_TOK_NONE -4 /* The string to be scanned is empty */
+#define XML_TOK_TRAILING_CR -3 /* A CR at the end of the scan;
+ might be part of CRLF sequence */
+#define XML_TOK_PARTIAL_CHAR -2 /* only part of a multibyte sequence */
+#define XML_TOK_PARTIAL -1 /* only part of a token */
#define XML_TOK_INVALID 0
/* The following tokens are returned by XmlContentTok; some are also
- returned by XmlAttributeValueTok, XmlEntityTok, XmlCdataSectionTok */
-
+ returned by XmlAttributeValueTok, XmlEntityTok, XmlCdataSectionTok.
+*/
#define XML_TOK_START_TAG_WITH_ATTS 1
#define XML_TOK_START_TAG_NO_ATTS 2
#define XML_TOK_EMPTY_ELEMENT_WITH_ATTS 3 /* empty element tag <e/> */
@@ -33,22 +34,24 @@ extern "C" {
#define XML_TOK_DATA_NEWLINE 7
#define XML_TOK_CDATA_SECT_OPEN 8
#define XML_TOK_ENTITY_REF 9
-#define XML_TOK_CHAR_REF 10 /* numeric character reference */
+#define XML_TOK_CHAR_REF 10 /* numeric character reference */
-/* The following tokens may be returned by both XmlPrologTok and XmlContentTok */
-#define XML_TOK_PI 11 /* processing instruction */
-#define XML_TOK_XML_DECL 12 /* XML decl or text decl */
+/* The following tokens may be returned by both XmlPrologTok and
+ XmlContentTok.
+*/
+#define XML_TOK_PI 11 /* processing instruction */
+#define XML_TOK_XML_DECL 12 /* XML decl or text decl */
#define XML_TOK_COMMENT 13
-#define XML_TOK_BOM 14 /* Byte order mark */
+#define XML_TOK_BOM 14 /* Byte order mark */
/* The following tokens are returned only by XmlPrologTok */
#define XML_TOK_PROLOG_S 15
-#define XML_TOK_DECL_OPEN 16 /* <!foo */
-#define XML_TOK_DECL_CLOSE 17 /* > */
+#define XML_TOK_DECL_OPEN 16 /* <!foo */
+#define XML_TOK_DECL_CLOSE 17 /* > */
#define XML_TOK_NAME 18
#define XML_TOK_NMTOKEN 19
-#define XML_TOK_POUND_NAME 20 /* #name */
-#define XML_TOK_OR 21 /* | */
+#define XML_TOK_POUND_NAME 20 /* #name */
+#define XML_TOK_OR 21 /* | */
#define XML_TOK_PERCENT 22
#define XML_TOK_OPEN_PAREN 23
#define XML_TOK_CLOSE_PAREN 24
@@ -59,14 +62,14 @@ extern "C" {
#define XML_TOK_INSTANCE_START 29
/* The following occur only in element type declarations */
-#define XML_TOK_NAME_QUESTION 30 /* name? */
-#define XML_TOK_NAME_ASTERISK 31 /* name* */
-#define XML_TOK_NAME_PLUS 32 /* name+ */
-#define XML_TOK_COND_SECT_OPEN 33 /* <![ */
-#define XML_TOK_COND_SECT_CLOSE 34 /* ]]> */
-#define XML_TOK_CLOSE_PAREN_QUESTION 35 /* )? */
-#define XML_TOK_CLOSE_PAREN_ASTERISK 36 /* )* */
-#define XML_TOK_CLOSE_PAREN_PLUS 37 /* )+ */
+#define XML_TOK_NAME_QUESTION 30 /* name? */
+#define XML_TOK_NAME_ASTERISK 31 /* name* */
+#define XML_TOK_NAME_PLUS 32 /* name+ */
+#define XML_TOK_COND_SECT_OPEN 33 /* <![ */
+#define XML_TOK_COND_SECT_CLOSE 34 /* ]]> */
+#define XML_TOK_CLOSE_PAREN_QUESTION 35 /* )? */
+#define XML_TOK_CLOSE_PAREN_ASTERISK 36 /* )* */
+#define XML_TOK_CLOSE_PAREN_PLUS 37 /* )+ */
#define XML_TOK_COMMA 38
/* The following token is returned only by XmlAttributeValueTok */
@@ -75,8 +78,9 @@ extern "C" {
/* The following token is returned only by XmlCdataSectionTok */
#define XML_TOK_CDATA_SECT_CLOSE 40
-/* With namespace processing this is returned by XmlPrologTok
- for a name with a colon. */
+/* With namespace processing this is returned by XmlPrologTok for a
+ name with a colon.
+*/
#define XML_TOK_PREFIXED_NAME 41
#ifdef XML_DTD
@@ -121,64 +125,73 @@ typedef struct {
struct encoding;
typedef struct encoding ENCODING;
+typedef int (PTRCALL *SCANNER)(const ENCODING *,
+ const char *,
+ const char *,
+ const char **);
+
struct encoding {
- int (*scanners[XML_N_STATES])(const ENCODING *,
- const char *,
- const char *,
- const char **);
- int (*literalScanners[XML_N_LITERAL_TYPES])(const ENCODING *,
- const char *,
- const char *,
- const char **);
- int (*sameName)(const ENCODING *,
- const char *, const char *);
- int (*nameMatchesAscii)(const ENCODING *,
- const char *, const char *, const char *);
- int (*nameLength)(const ENCODING *, const char *);
- const char *(*skipS)(const ENCODING *, const char *);
- int (*getAtts)(const ENCODING *enc, const char *ptr,
- int attsMax, ATTRIBUTE *atts);
- int (*charRefNumber)(const ENCODING *enc, const char *ptr);
- int (*predefinedEntityName)(const ENCODING *, const char *, const char *);
- void (*updatePosition)(const ENCODING *,
- const char *ptr,
- const char *end,
- POSITION *);
- int (*isPublicId)(const ENCODING *enc, const char *ptr, const char *end,
- const char **badPtr);
- void (*utf8Convert)(const ENCODING *enc,
- const char **fromP,
- const char *fromLim,
- char **toP,
- const char *toLim);
- void (*utf16Convert)(const ENCODING *enc,
- const char **fromP,
- const char *fromLim,
- unsigned short **toP,
- const unsigned short *toLim);
+ SCANNER scanners[XML_N_STATES];
+ SCANNER literalScanners[XML_N_LITERAL_TYPES];
+ int (PTRCALL *sameName)(const ENCODING *,
+ const char *,
+ const char *);
+ int (PTRCALL *nameMatchesAscii)(const ENCODING *,
+ const char *,
+ const char *,
+ const char *);
+ int (PTRFASTCALL *nameLength)(const ENCODING *, const char *);
+ const char *(PTRFASTCALL *skipS)(const ENCODING *, const char *);
+ int (PTRCALL *getAtts)(const ENCODING *enc,
+ const char *ptr,
+ int attsMax,
+ ATTRIBUTE *atts);
+ int (PTRFASTCALL *charRefNumber)(const ENCODING *enc, const char *ptr);
+ int (PTRCALL *predefinedEntityName)(const ENCODING *,
+ const char *,
+ const char *);
+ void (PTRCALL *updatePosition)(const ENCODING *,
+ const char *ptr,
+ const char *end,
+ POSITION *);
+ int (PTRCALL *isPublicId)(const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr);
+ void (PTRCALL *utf8Convert)(const ENCODING *enc,
+ const char **fromP,
+ const char *fromLim,
+ char **toP,
+ const char *toLim);
+ void (PTRCALL *utf16Convert)(const ENCODING *enc,
+ const char **fromP,
+ const char *fromLim,
+ unsigned short **toP,
+ const unsigned short *toLim);
int minBytesPerChar;
char isUtf8;
char isUtf16;
};
-/*
-Scan the string starting at ptr until the end of the next complete token,
-but do not scan past eptr. Return an integer giving the type of token.
+/* Scan the string starting at ptr until the end of the next complete
+ token, but do not scan past eptr. Return an integer giving the
+ type of token.
-Return XML_TOK_NONE when ptr == eptr; nextTokPtr will not be set.
+ Return XML_TOK_NONE when ptr == eptr; nextTokPtr will not be set.
-Return XML_TOK_PARTIAL when the string does not contain a complete token;
-nextTokPtr will not be set.
+ Return XML_TOK_PARTIAL when the string does not contain a complete
+ token; nextTokPtr will not be set.
-Return XML_TOK_INVALID when the string does not start a valid token; nextTokPtr
-will be set to point to the character which made the token invalid.
+ Return XML_TOK_INVALID when the string does not start a valid
+ token; nextTokPtr will be set to point to the character which made
+ the token invalid.
-Otherwise the string starts with a valid token; nextTokPtr will be set to point
-to the character following the end of that token.
+ Otherwise the string starts with a valid token; nextTokPtr will be
+ set to point to the character following the end of that token.
-Each data character counts as a single token, but adjacent data characters
-may be returned together. Similarly for characters in the prolog outside
-literals, comments and processing instructions.
+ Each data character counts as a single token, but adjacent data
+ characters may be returned together. Similarly for characters in
+ the prolog outside literals, comments and processing instructions.
*/
@@ -201,9 +214,9 @@ literals, comments and processing instru
#endif /* XML_DTD */
-/* This is used for performing a 2nd-level tokenization on
-the content of a literal that has already been returned by XmlTok. */
-
+/* This is used for performing a 2nd-level tokenization on the content
+ of a literal that has already been returned by XmlTok.
+*/
#define XmlLiteralTok(enc, literalType, ptr, end, nextTokPtr) \
(((enc)->literalScanners[literalType])(enc, ptr, end, nextTokPtr))
@@ -250,48 +263,51 @@ typedef struct {
const ENCODING **encPtr;
} INIT_ENCODING;
-int XmlParseXmlDecl(int isGeneralTextEntity,
- const ENCODING *enc,
- const char *ptr,
- const char *end,
- const char **badPtr,
- const char **versionPtr,
- const char **versionEndPtr,
- const char **encodingNamePtr,
- const ENCODING **namedEncodingPtr,
- int *standalonePtr);
-
-int XmlInitEncoding(INIT_ENCODING *, const ENCODING **, const char *name);
-const ENCODING *XmlGetUtf8InternalEncoding(void);
-const ENCODING *XmlGetUtf16InternalEncoding(void);
-int XmlUtf8Encode(int charNumber, char *buf);
-int XmlUtf16Encode(int charNumber, unsigned short *buf);
+int XmlParseXmlDecl(int isGeneralTextEntity,
+ const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr,
+ const char **versionPtr,
+ const char **versionEndPtr,
+ const char **encodingNamePtr,
+ const ENCODING **namedEncodingPtr,
+ int *standalonePtr);
+
+int XmlInitEncoding(INIT_ENCODING *, const ENCODING **, const char *name);
+const ENCODING *XmlGetUtf8InternalEncoding(void);
+const ENCODING *XmlGetUtf16InternalEncoding(void);
+int FASTCALL XmlUtf8Encode(int charNumber, char *buf);
+int FASTCALL XmlUtf16Encode(int charNumber, unsigned short *buf);
+int XmlSizeOfUnknownEncoding(void);
+
+typedef int (*CONVERTER)(void *userData, const char *p);
-int XmlSizeOfUnknownEncoding(void);
-ENCODING *
+ENCODING *
XmlInitUnknownEncoding(void *mem,
- int *table,
- int (*conv)(void *userData, const char *p),
- void *userData);
-
-int XmlParseXmlDeclNS(int isGeneralTextEntity,
- const ENCODING *enc,
- const char *ptr,
- const char *end,
- const char **badPtr,
- const char **versionPtr,
- const char **versionEndPtr,
- const char **encodingNamePtr,
- const ENCODING **namedEncodingPtr,
- int *standalonePtr);
-int XmlInitEncodingNS(INIT_ENCODING *, const ENCODING **, const char *name);
-const ENCODING *XmlGetUtf8InternalEncodingNS(void);
-const ENCODING *XmlGetUtf16InternalEncodingNS(void);
-ENCODING *
+ int *table,
+ CONVERTER convert,
+ void *userData);
+
+int XmlParseXmlDeclNS(int isGeneralTextEntity,
+ const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr,
+ const char **versionPtr,
+ const char **versionEndPtr,
+ const char **encodingNamePtr,
+ const ENCODING **namedEncodingPtr,
+ int *standalonePtr);
+
+int XmlInitEncodingNS(INIT_ENCODING *, const ENCODING **, const char *name);
+const ENCODING *XmlGetUtf8InternalEncodingNS(void);
+const ENCODING *XmlGetUtf16InternalEncodingNS(void);
+ENCODING *
XmlInitUnknownEncodingNS(void *mem,
- int *table,
- int (*conv)(void *userData, const char *p),
- void *userData);
+ int *table,
+ CONVERTER convert,
+ void *userData);
#ifdef __cplusplus
}
#endif