Swift
summaryrefslogtreecommitdiffstats
path: root/3rdParty/Expat/src/xmltok.c
diff options
context:
space:
mode:
authorTobias Markmann <tm@ayena.de>2017-06-20 12:17:48 (GMT)
committerTobias Markmann <tm@ayena.de>2017-06-20 14:43:05 (GMT)
commita507a88a189bb603c9f2d686c9c8dafca49c053d (patch)
treeac32d8281fdefb83f3a0d9ef79c3983d98bed24d /3rdParty/Expat/src/xmltok.c
parentb807e3fa975cf25e5e901b59643419a5a73a12fe (diff)
downloadswift-a507a88a189bb603c9f2d686c9c8dafca49c053d.zip
swift-a507a88a189bb603c9f2d686c9c8dafca49c053d.tar.bz2
Update 3rdParty/Expat to version 2.2.1
Test-Information: Build successfully on macOS 10.12.5 with clang trunk and ./scons test=all try_libxml=no try_expat=no . Change-Id: I0cc0680086ea40f92bbfa7296d10beb08cc657e3
Diffstat (limited to '3rdParty/Expat/src/xmltok.c')
-rw-r--r--3rdParty/Expat/src/xmltok.c244
1 files changed, 170 insertions, 74 deletions
diff --git a/3rdParty/Expat/src/xmltok.c b/3rdParty/Expat/src/xmltok.c
index bf09dfc..cdf0720 100644
--- a/3rdParty/Expat/src/xmltok.c
+++ b/3rdParty/Expat/src/xmltok.c
@@ -1,254 +1,259 @@
/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
See the file COPYING for copying permission.
*/
#include <stddef.h>
-#ifdef COMPILED_FROM_DSP
+#ifdef _WIN32
#include "winconfig.h"
-#elif defined(MACOS_CLASSIC)
-#include "macconfig.h"
-#elif defined(__amigaos__)
-#include "amigaconfig.h"
-#elif defined(__WATCOMC__)
-#include "watcomconfig.h"
#else
#ifdef HAVE_EXPAT_CONFIG_H
#include <expat_config.h>
#endif
-#endif /* ndef COMPILED_FROM_DSP */
+#endif /* ndef _WIN32 */
#include "expat_external.h"
#include "internal.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)))
+ (namingBitmap[(pages[hi] << 3) + ((lo) >> 5)] & (1u << ((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)))
+ & (1u << (((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)))
+ & (1u << (((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))
/* 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_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 PTRFASTCALL
-isNever(const ENCODING *enc, const char *p)
+isNever(const ENCODING *UNUSED_P(enc), const char *UNUSED_P(p))
{
return 0;
}
static int PTRFASTCALL
-utf8_isName2(const ENCODING *enc, const char *p)
+utf8_isName2(const ENCODING *UNUSED_P(enc), const char *p)
{
return UTF8_GET_NAMING2(namePages, (const unsigned char *)p);
}
static int PTRFASTCALL
-utf8_isName3(const ENCODING *enc, const char *p)
+utf8_isName3(const ENCODING *UNUSED_P(enc), const char *p)
{
return UTF8_GET_NAMING3(namePages, (const unsigned char *)p);
}
#define utf8_isName4 isNever
static int PTRFASTCALL
-utf8_isNmstrt2(const ENCODING *enc, const char *p)
+utf8_isNmstrt2(const ENCODING *UNUSED_P(enc), const char *p)
{
return UTF8_GET_NAMING2(nmstrtPages, (const unsigned char *)p);
}
static int PTRFASTCALL
-utf8_isNmstrt3(const ENCODING *enc, const char *p)
+utf8_isNmstrt3(const ENCODING *UNUSED_P(enc), const char *p)
{
return UTF8_GET_NAMING3(nmstrtPages, (const unsigned char *)p);
}
#define utf8_isNmstrt4 isNever
static int PTRFASTCALL
-utf8_isInvalid2(const ENCODING *enc, const char *p)
+utf8_isInvalid2(const ENCODING *UNUSED_P(enc), const char *p)
{
return UTF8_INVALID2((const unsigned char *)p);
}
static int PTRFASTCALL
-utf8_isInvalid3(const ENCODING *enc, const char *p)
+utf8_isInvalid3(const ENCODING *UNUSED_P(enc), const char *p)
{
return UTF8_INVALID3((const unsigned char *)p);
}
static int PTRFASTCALL
-utf8_isInvalid4(const ENCODING *enc, const char *p)
+utf8_isInvalid4(const ENCODING *UNUSED_P(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 (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 (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) \
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
+#define NULL_VTABLE \
+ /* isName2 */ NULL, \
+ /* isName3 */ NULL, \
+ /* isName4 */ NULL, \
+ /* isNmstrt2 */ NULL, \
+ /* isNmstrt3 */ NULL, \
+ /* isNmstrt4 */ NULL, \
+ /* isInvalid2 */ NULL, \
+ /* isInvalid3 */ NULL, \
+ /* isInvalid4 */ NULL
+
static int FASTCALL 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 PTRFASTCALL
sb_byteType(const ENCODING *enc, const char *p)
{
return SB_BYTE_TYPE(enc, p);
}
#define BYTE_TYPE(enc, p) \
(AS_NORMAL_ENCODING(enc)->byteType(enc, p))
#else
#define BYTE_TYPE(enc, p) SB_BYTE_TYPE(enc, p)
@@ -291,352 +296,438 @@ 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
#define XML_TOK_IMPL_C
#include "xmltok_impl.c"
#undef XML_TOK_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 PTRCALL
-utf8_toUtf8(const ENCODING *enc,
+void
+align_limit_to_full_utf8_characters(const char * from, const char ** fromLimRef)
+{
+ const char * fromLim = *fromLimRef;
+ size_t walked = 0;
+ for (; fromLim > from; fromLim--, walked++) {
+ const unsigned char prev = (unsigned char)fromLim[-1];
+ if ((prev & 0xf8u) == 0xf0u) { /* 4-byte character, lead by 0b11110xxx byte */
+ if (walked + 1 >= 4) {
+ fromLim += 4 - 1;
+ break;
+ } else {
+ walked = 0;
+ }
+ } else if ((prev & 0xf0u) == 0xe0u) { /* 3-byte character, lead by 0b1110xxxx byte */
+ if (walked + 1 >= 3) {
+ fromLim += 3 - 1;
+ break;
+ } else {
+ walked = 0;
+ }
+ } else if ((prev & 0xe0u) == 0xc0u) { /* 2-byte character, lead by 0b110xxxxx byte */
+ if (walked + 1 >= 2) {
+ fromLim += 2 - 1;
+ break;
+ } else {
+ walked = 0;
+ }
+ } else if ((prev & 0x80u) == 0x00u) { /* 1-byte character, matching 0b0xxxxxxx */
+ break;
+ }
+ }
+ *fromLimRef = fromLim;
+}
+
+static enum XML_Convert_Result PTRCALL
+utf8_toUtf8(const ENCODING *UNUSED_P(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++)
+ const char *fromLimInitial = fromLim;
+
+ /* Avoid copying partial characters. */
+ align_limit_to_full_utf8_characters(*fromP, &fromLim);
+
+ for (to = *toP, from = *fromP; (from < fromLim) && (to < toLim); from++, to++)
*to = *from;
*fromP = from;
*toP = to;
+
+ if (fromLim < fromLimInitial)
+ return XML_CONVERT_INPUT_INCOMPLETE;
+ else if ((to == toLim) && (from < fromLim))
+ return XML_CONVERT_OUTPUT_EXHAUSTED;
+ else
+ return XML_CONVERT_COMPLETED;
}
-static void PTRCALL
+static enum XML_Convert_Result PTRCALL
utf8_toUtf16(const ENCODING *enc,
const char **fromP, const char *fromLim,
unsigned short **toP, const unsigned short *toLim)
{
+ enum XML_Convert_Result res = XML_CONVERT_COMPLETED;
unsigned short *to = *toP;
const char *from = *fromP;
- while (from != fromLim && to != toLim) {
+ while (from < fromLim && to < toLim) {
switch (((struct normal_encoding *)enc)->type[(unsigned char)*from]) {
case BT_LEAD2:
+ if (fromLim - from < 2) {
+ res = XML_CONVERT_INPUT_INCOMPLETE;
+ goto after;
+ }
*to++ = (unsigned short)(((from[0] & 0x1f) << 6) | (from[1] & 0x3f));
from += 2;
break;
case BT_LEAD3:
+ if (fromLim - from < 3) {
+ res = XML_CONVERT_INPUT_INCOMPLETE;
+ goto after;
+ }
*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)
+ if (toLim - to < 2) {
+ res = XML_CONVERT_OUTPUT_EXHAUSTED;
+ goto after;
+ }
+ if (fromLim - from < 4) {
+ res = XML_CONVERT_INPUT_INCOMPLETE;
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:
*to++ = *from++;
break;
}
}
+ if (from < fromLim)
+ res = XML_CONVERT_OUTPUT_EXHAUSTED;
after:
*fromP = from;
*toP = to;
+ return res;
}
#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 PTRCALL
-latin1_toUtf8(const ENCODING *enc,
+static enum XML_Convert_Result PTRCALL
+latin1_toUtf8(const ENCODING *UNUSED_P(enc),
const char **fromP, const char *fromLim,
char **toP, const char *toLim)
{
for (;;) {
unsigned char c;
if (*fromP == fromLim)
- break;
+ return XML_CONVERT_COMPLETED;
c = (unsigned char)**fromP;
if (c & 0x80) {
if (toLim - *toP < 2)
- break;
+ return XML_CONVERT_OUTPUT_EXHAUSTED;
*(*toP)++ = (char)((c >> 6) | UTF8_cval2);
*(*toP)++ = (char)((c & 0x3f) | 0x80);
(*fromP)++;
}
else {
if (*toP == toLim)
- break;
+ return XML_CONVERT_OUTPUT_EXHAUSTED;
*(*toP)++ = *(*fromP)++;
}
}
}
-static void PTRCALL
-latin1_toUtf16(const ENCODING *enc,
+static enum XML_Convert_Result PTRCALL
+latin1_toUtf16(const ENCODING *UNUSED_P(enc),
const char **fromP, const char *fromLim,
unsigned short **toP, const unsigned short *toLim)
{
- while (*fromP != fromLim && *toP != toLim)
+ while (*fromP < fromLim && *toP < toLim)
*(*toP)++ = (unsigned char)*(*fromP)++;
+
+ if ((*toP == toLim) && (*fromP < fromLim))
+ return XML_CONVERT_OUTPUT_EXHAUSTED;
+ else
+ return XML_CONVERT_COMPLETED;
}
#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_)
+ STANDARD_VTABLE(sb_) NULL_VTABLE
};
#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_)
+ STANDARD_VTABLE(sb_) NULL_VTABLE
};
-static void PTRCALL
-ascii_toUtf8(const ENCODING *enc,
+static enum XML_Convert_Result PTRCALL
+ascii_toUtf8(const ENCODING *UNUSED_P(enc),
const char **fromP, const char *fromLim,
char **toP, const char *toLim)
{
- while (*fromP != fromLim && *toP != toLim)
+ while (*fromP < fromLim && *toP < toLim)
*(*toP)++ = *(*fromP)++;
+
+ if ((*toP == toLim) && (*fromP < fromLim))
+ return XML_CONVERT_OUTPUT_EXHAUSTED;
+ else
+ return XML_CONVERT_COMPLETED;
}
#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_)
+ STANDARD_VTABLE(sb_) NULL_VTABLE
};
#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_)
+ STANDARD_VTABLE(sb_) NULL_VTABLE
};
static int PTRFASTCALL
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 PTRCALL \
-E ## toUtf8(const ENCODING *enc, \
+static enum XML_Convert_Result PTRCALL \
+E ## toUtf8(const ENCODING *UNUSED_P(enc), \
const char **fromP, const char *fromLim, \
char **toP, const char *toLim) \
{ \
- const char *from; \
- for (from = *fromP; from != fromLim; from += 2) { \
+ const char *from = *fromP; \
+ fromLim = from + (((fromLim - from) >> 1) << 1); /* shrink to even */ \
+ for (; 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; \
+ return XML_CONVERT_OUTPUT_EXHAUSTED; \
} \
*(*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; \
+ return XML_CONVERT_OUTPUT_EXHAUSTED; \
} \
*(*toP)++ = ((lo >> 6) | (hi << 2) | UTF8_cval2); \
*(*toP)++ = ((lo & 0x3f) | 0x80); \
break; \
default: \
if (toLim - *toP < 3) { \
*fromP = from; \
- return; \
+ return XML_CONVERT_OUTPUT_EXHAUSTED; \
} \
/* 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; \
+ return XML_CONVERT_OUTPUT_EXHAUSTED; \
+ } \
+ if (fromLim - from < 4) { \
+ *fromP = from; \
+ return XML_CONVERT_INPUT_INCOMPLETE; \
} \
plane = (((hi & 0x3) << 2) | ((lo >> 6) & 0x3)) + 1; \
*(*toP)++ = ((plane >> 2) | UTF8_cval4); \
*(*toP)++ = (((lo >> 2) & 0xF) | ((plane & 0x3) << 4) | 0x80); \
from += 2; \
lo2 = GET_LO(from); \
*(*toP)++ = (((lo & 0x3) << 4) \
| ((GET_HI(from) & 0x3) << 2) \
| (lo2 >> 6) \
| 0x80); \
*(*toP)++ = ((lo2 & 0x3f) | 0x80); \
break; \
} \
} \
*fromP = from; \
+ if (from < fromLim) \
+ return XML_CONVERT_INPUT_INCOMPLETE; \
+ else \
+ return XML_CONVERT_COMPLETED; \
}
#define DEFINE_UTF16_TO_UTF16(E) \
-static void PTRCALL \
-E ## toUtf16(const ENCODING *enc, \
+static enum XML_Convert_Result PTRCALL \
+E ## toUtf16(const ENCODING *UNUSED_P(enc), \
const char **fromP, const char *fromLim, \
unsigned short **toP, const unsigned short *toLim) \
{ \
+ enum XML_Convert_Result res = XML_CONVERT_COMPLETED; \
+ fromLim = *fromP + (((fromLim - *fromP) >> 1) << 1); /* shrink to even */ \
/* Avoid copying first half only of surrogate */ \
if (fromLim - *fromP > ((toLim - *toP) << 1) \
- && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) \
+ && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) { \
fromLim -= 2; \
- for (; *fromP != fromLim && *toP != toLim; *fromP += 2) \
+ res = XML_CONVERT_INPUT_INCOMPLETE; \
+ } \
+ for (; *fromP < fromLim && *toP < toLim; *fromP += 2) \
*(*toP)++ = (GET_HI(*fromP) << 8) | GET_LO(*fromP); \
+ if ((*toP == toLim) && (*fromP < fromLim)) \
+ return XML_CONVERT_OUTPUT_EXHAUSTED; \
+ else \
+ return res; \
}
#define SET2(ptr, ch) \
(((ptr)[0] = ((ch) & 0xff)), ((ptr)[1] = ((ch) >> 8)))
#define GET_LO(ptr) ((unsigned char)(ptr)[0])
#define GET_HI(ptr) ((unsigned char)(ptr)[1])
DEFINE_UTF16_TO_UTF8(little2_)
DEFINE_UTF16_TO_UTF16(little2_)
#undef SET2
#undef GET_LO
#undef GET_HI
#define SET2(ptr, ch) \
(((ptr)[0] = ((ch) >> 8)), ((ptr)[1] = ((ch) & 0xFF)))
#define GET_LO(ptr) ((unsigned char)(ptr)[1])
#define GET_HI(ptr) ((unsigned char)(ptr)[0])
DEFINE_UTF16_TO_UTF8(big2_)
DEFINE_UTF16_TO_UTF16(big2_)
#undef SET2
#undef GET_LO
#undef GET_HI
#define LITTLE2_BYTE_TYPE(enc, p) \
((p)[1] == 0 \
? ((struct normal_encoding *)(enc))->type[(unsigned char)*(p)] \
: unicode_byte_type((p)[1], (p)[0]))
@@ -699,106 +790,106 @@ little2_isNmstrtMin(const ENCODING *enc, const char *p)
#define XML_TOK_IMPL_C
#include "xmltok_impl.c"
#undef XML_TOK_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
#endif /* not XML_MIN_SIZE */
#ifdef XML_NS
static const struct normal_encoding little2_encoding_ns = {
{ VTABLE, 2, 0,
#if BYTEORDER == 1234
1
#else
0
#endif
},
{
#include "asciitab.h"
#include "latin1tab.h"
},
- STANDARD_VTABLE(little2_)
+ STANDARD_VTABLE(little2_) NULL_VTABLE
};
#endif
static const struct normal_encoding little2_encoding = {
{ VTABLE, 2, 0,
#if BYTEORDER == 1234
1
#else
0
#endif
},
{
#define BT_COLON BT_NMSTRT
#include "asciitab.h"
#undef BT_COLON
#include "latin1tab.h"
},
- STANDARD_VTABLE(little2_)
+ STANDARD_VTABLE(little2_) NULL_VTABLE
};
#if BYTEORDER != 4321
#ifdef XML_NS
static const struct normal_encoding internal_little2_encoding_ns = {
{ VTABLE, 2, 0, 1 },
{
#include "iasciitab.h"
#include "latin1tab.h"
},
- STANDARD_VTABLE(little2_)
+ STANDARD_VTABLE(little2_) NULL_VTABLE
};
#endif
static const struct normal_encoding internal_little2_encoding = {
{ VTABLE, 2, 0, 1 },
{
#define BT_COLON BT_NMSTRT
#include "iasciitab.h"
#undef BT_COLON
#include "latin1tab.h"
},
- STANDARD_VTABLE(little2_)
+ STANDARD_VTABLE(little2_) NULL_VTABLE
};
#endif
#define BIG2_BYTE_TYPE(enc, p) \
((p)[0] == 0 \
? ((struct normal_encoding *)(enc))->type[(unsigned char)(p)[1]] \
: unicode_byte_type((p)[0], (p)[1]))
#define BIG2_BYTE_TO_ASCII(enc, p) ((p)[0] == 0 ? (p)[1] : -1)
#define BIG2_CHAR_MATCHES(enc, p, c) ((p)[0] == 0 && (p)[1] == c)
#define BIG2_IS_NAME_CHAR_MINBPC(enc, p) \
UCS2_GET_NAMING(namePages, (unsigned char)p[0], (unsigned char)p[1])
#define BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p) \
UCS2_GET_NAMING(nmstrtPages, (unsigned char)p[0], (unsigned char)p[1])
#ifdef XML_MIN_SIZE
static int PTRFASTCALL
big2_byteType(const ENCODING *enc, const char *p)
{
return BIG2_BYTE_TYPE(enc, p);
}
static int PTRFASTCALL
big2_byteToAscii(const ENCODING *enc, const char *p)
{
return BIG2_BYTE_TO_ASCII(enc, p);
}
@@ -840,132 +931,132 @@ big2_isNmstrtMin(const ENCODING *enc, const char *p)
#define XML_TOK_IMPL_C
#include "xmltok_impl.c"
#undef XML_TOK_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
#endif /* not XML_MIN_SIZE */
#ifdef XML_NS
static const struct normal_encoding big2_encoding_ns = {
{ VTABLE, 2, 0,
#if BYTEORDER == 4321
1
#else
0
#endif
},
{
#include "asciitab.h"
#include "latin1tab.h"
},
- STANDARD_VTABLE(big2_)
+ STANDARD_VTABLE(big2_) NULL_VTABLE
};
#endif
static const struct normal_encoding big2_encoding = {
{ VTABLE, 2, 0,
#if BYTEORDER == 4321
1
#else
0
#endif
},
{
#define BT_COLON BT_NMSTRT
#include "asciitab.h"
#undef BT_COLON
#include "latin1tab.h"
},
- STANDARD_VTABLE(big2_)
+ STANDARD_VTABLE(big2_) NULL_VTABLE
};
#if BYTEORDER != 1234
#ifdef XML_NS
static const struct normal_encoding internal_big2_encoding_ns = {
{ VTABLE, 2, 0, 1 },
{
#include "iasciitab.h"
#include "latin1tab.h"
},
- STANDARD_VTABLE(big2_)
+ STANDARD_VTABLE(big2_) NULL_VTABLE
};
#endif
static const struct normal_encoding internal_big2_encoding = {
{ VTABLE, 2, 0, 1 },
{
#define BT_COLON BT_NMSTRT
#include "iasciitab.h"
#undef BT_COLON
#include "latin1tab.h"
},
- STANDARD_VTABLE(big2_)
+ STANDARD_VTABLE(big2_) NULL_VTABLE
};
#endif
#undef PREFIX
static int FASTCALL
streqci(const char *s1, const char *s2)
{
for (;;) {
char c1 = *s1++;
char c2 = *s2++;
if (ASCII_a <= c1 && c1 <= ASCII_z)
c1 += ASCII_A - ASCII_a;
if (ASCII_a <= c2 && c2 <= ASCII_z)
c2 += ASCII_A - ASCII_a;
if (c1 != c2)
return 0;
if (!c1)
break;
}
return 1;
}
static void PTRCALL
-initUpdatePosition(const ENCODING *enc, const char *ptr,
+initUpdatePosition(const ENCODING *UNUSED_P(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)
{
char buf[1];
char *p = buf;
XmlUtf8Convert(enc, &ptr, end, &p, p + 1);
if (p == buf)
return -1;
else
return buf[0];
}
static int FASTCALL
isSpace(int c)
{
switch (c) {
case 0x20:
case 0xD:
case 0xA:
case 0x9:
return 1;
}
return 0;
}
@@ -1261,112 +1352,117 @@ XmlSizeOfUnknownEncoding(void)
}
static int PTRFASTCALL
unknown_isName(const ENCODING *enc, const char *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 PTRFASTCALL
unknown_isNmstrt(const ENCODING *enc, const char *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 PTRFASTCALL
unknown_isInvalid(const ENCODING *enc, const char *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 PTRCALL
+static enum XML_Convert_Result 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;
+ return XML_CONVERT_COMPLETED;
utf8 = uenc->utf8[(unsigned char)**fromP];
n = *utf8++;
if (n == 0) {
int c = uenc->convert(uenc->userData, *fromP);
n = XmlUtf8Encode(c, buf);
if (n > toLim - *toP)
- break;
+ return XML_CONVERT_OUTPUT_EXHAUSTED;
utf8 = buf;
*fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP]
- (BT_LEAD2 - 2));
}
else {
if (n > toLim - *toP)
- break;
+ return XML_CONVERT_OUTPUT_EXHAUSTED;
(*fromP)++;
}
do {
*(*toP)++ = *utf8++;
} while (--n != 0);
}
}
-static void PTRCALL
+static enum XML_Convert_Result 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) {
+ while (*fromP < fromLim && *toP < toLim) {
unsigned short c = uenc->utf16[(unsigned char)**fromP];
if (c == 0) {
c = (unsigned short)
uenc->convert(uenc->userData, *fromP);
*fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP]
- (BT_LEAD2 - 2));
}
else
(*fromP)++;
*(*toP)++ = c;
}
+
+ if ((*toP == toLim) && (*fromP < fromLim))
+ return XML_CONVERT_OUTPUT_EXHAUSTED;
+ else
+ return XML_CONVERT_COMPLETED;
}
ENCODING *
XmlInitUnknownEncoding(void *mem,
int *table,
CONVERTER convert,
void *userData)
{
int i;
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)
return 0;
for (i = 0; i < 256; i++) {
int c = table[i];
if (c == -1) {
e->normal.type[i] = BT_MALFORM;
/* This shouldn't really get used. */
e->utf16[i] = 0xFFFF;
e->utf8[i][0] = 1;
e->utf8[i][1] = 0;
}
else if (c < 0) {
if (c < -4)
return 0;
e->normal.type[i] = (unsigned char)(BT_LEAD2 - (c + 2));
@@ -1476,61 +1572,61 @@ getEncodingIndex(const char *name)
if (streqci(name, encodingNames[i]))
return i;
return UNKNOWN_ENC;
}
/* 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.
*/
static int
initScan(const ENCODING * const *encodingTable,
const INIT_ENCODING *enc,
int state,
const char *ptr,
const char *end,
const char **nextTokPtr)
{
const ENCODING **encPtr;
- if (ptr == end)
+ if (ptr >= end)
return XML_TOK_NONE;
encPtr = enc->encPtr;
if (ptr + 1 == end) {
/* only a single byte available for auto-detection */
#ifndef XML_DTD /* FIXME */
/* a well-formed document entity must have more than one byte */
if (state != XML_CONTENT_STATE)
return XML_TOK_PARTIAL;
#endif
/* so we're parsing an external text entity... */
/* if UTF-16 was externally specified, then we need at least 2 bytes */
switch (INIT_ENC_INDEX(enc)) {
case UTF_16_ENC:
case UTF_16LE_ENC:
case UTF_16BE_ENC:
return XML_TOK_PARTIAL;
}
switch ((unsigned char)*ptr) {
case 0xFE:
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;
/* fall through */
case 0x00:
case 0x3C:
return XML_TOK_PARTIAL;
}
}