static int PTRCALL
PREFIX(scanComment)(const ENCODING *enc, const char *ptr,
const char *end, const char **nextTokPtr)
{
if (ptr != end) {
if (!CHAR_MATCHES(enc, ptr, ASCII_MINUS)) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
ptr += MINBPC(enc);
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
INVALID_CASES(ptr, nextTokPtr)
case BT_MINUS:
if ((ptr += MINBPC(enc)) == end)
return XML_TOK_PARTIAL;
if (CHAR_MATCHES(enc, ptr, ASCII_MINUS)) {
if ((ptr += MINBPC(enc)) == end)
return XML_TOK_PARTIAL;
if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_COMMENT;
}
break;
default:
ptr += MINBPC(enc);
break;
}
}
}
return XML_TOK_PARTIAL;
}
static int PTRCALL
PREFIX(scanPi)(const ENCODING *enc, const char *ptr,
const char *end, const char **nextTokPtr)
{
int tok;
const char *target = ptr;
REQUIRE_CHAR(enc, ptr, end);
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
while (HAS_CHAR(enc, ptr, end)) {
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
case BT_S: case BT_CR: case BT_LF:
if (!PREFIX(checkPiTarget)(enc, target, ptr, &tok)) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
ptr += MINBPC(enc);
while (HAS_CHAR(enc, ptr, end)) {
switch (BYTE_TYPE(enc, ptr)) {
INVALID_CASES(ptr, nextTokPtr)
case BT_QUEST:
ptr += MINBPC(enc);
REQUIRE_CHAR(enc, ptr, end);
if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
*nextTokPtr = ptr + MINBPC(enc);
return tok;
}
break;
default:
ptr += MINBPC(enc);
break;
}
}
return XML_TOK_PARTIAL;
case BT_QUEST:
if (!PREFIX(checkPiTarget)(enc, target, ptr, &tok)) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
ptr += MINBPC(enc);
REQUIRE_CHAR(enc, ptr, end);
if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
*nextTokPtr = ptr + MINBPC(enc);
return tok;
}
/* fall through */
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
return XML_TOK_PARTIAL;
}
static int PTRCALL
PREFIX(predefinedEntityName)(const ENCODING *enc, const char *ptr,
const char *end)
{
UNUSED(enc);
switch ((end - ptr)/MINBPC(enc)) {
case 2:
if (CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_t)) {
switch (BYTE_TO_ASCII(enc, ptr)) {
case ASCII_l:
return ASCII_LT;
case ASCII_g:
return ASCII_GT;
}
}
break;
case 3:
if (CHAR_MATCHES(enc, ptr, ASCII_a)) {
ptr += MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_m)) {
ptr += MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_p))
return ASCII_AMP;
}
}
break;
case 4:
switch (BYTE_TO_ASCII(enc, ptr)) {
case ASCII_q:
ptr += MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_u)) {
ptr += MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_o)) {
ptr += MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_t))
return ASCII_QUOT;
}
}
break;
case ASCII_a:
ptr += MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_p)) {
ptr += MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_o)) {
ptr += MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_s))
return ASCII_APOS;
}
}
break;
}
}
return 0;
}
static
int PREFIX(predefinedEntityName)(const ENCODING *enc, const char *ptr, const char *end)
{
(void)enc;
switch (end - ptr) {
case 2 * MINBPC:
if (CHAR_MATCHES(enc, ptr + MINBPC, 't')) {
switch (BYTE_TO_ASCII(enc, ptr)) {
case 'l':
return '<';
case 'g':
return '>';
}
}
break;
case 3 * MINBPC:
if (CHAR_MATCHES(enc, ptr, 'a')) {
ptr += MINBPC;
if (CHAR_MATCHES(enc, ptr, 'm')) {
ptr += MINBPC;
if (CHAR_MATCHES(enc, ptr, 'p'))
return '&';
}
}
break;
case 4 * MINBPC:
switch (BYTE_TO_ASCII(enc, ptr)) {
case 'q':
ptr += MINBPC;
if (CHAR_MATCHES(enc, ptr, 'u')) {
ptr += MINBPC;
if (CHAR_MATCHES(enc, ptr, 'o')) {
ptr += MINBPC;
if (CHAR_MATCHES(enc, ptr, 't'))
return '"';
}
}
break;
case 'a':
ptr += MINBPC;
if (CHAR_MATCHES(enc, ptr, 'p')) {
ptr += MINBPC;
if (CHAR_MATCHES(enc, ptr, 'o')) {
ptr += MINBPC;
if (CHAR_MATCHES(enc, ptr, 's'))
return '\'';
}
}
break;
}
}
return 0;
}
static int PTRFASTCALL
PREFIX(charRefNumber)(const ENCODING *enc, const char *ptr)
{
int result = 0;
/* skip &# */
ptr += 2*MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_x)) {
for (ptr += MINBPC(enc);
!CHAR_MATCHES(enc, ptr, ASCII_SEMI);
ptr += MINBPC(enc)) {
int c = BYTE_TO_ASCII(enc, ptr);
switch (c) {
case ASCII_0: case ASCII_1: case ASCII_2: case ASCII_3: case ASCII_4:
case ASCII_5: case ASCII_6: case ASCII_7: case ASCII_8: case ASCII_9:
result <<= 4;
result |= (c - ASCII_0);
break;
case ASCII_A: case ASCII_B: case ASCII_C:
case ASCII_D: case ASCII_E: case ASCII_F:
result <<= 4;
result += 10 + (c - ASCII_A);
break;
case ASCII_a: case ASCII_b: case ASCII_c:
case ASCII_d: case ASCII_e: case ASCII_f:
result <<= 4;
result += 10 + (c - ASCII_a);
break;
}
if (result >= 0x110000)
return -1;
}
}
else {
for (; !CHAR_MATCHES(enc, ptr, ASCII_SEMI); ptr += MINBPC(enc)) {
int c = BYTE_TO_ASCII(enc, ptr);
result *= 10;
result += (c - ASCII_0);
if (result >= 0x110000)
return -1;
}
}
// Some faggots compose mails with ancient control symbols in headers.
// Libexpat tokenizer breaks on them (so far 'Form Feed' and 'DC2' were found).
if (result == 0x0C) {
result = 0x0A;
}
else if (result == 0x12) {
result = ' ';
}
return checkCharRefNumber(result);
}
static int PTRCALL
PREFIX(ignoreSectionTok)(const ENCODING *enc, const char *ptr,
const char *end, const char **nextTokPtr)
{
int level = 0;
if (MINBPC(enc) > 1) {
size_t n = end - ptr;
if (n & (MINBPC(enc) - 1)) {
n &= ~(MINBPC(enc) - 1);
end = ptr + n;
}
}
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
INVALID_CASES(ptr, nextTokPtr)
case BT_LT:
if ((ptr += MINBPC(enc)) == end)
return XML_TOK_PARTIAL;
if (CHAR_MATCHES(enc, ptr, ASCII_EXCL)) {
if ((ptr += MINBPC(enc)) == end)
return XML_TOK_PARTIAL;
if (CHAR_MATCHES(enc, ptr, ASCII_LSQB)) {
++level;
ptr += MINBPC(enc);
}
}
break;
case BT_RSQB:
if ((ptr += MINBPC(enc)) == end)
return XML_TOK_PARTIAL;
if (CHAR_MATCHES(enc, ptr, ASCII_RSQB)) {
if ((ptr += MINBPC(enc)) == end)
return XML_TOK_PARTIAL;
if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
ptr += MINBPC(enc);
if (level == 0) {
*nextTokPtr = ptr;
return XML_TOK_IGNORE_SECT;
}
--level;
}
}
break;
default:
ptr += MINBPC(enc);
break;
}
}
return XML_TOK_PARTIAL;
}
static int PTRCALL
PREFIX(scanCharRef)(const ENCODING *enc, const char *ptr,
const char *end, const char **nextTokPtr)
{
if (ptr != end) {
if (CHAR_MATCHES(enc, ptr, ASCII_x))
return PREFIX(scanHexCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
switch (BYTE_TYPE(enc, ptr)) {
case BT_DIGIT:
break;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
switch (BYTE_TYPE(enc, ptr)) {
case BT_DIGIT:
break;
case BT_SEMI:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_CHAR_REF;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
}
return XML_TOK_PARTIAL;
}
static
int PREFIX(nameMatchesAscii)(const ENCODING *enc, const char *ptr1, const char *ptr2)
{
for (; *ptr2; ptr1 += MINBPC(enc), ptr2++) {
if (!CHAR_MATCHES(enc, ptr1, *ptr2))
return 0;
}
switch (BYTE_TYPE(enc, ptr1)) {
case BT_LEAD2:
case BT_LEAD3:
case BT_LEAD4:
case BT_NONASCII:
case BT_NMSTRT:
#ifdef XML_NS
case BT_COLON:
#endif
case BT_HEX:
case BT_DIGIT:
case BT_NAME:
case BT_MINUS:
return 0;
default:
return 1;
}
}
static int PTRFASTCALL
PREFIX(charRefNumber)(const ENCODING *enc, const char *ptr)
{
int result = 0;
UNUSED(enc);
/* skip &# */
ptr += 2*MINBPC(enc);
if (CHAR_MATCHES(enc, ptr, ASCII_x)) {
for (ptr += MINBPC(enc);
!CHAR_MATCHES(enc, ptr, ASCII_SEMI);
ptr += MINBPC(enc)) {
int c = BYTE_TO_ASCII(enc, ptr);
switch (c) {
case ASCII_0: case ASCII_1: case ASCII_2: case ASCII_3: case ASCII_4:
case ASCII_5: case ASCII_6: case ASCII_7: case ASCII_8: case ASCII_9:
result <<= 4;
result |= (c - ASCII_0);
break;
case ASCII_A: case ASCII_B: case ASCII_C:
case ASCII_D: case ASCII_E: case ASCII_F:
result <<= 4;
result += 10 + (c - ASCII_A);
break;
case ASCII_a: case ASCII_b: case ASCII_c:
case ASCII_d: case ASCII_e: case ASCII_f:
result <<= 4;
result += 10 + (c - ASCII_a);
break;
}
if (result >= 0x110000)
return -1;
}
}
else {
for (; !CHAR_MATCHES(enc, ptr, ASCII_SEMI); ptr += MINBPC(enc)) {
int c = BYTE_TO_ASCII(enc, ptr);
result *= 10;
result += (c - ASCII_0);
if (result >= 0x110000)
return -1;
}
}
return checkCharRefNumber(result);
}
static int PTRCALL
PREFIX(isPublicId)(const ENCODING *enc, const char *ptr, const char *end,
const char **badPtr)
{
ptr += MINBPC(enc);
end -= MINBPC(enc);
for (; HAS_CHAR(enc, ptr, end); ptr += MINBPC(enc)) {
switch (BYTE_TYPE(enc, ptr)) {
case BT_DIGIT:
case BT_HEX:
case BT_MINUS:
case BT_APOS:
case BT_LPAR:
case BT_RPAR:
case BT_PLUS:
case BT_COMMA:
case BT_SOL:
case BT_EQUALS:
case BT_QUEST:
case BT_CR:
case BT_LF:
case BT_SEMI:
case BT_EXCL:
case BT_AST:
case BT_PERCNT:
case BT_NUM:
#ifdef XML_NS
case BT_COLON:
#endif
break;
case BT_S:
if (CHAR_MATCHES(enc, ptr, ASCII_TAB)) {
*badPtr = ptr;
return 0;
}
break;
case BT_NAME:
case BT_NMSTRT:
if (!(BYTE_TO_ASCII(enc, ptr) & ~0x7f))
break;
/* fall through */
default:
switch (BYTE_TO_ASCII(enc, ptr)) {
case 0x24: /* $ */
case 0x40: /* @ */
break;
default:
*badPtr = ptr;
return 0;
}
break;
}
}
return 1;
}
static int PTRCALL
PREFIX(nameMatchesAscii)(const ENCODING *enc, const char *ptr1,
const char *end1, const char *ptr2)
{
for (; *ptr2; ptr1 += MINBPC(enc), ptr2++) {
if (ptr1 == end1)
return 0;
if (!CHAR_MATCHES(enc, ptr1, *ptr2))
return 0;
}
return ptr1 == end1;
}
static
int PREFIX(charRefNumber)(const ENCODING *enc, const char *ptr)
{
int result = 0;
(void)enc;
/* skip &# */
ptr += 2*MINBPC;
if (CHAR_MATCHES(enc, ptr, 'x')) {
for (ptr += MINBPC; !CHAR_MATCHES(enc, ptr, ';'); ptr += MINBPC) {
int c = BYTE_TO_ASCII(enc, ptr);
switch (c) {
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
result <<= 4;
result |= (c - '0');
break;
case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
result <<= 4;
result += 10 + (c - 'A');
break;
case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
result <<= 4;
result += 10 + (c - 'a');
break;
}
if (result >= 0x110000)
return -1;
}
}
else {
for (; !CHAR_MATCHES(enc, ptr, ';'); ptr += MINBPC) {
int c = BYTE_TO_ASCII(enc, ptr);
result *= 10;
result += (c - '0');
if (result >= 0x110000)
return -1;
}
}
return checkCharRefNumber(result);
}
static
int PREFIX(isPublicId)(const ENCODING *enc, const char *ptr, const char *end,
const char **badPtr)
{
ptr += MINBPC;
end -= MINBPC;
for (; ptr != end; ptr += MINBPC) {
switch (BYTE_TYPE(enc, ptr)) {
case BT_DIGIT:
case BT_HEX:
case BT_MINUS:
case BT_APOS:
case BT_LPAR:
case BT_RPAR:
case BT_PLUS:
case BT_COMMA:
case BT_SOL:
case BT_EQUALS:
case BT_QUEST:
case BT_CR:
case BT_LF:
case BT_SEMI:
case BT_EXCL:
case BT_AST:
case BT_PERCNT:
case BT_NUM:
break;
case BT_S:
if (CHAR_MATCHES(enc, ptr, '\t')) {
*badPtr = ptr;
return 0;
}
break;
case BT_NAME:
case BT_NMSTRT:
if (!(BYTE_TO_ASCII(enc, ptr) & ~0x7f))
break;
default:
switch (BYTE_TO_ASCII(enc, ptr)) {
case 0x24: /* $ */
case 0x40: /* @ */
break;
default:
*badPtr = ptr;
return 0;
}
break;
}
}
return 1;
}
static
int PREFIX(scanCdataSection)(const ENCODING *enc, const char *ptr, const char *end,
const char **nextTokPtr)
{
int i;
/* CDATA[ */
if (end - ptr < 6 * MINBPC(enc))
return XML_TOK_PARTIAL;
for (i = 0; i < 6; i++, ptr += MINBPC(enc)) {
if (!CHAR_MATCHES(enc, ptr, "CDATA["[i])) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
*nextTokPtr = ptr;
return XML_TOK_CDATA_SECT_OPEN;
}
static int PTRCALL
PREFIX(nameMatchesAscii)(const ENCODING *UNUSED_P(enc), const char *ptr1,
const char *end1, const char *ptr2)
{
for (; *ptr2; ptr1 += MINBPC(enc), ptr2++) {
if (end1 - ptr1 < MINBPC(enc)) {
/* This line cannot be executed. The incoming data has already
* been tokenized once, so incomplete characters like this have
* already been eliminated from the input. Retaining the
* paranoia check is still valuable, however.
*/
return 0; /* LCOV_EXCL_LINE */
}
if (!CHAR_MATCHES(enc, ptr1, *ptr2))
return 0;
}
return ptr1 == end1;
}
static int PTRCALL
PREFIX(scanCdataSection)(const ENCODING *UNUSED_P(enc), const char *ptr,
const char *end, const char **nextTokPtr)
{
static const char CDATA_LSQB[] = { ASCII_C, ASCII_D, ASCII_A,
ASCII_T, ASCII_A, ASCII_LSQB };
int i;
/* CDATA[ */
REQUIRE_CHARS(enc, ptr, end, 6);
for (i = 0; i < 6; i++, ptr += MINBPC(enc)) {
if (!CHAR_MATCHES(enc, ptr, CDATA_LSQB[i])) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
*nextTokPtr = ptr;
return XML_TOK_CDATA_SECT_OPEN;
}
static
int PREFIX(scanCdataSection)(const ENCODING *enc, const char *ptr, const char *end,
const char **nextTokPtr)
{
static const char CDATA_LSQB[] = { ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, ASCII_LSQB };
int i;
cmExpatUnused(enc);
/* CDATA[ */
if (end - ptr < 6 * MINBPC(enc))
return XML_TOK_PARTIAL;
for (i = 0; i < 6; i++, ptr += MINBPC(enc)) {
if (!CHAR_MATCHES(enc, ptr, CDATA_LSQB[i])) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
*nextTokPtr = ptr;
return XML_TOK_CDATA_SECT_OPEN;
}
static int PTRCALL
PREFIX(contentTok)(const ENCODING *enc, const char *ptr, const char *end,
const char **nextTokPtr)
{
if (ptr == end)
return XML_TOK_NONE;
if (MINBPC(enc) > 1) {
size_t n = end - ptr;
if (n & (MINBPC(enc) - 1)) {
n &= ~(MINBPC(enc) - 1);
if (n == 0)
return XML_TOK_PARTIAL;
end = ptr + n;
}
}
switch (BYTE_TYPE(enc, ptr)) {
case BT_LT:
return PREFIX(scanLt)(enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_AMP:
return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_CR:
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_TRAILING_CR;
if (BYTE_TYPE(enc, ptr) == BT_LF)
ptr += MINBPC(enc);
*nextTokPtr = ptr;
return XML_TOK_DATA_NEWLINE;
case BT_LF:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_DATA_NEWLINE;
case BT_RSQB:
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_TRAILING_RSQB;
if (!CHAR_MATCHES(enc, ptr, ASCII_RSQB))
break;
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_TRAILING_RSQB;
if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
ptr -= MINBPC(enc);
break;
}
*nextTokPtr = ptr;
return XML_TOK_INVALID;
INVALID_CASES(ptr, nextTokPtr)
default:
ptr += MINBPC(enc);
break;
}
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: \
if (end - ptr < n || IS_INVALID_CHAR(enc, ptr, n)) { \
*nextTokPtr = ptr; \
return XML_TOK_DATA_CHARS; \
} \
ptr += n; \
break;
LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
#undef LEAD_CASE
case BT_RSQB:
if (ptr + MINBPC(enc) != end) {
if (!CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_RSQB)) {
ptr += MINBPC(enc);
break;
}
if (ptr + 2*MINBPC(enc) != end) {
if (!CHAR_MATCHES(enc, ptr + 2*MINBPC(enc), ASCII_GT)) {
ptr += MINBPC(enc);
break;
}
*nextTokPtr = ptr + 2*MINBPC(enc);
return XML_TOK_INVALID;
}
}
/* fall through */
case BT_AMP:
case BT_LT:
case BT_NONXML:
case BT_MALFORM:
case BT_TRAIL:
case BT_CR:
case BT_LF:
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
default:
ptr += MINBPC(enc);
break;
}
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
}
static int PTRCALL
PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end,
const char **nextTokPtr)
{
#ifdef XML_NS
int hadColon;
#endif
if (ptr == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
case BT_EXCL:
if ((ptr += MINBPC(enc)) == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
case BT_MINUS:
return PREFIX(scanComment)(enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_LSQB:
return PREFIX(scanCdataSection)(enc, ptr + MINBPC(enc),
end, nextTokPtr);
}
*nextTokPtr = ptr;
return XML_TOK_INVALID;
case BT_QUEST:
return PREFIX(scanPi)(enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_SOL:
return PREFIX(scanEndTag)(enc, ptr + MINBPC(enc), end, nextTokPtr);
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
#ifdef XML_NS
hadColon = 0;
#endif
/* we have a start-tag */
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
#ifdef XML_NS
case BT_COLON:
if (hadColon) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
hadColon = 1;
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
break;
#endif
case BT_S: case BT_CR: case BT_LF:
{
ptr += MINBPC(enc);
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
case BT_GT:
goto gt;
case BT_SOL:
goto sol;
case BT_S: case BT_CR: case BT_LF:
ptr += MINBPC(enc);
continue;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
return PREFIX(scanAtts)(enc, ptr, end, nextTokPtr);
}
return XML_TOK_PARTIAL;
}
case BT_GT:
gt:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_START_TAG_NO_ATTS;
case BT_SOL:
sol:
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_EMPTY_ELEMENT_NO_ATTS;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
return XML_TOK_PARTIAL;
}
static int PTRCALL
PREFIX(scanAtts)(const ENCODING *enc, const char *ptr, const char *end,
const char **nextTokPtr)
{
#ifdef XML_NS
int hadColon = 0;
#endif
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
#ifdef XML_NS
case BT_COLON:
if (hadColon) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
hadColon = 1;
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
break;
#endif
case BT_S: case BT_CR: case BT_LF:
for (;;) {
int t;
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
t = BYTE_TYPE(enc, ptr);
if (t == BT_EQUALS)
break;
switch (t) {
case BT_S:
case BT_LF:
case BT_CR:
break;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
/* fall through */
case BT_EQUALS:
{
int open;
#ifdef XML_NS
hadColon = 0;
#endif
for (;;) {
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
open = BYTE_TYPE(enc, ptr);
if (open == BT_QUOT || open == BT_APOS)
break;
switch (open) {
case BT_S:
case BT_LF:
case BT_CR:
break;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
ptr += MINBPC(enc);
/* in attribute value */
for (;;) {
int t;
if (ptr == end)
return XML_TOK_PARTIAL;
t = BYTE_TYPE(enc, ptr);
if (t == open)
break;
switch (t) {
INVALID_CASES(ptr, nextTokPtr)
case BT_AMP:
{
int tok = PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, &ptr);
if (tok <= 0) {
if (tok == XML_TOK_INVALID)
*nextTokPtr = ptr;
return tok;
}
break;
}
case BT_LT:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
default:
ptr += MINBPC(enc);
break;
}
}
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
case BT_S:
case BT_CR:
case BT_LF:
break;
case BT_SOL:
goto sol;
case BT_GT:
goto gt;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
/* ptr points to closing quote */
for (;;) {
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
case BT_S: case BT_CR: case BT_LF:
continue;
case BT_GT:
gt:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_START_TAG_WITH_ATTS;
case BT_SOL:
sol:
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_EMPTY_ELEMENT_WITH_ATTS;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
break;
}
break;
}
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
return XML_TOK_PARTIAL;
}
static int PTRCALL
PREFIX(cdataSectionTok)(const ENCODING *enc, const char *ptr,
const char *end, const char **nextTokPtr)
{
if (ptr == end)
return XML_TOK_NONE;
if (MINBPC(enc) > 1) {
size_t n = end - ptr;
if (n & (MINBPC(enc) - 1)) {
n &= ~(MINBPC(enc) - 1);
if (n == 0)
return XML_TOK_PARTIAL;
end = ptr + n;
}
}
switch (BYTE_TYPE(enc, ptr)) {
case BT_RSQB:
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
if (!CHAR_MATCHES(enc, ptr, ASCII_RSQB))
break;
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
ptr -= MINBPC(enc);
break;
}
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_CDATA_SECT_CLOSE;
case BT_CR:
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
if (BYTE_TYPE(enc, ptr) == BT_LF)
ptr += MINBPC(enc);
*nextTokPtr = ptr;
return XML_TOK_DATA_NEWLINE;
case BT_LF:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_DATA_NEWLINE;
INVALID_CASES(ptr, nextTokPtr)
default:
ptr += MINBPC(enc);
break;
}
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: \
if (end - ptr < n || IS_INVALID_CHAR(enc, ptr, n)) { \
*nextTokPtr = ptr; \
return XML_TOK_DATA_CHARS; \
} \
ptr += n; \
break;
LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
#undef LEAD_CASE
case BT_NONXML:
case BT_MALFORM:
case BT_TRAIL:
case BT_CR:
case BT_LF:
case BT_RSQB:
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
default:
ptr += MINBPC(enc);
break;
}
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
}
static int PTRCALL
PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end,
const char **nextTokPtr)
{
int tok;
if (ptr == end)
return XML_TOK_NONE;
if (MINBPC(enc) > 1) {
size_t n = end - ptr;
if (n & (MINBPC(enc) - 1)) {
n &= ~(MINBPC(enc) - 1);
if (n == 0)
return XML_TOK_PARTIAL;
end = ptr + n;
}
}
switch (BYTE_TYPE(enc, ptr)) {
case BT_QUOT:
return PREFIX(scanLit)(BT_QUOT, enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_APOS:
return PREFIX(scanLit)(BT_APOS, enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_LT:
{
ptr += MINBPC(enc);
if (ptr == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
case BT_EXCL:
return PREFIX(scanDecl)(enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_QUEST:
return PREFIX(scanPi)(enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_NMSTRT:
case BT_HEX:
case BT_NONASCII:
case BT_LEAD2:
case BT_LEAD3:
case BT_LEAD4:
*nextTokPtr = ptr - MINBPC(enc);
return XML_TOK_INSTANCE_START;
}
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
case BT_CR:
if (ptr + MINBPC(enc) == end) {
*nextTokPtr = end;
/* indicate that this might be part of a CR/LF pair */
return -XML_TOK_PROLOG_S;
}
/* fall through */
case BT_S: case BT_LF:
for (;;) {
ptr += MINBPC(enc);
if (ptr == end)
break;
switch (BYTE_TYPE(enc, ptr)) {
case BT_S: case BT_LF:
break;
case BT_CR:
/* don't split CR/LF pair */
if (ptr + MINBPC(enc) != end)
break;
/* fall through */
default:
*nextTokPtr = ptr;
return XML_TOK_PROLOG_S;
}
}
*nextTokPtr = ptr;
return XML_TOK_PROLOG_S;
case BT_PERCNT:
return PREFIX(scanPercent)(enc, ptr + MINBPC(enc), end, nextTokPtr);
case BT_COMMA:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_COMMA;
case BT_LSQB:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_OPEN_BRACKET;
case BT_RSQB:
ptr += MINBPC(enc);
if (ptr == end)
return -XML_TOK_CLOSE_BRACKET;
if (CHAR_MATCHES(enc, ptr, ASCII_RSQB)) {
if (ptr + MINBPC(enc) == end)
return XML_TOK_PARTIAL;
if (CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_GT)) {
*nextTokPtr = ptr + 2*MINBPC(enc);
return XML_TOK_COND_SECT_CLOSE;
}
}
*nextTokPtr = ptr;
return XML_TOK_CLOSE_BRACKET;
case BT_LPAR:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_OPEN_PAREN;
case BT_RPAR:
ptr += MINBPC(enc);
if (ptr == end)
return -XML_TOK_CLOSE_PAREN;
switch (BYTE_TYPE(enc, ptr)) {
case BT_AST:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_CLOSE_PAREN_ASTERISK;
case BT_QUEST:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_CLOSE_PAREN_QUESTION;
case BT_PLUS:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_CLOSE_PAREN_PLUS;
case BT_CR: case BT_LF: case BT_S:
case BT_GT: case BT_COMMA: case BT_VERBAR:
case BT_RPAR:
*nextTokPtr = ptr;
return XML_TOK_CLOSE_PAREN;
}
*nextTokPtr = ptr;
return XML_TOK_INVALID;
case BT_VERBAR:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_OR;
case BT_GT:
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_DECL_CLOSE;
case BT_NUM:
return PREFIX(scanPoundName)(enc, ptr + MINBPC(enc), end, nextTokPtr);
#define LEAD_CASE(n) \
case BT_LEAD ## n: \
if (end - ptr < n) \
return XML_TOK_PARTIAL_CHAR; \
if (IS_NMSTRT_CHAR(enc, ptr, n)) { \
ptr += n; \
tok = XML_TOK_NAME; \
break; \
} \
if (IS_NAME_CHAR(enc, ptr, n)) { \
ptr += n; \
tok = XML_TOK_NMTOKEN; \
break; \
} \
*nextTokPtr = ptr; \
return XML_TOK_INVALID;
LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
#undef LEAD_CASE
case BT_NMSTRT:
case BT_HEX:
tok = XML_TOK_NAME;
ptr += MINBPC(enc);
break;
case BT_DIGIT:
case BT_NAME:
case BT_MINUS:
#ifdef XML_NS
case BT_COLON:
#endif
tok = XML_TOK_NMTOKEN;
ptr += MINBPC(enc);
break;
case BT_NONASCII:
if (IS_NMSTRT_CHAR_MINBPC(enc, ptr)) {
ptr += MINBPC(enc);
tok = XML_TOK_NAME;
break;
}
if (IS_NAME_CHAR_MINBPC(enc, ptr)) {
ptr += MINBPC(enc);
tok = XML_TOK_NMTOKEN;
break;
}
/* fall through */
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
case BT_GT: case BT_RPAR: case BT_COMMA:
case BT_VERBAR: case BT_LSQB: case BT_PERCNT:
case BT_S: case BT_CR: case BT_LF:
*nextTokPtr = ptr;
return tok;
#ifdef XML_NS
case BT_COLON:
ptr += MINBPC(enc);
switch (tok) {
case XML_TOK_NAME:
if (ptr == end)
return XML_TOK_PARTIAL;
tok = XML_TOK_PREFIXED_NAME;
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
default:
tok = XML_TOK_NMTOKEN;
break;
}
break;
case XML_TOK_PREFIXED_NAME:
tok = XML_TOK_NMTOKEN;
break;
}
break;
#endif
case BT_PLUS:
if (tok == XML_TOK_NMTOKEN) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_NAME_PLUS;
case BT_AST:
if (tok == XML_TOK_NMTOKEN) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_NAME_ASTERISK;
case BT_QUEST:
if (tok == XML_TOK_NMTOKEN) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_NAME_QUESTION;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
return -tok;
}
static
int PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end,
const char **nextTokPtr)
{
if (ptr == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
case BT_EXCL:
if ((ptr += MINBPC) == end)
return XML_TOK_PARTIAL;
switch (BYTE_TYPE(enc, ptr)) {
case BT_MINUS:
return PREFIX(scanComment)(enc, ptr + MINBPC, end, nextTokPtr);
case BT_LSQB:
return PREFIX(scanCdataSection)(enc, ptr + MINBPC, end, nextTokPtr);
}
*nextTokPtr = ptr;
return XML_TOK_INVALID;
case BT_QUEST:
return PREFIX(scanPi)(enc, ptr + MINBPC, end, nextTokPtr);
case BT_SOL:
return PREFIX(scanEndTag)(enc, ptr + MINBPC, end, nextTokPtr);
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
/* we have a start-tag */
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
case BT_S: case BT_CR: case BT_LF:
{
ptr += MINBPC;
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
case BT_GT:
goto gt;
case BT_SOL:
goto sol;
case BT_S: case BT_CR: case BT_LF:
ptr += MINBPC;
continue;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
return PREFIX(scanAtts)(enc, ptr, end, nextTokPtr);
}
return XML_TOK_PARTIAL;
}
case BT_GT:
gt:
*nextTokPtr = ptr + MINBPC;
return XML_TOK_START_TAG_NO_ATTS;
case BT_SOL:
sol:
ptr += MINBPC;
if (ptr == end)
return XML_TOK_PARTIAL;
if (!CHAR_MATCHES(enc, ptr, '>')) {
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
*nextTokPtr = ptr + MINBPC;
return XML_TOK_EMPTY_ELEMENT_NO_ATTS;
default:
*nextTokPtr = ptr;
return XML_TOK_INVALID;
}
}
return XML_TOK_PARTIAL;
}
