static int PTRFASTCALL
PREFIX(nameLength)(const ENCODING *enc, const char *ptr)
{
const char *start = ptr;
for (;;) {
switch (BYTE_TYPE(enc, ptr)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: ptr += n; break;
LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
#undef LEAD_CASE
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:
ptr += MINBPC(enc);
break;
default:
return ptr - start;
}
}
}
static void PTRCALL
PREFIX(updatePosition)(const ENCODING *enc,
const char *ptr,
const char *end,
POSITION *pos)
{
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: \
ptr += n; \
break;
LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
#undef LEAD_CASE
case BT_LF:
pos->columnNumber = (unsigned)-1;
pos->lineNumber++;
ptr += MINBPC(enc);
break;
case BT_CR:
pos->lineNumber++;
ptr += MINBPC(enc);
if (ptr != end && BYTE_TYPE(enc, ptr) == BT_LF)
ptr += MINBPC(enc);
pos->columnNumber = (unsigned)-1;
break;
default:
ptr += MINBPC(enc);
break;
}
pos->columnNumber++;
}
}
static
int PREFIX(sameName)(const ENCODING *enc, const char *ptr1, const char *ptr2)
{
for (;;) {
switch (BYTE_TYPE(enc, ptr1)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: \
if (*ptr1++ != *ptr2++) \
return 0;
LEAD_CASE(4) LEAD_CASE(3) LEAD_CASE(2)
#undef LEAD_CASE
/* fall through */
if (*ptr1++ != *ptr2++)
return 0;
break;
case BT_NONASCII:
case BT_NMSTRT:
case BT_HEX:
case BT_DIGIT:
case BT_NAME:
case BT_MINUS:
if (*ptr2++ != *ptr1++)
return 0;
#if MINBPC > 1
if (*ptr2++ != *ptr1++)
return 0;
#if MINBPC > 2
if (*ptr2++ != *ptr1++)
return 0;
#if MINBPC > 3
if (*ptr2++ != *ptr1++)
return 0;
#endif
#endif
#endif
break;
default:
#if MINBPC == 1
if (*ptr1 == *ptr2)
return 1;
#endif
switch (BYTE_TYPE(enc, ptr2)) {
case BT_LEAD2:
case BT_LEAD3:
case BT_LEAD4:
case BT_NONASCII:
case BT_NMSTRT:
case BT_HEX:
case BT_DIGIT:
case BT_NAME:
case BT_MINUS:
return 0;
default:
return 1;
}
}
}
/* not reached */
}
static int PTRCALL
PREFIX(attributeValueTok)(const ENCODING *enc, const char *ptr,
const char *end, const char **nextTokPtr)
{
const char *start;
if (ptr == end)
return XML_TOK_NONE;
start = ptr;
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: ptr += n; break;
LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
#undef LEAD_CASE
case BT_AMP:
if (ptr == start)
return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_LT:
/* this is for inside entity references */
*nextTokPtr = ptr;
return XML_TOK_INVALID;
case BT_LF:
if (ptr == start) {
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_DATA_NEWLINE;
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_CR:
if (ptr == start) {
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;
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_S:
if (ptr == start) {
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_ATTRIBUTE_VALUE_S;
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
default:
ptr += MINBPC(enc);
break;
}
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
}
static int PTRCALL
PREFIX(entityValueTok)(const ENCODING *enc, const char *ptr,
const char *end, const char **nextTokPtr)
{
const char *start;
if (ptr == end)
return XML_TOK_NONE;
start = ptr;
while (ptr != end) {
switch (BYTE_TYPE(enc, ptr)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: ptr += n; break;
LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
#undef LEAD_CASE
case BT_AMP:
if (ptr == start)
return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_PERCNT:
if (ptr == start) {
int tok = PREFIX(scanPercent)(enc, ptr + MINBPC(enc),
end, nextTokPtr);
return ((tok == XML_TOK_PERCENT || tok == -XML_TOK_PERCENT)
? XML_TOK_INVALID : tok);
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_LF:
if (ptr == start) {
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_DATA_NEWLINE;
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_CR:
if (ptr == start) {
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;
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
default:
ptr += MINBPC(enc);
break;
}
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
}
static int PTRCALL
PREFIX (getAtts) (const ENCODING * enc, const char *ptr, int attsMax, ATTRIBUTE * atts) {
enum { other, inName, inValue } state = inName;
int nAtts = 0;
int open = 0; /* defined when state == inValue;
initialization just to shut up compilers */
for (ptr += MINBPC (enc);; ptr += MINBPC (enc)) {
switch (BYTE_TYPE (enc, ptr)) {
#define START_NAME \
if (state == other) { \
if (nAtts < attsMax) { \
atts[nAtts].name = ptr; \
atts[nAtts].normalized = 1; \
} \
state = inName; \
}
#define LEAD_CASE(n) \
case BT_LEAD ## n: START_NAME ptr += (n - MINBPC(enc)); break;
LEAD_CASE (2) LEAD_CASE (3) LEAD_CASE (4)
#undef LEAD_CASE
case BT_NONASCII:
case BT_NMSTRT:
case BT_HEX:
START_NAME break;
#undef START_NAME
case BT_QUOT:
if (state != inValue) {
if (nAtts < attsMax)
atts[nAtts].valuePtr = ptr + MINBPC (enc);
state = inValue;
open = BT_QUOT;
} else if (open == BT_QUOT) {
state = other;
if (nAtts < attsMax)
atts[nAtts].valueEnd = ptr;
nAtts++;
}
break;
case BT_APOS:
if (state != inValue) {
if (nAtts < attsMax)
atts[nAtts].valuePtr = ptr + MINBPC (enc);
state = inValue;
open = BT_APOS;
} else if (open == BT_APOS) {
state = other;
if (nAtts < attsMax)
atts[nAtts].valueEnd = ptr;
nAtts++;
}
break;
case BT_AMP:
if (nAtts < attsMax)
atts[nAtts].normalized = 0;
break;
case BT_S:
if (state == inName)
state = other;
else if (state == inValue
&& nAtts < attsMax
&& atts[nAtts].normalized
&& (ptr == atts[nAtts].valuePtr
|| BYTE_TO_ASCII (enc, ptr) != ASCII_SPACE
|| BYTE_TO_ASCII (enc, ptr + MINBPC (enc)) == ASCII_SPACE
|| BYTE_TYPE (enc, ptr + MINBPC (enc)) == open))
atts[nAtts].normalized = 0;
break;
case BT_CR:
case BT_LF:
/* This case ensures that the first attribute name is counted
Apart from that we could just change state on the quote. */
if (state == inName)
state = other;
else if (state == inValue && nAtts < attsMax)
atts[nAtts].normalized = 0;
break;
case BT_GT:
case BT_SOL:
if (state != inValue)
return nAtts;
break;
default:
break;
}
}
/* not reached */
}
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 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(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(sameName)(const ENCODING *enc, const char *ptr1, const char *ptr2)
{
for (;;) {
switch (BYTE_TYPE(enc, ptr1)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: \
if (*ptr1++ != *ptr2++) \
return 0;
LEAD_CASE(4) LEAD_CASE(3) LEAD_CASE(2)
#undef LEAD_CASE
/* fall through */
if (*ptr1++ != *ptr2++)
return 0;
break;
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:
if (*ptr2++ != *ptr1++)
return 0;
if (MINBPC(enc) > 1) {
if (*ptr2++ != *ptr1++)
return 0;
if (MINBPC(enc) > 2) {
if (*ptr2++ != *ptr1++)
return 0;
if (MINBPC(enc) > 3) {
if (*ptr2++ != *ptr1++)
return 0;
}
}
}
break;
default:
if (MINBPC(enc) == 1 && *ptr1 == *ptr2)
return 1;
switch (BYTE_TYPE(enc, ptr2)) {
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;
}
}
}
/* not reached */
}
static int PTRCALL
PREFIX(entityValueTok)(const ENCODING *enc, const char *ptr,
const char *end, const char **nextTokPtr)
{
const char *start;
if (ptr >= end)
return XML_TOK_NONE;
else if (! HAS_CHAR(enc, ptr, end)) {
/* 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 XML_TOK_PARTIAL; /* LCOV_EXCL_LINE */
}
start = ptr;
while (HAS_CHAR(enc, ptr, end)) {
switch (BYTE_TYPE(enc, ptr)) {
#define LEAD_CASE(n) \
case BT_LEAD ## n: ptr += n; break;
LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
#undef LEAD_CASE
case BT_AMP:
if (ptr == start)
return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_PERCNT:
if (ptr == start) {
int tok = PREFIX(scanPercent)(enc, ptr + MINBPC(enc),
end, nextTokPtr);
return (tok == XML_TOK_PERCENT) ? XML_TOK_INVALID : tok;
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_LF:
if (ptr == start) {
*nextTokPtr = ptr + MINBPC(enc);
return XML_TOK_DATA_NEWLINE;
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
case BT_CR:
if (ptr == start) {
ptr += MINBPC(enc);
if (! HAS_CHAR(enc, ptr, end))
return XML_TOK_TRAILING_CR;
if (BYTE_TYPE(enc, ptr) == BT_LF)
ptr += MINBPC(enc);
*nextTokPtr = ptr;
return XML_TOK_DATA_NEWLINE;
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
default:
ptr += MINBPC(enc);
break;
}
}
*nextTokPtr = ptr;
return XML_TOK_DATA_CHARS;
}
