U_CFUNC void
ustr_u32cat(struct UString *dst, UChar32 c, UErrorCode *status){
if(c > 0x10FFFF){
*status = U_ILLEGAL_CHAR_FOUND;
return;
}
if(c >0xFFFF){
ustr_ucat(dst, U16_LEAD(c), status);
ustr_ucat(dst, U16_TRAIL(c), status);
}else{
ustr_ucat(dst, (UChar) c, status);
}
}
U_CFUNC void
ustr_initChars(struct UString *s, const char* source, int32_t length, UErrorCode *status)
{
int i = 0;
if (U_FAILURE(*status)) return;
s->fChars = 0;
s->fLength = s->fCapacity = 0;
if (length == -1) {
length = (int32_t)uprv_strlen(source);
}
if(s->fCapacity < length) {
ustr_resize(s, ALLOCATION(length), status);
if(U_FAILURE(*status)) return;
}
for (; i < length; i++)
{
UChar charToAppend;
u_charsToUChars(source+i, &charToAppend, 1);
ustr_ucat(s, charToAppend, status);
/*
#if U_CHARSET_FAMILY==U_ASCII_FAMILY
ustr_ucat(s, (UChar)(uint8_t)(source[i]), status);
#elif U_CHARSET_FAMILY==U_EBCDIC_FAMILY
ustr_ucat(s, (UChar)asciiFromEbcdic[(uint8_t)(*cs++)], status);
#else
# error U_CHARSET_FAMILY is not valid
#endif
*/
}
}
/* Copy a string token into the given UnicodeString. Upon entry, we
have already read the first character of the string token, which is
not a whitespace character (but may be a QUOTE or ESCAPE). This
function reads all subsequent characters that belong with this
string, and copy them into the token parameter. The other
important, and slightly convoluted purpose of this function is to
merge adjacent strings. It looks forward a bit, and if the next
non comment, non whitespace item is a string, it reads it in as
well. If two adjacent strings are quoted, they are merged without
intervening space. Otherwise a single SPACE character is
inserted. */
static enum ETokenType getStringToken(UCHARBUF* buf,
UChar32 initialChar,
struct UString *token,
UErrorCode *status) {
UBool lastStringWasQuoted;
UChar32 c;
UChar target[3] = { '\0' };
UChar *pTarget = target;
int len=0;
UBool isFollowingCharEscaped=FALSE;
UBool isNLUnescaped = FALSE;
UChar32 prevC=0;
/* We are guaranteed on entry that initialChar is not a whitespace
character. If we are at the EOF, or have some other problem, it
doesn't matter; we still want to validly return the initialChar
(if nothing else) as a string token. */
if (U_FAILURE(*status)) {
return TOK_ERROR;
}
/* setup */
lastStringWasQuoted = FALSE;
c = initialChar;
ustr_setlen(token, 0, status);
if (U_FAILURE(*status)) {
return TOK_ERROR;
}
for (;;) {
if (c == QUOTE) {
if (!lastStringWasQuoted && token->fLength > 0) {
ustr_ucat(token, SPACE, status);
if (U_FAILURE(*status)) {
return TOK_ERROR;
}
}
lastStringWasQuoted = TRUE;
for (;;) {
c = ucbuf_getc(buf,status);
/* EOF reached */
if (c == U_EOF) {
return TOK_EOF;
}
/* Unterminated quoted strings */
if (U_FAILURE(*status)) {
return TOK_ERROR;
}
if (c == QUOTE && !isFollowingCharEscaped) {
break;
}
if (c == ESCAPE && !isFollowingCharEscaped) {
pTarget = target;
c = unescape(buf, status);
if (c == U_ERR) {
return TOK_ERROR;
}
if(c == CR || c == LF){
isNLUnescaped = TRUE;
}
}
if(c==ESCAPE && !isFollowingCharEscaped){
isFollowingCharEscaped = TRUE;
}else{
U_APPEND_CHAR32(c, pTarget,len);
pTarget = target;
ustr_uscat(token, pTarget,len, status);
isFollowingCharEscaped = FALSE;
len=0;
if(c == CR || c == LF){
if(isNLUnescaped == FALSE && prevC!=CR){
lineCount++;
}
isNLUnescaped = FALSE;
}
}
if (U_FAILURE(*status)) {
return TOK_ERROR;
}
prevC = c;
}
} else {
if (token->fLength > 0) {
ustr_ucat(token, SPACE, status);
if (U_FAILURE(*status)) {
return TOK_ERROR;
}
}
if(lastStringWasQuoted){
if(getShowWarning()){
warning(lineCount, "Mixing quoted and unquoted strings");
}
if(isStrict()){
return TOK_ERROR;
}
}
lastStringWasQuoted = FALSE;
/* if we reach here we are mixing
* quoted and unquoted strings
* warn in normal mode and error in
* pedantic mode
*/
if (c == ESCAPE) {
pTarget = target;
c = unescape(buf, status);
/* EOF reached */
if (c == U_EOF) {
return TOK_ERROR;
}
}
U_APPEND_CHAR32(c, pTarget,len);
pTarget = target;
ustr_uscat(token, pTarget,len, status);
len=0;
if (U_FAILURE(*status)) {
return TOK_ERROR;
}
for (;;) {
/* DON'T skip whitespace */
c = getNextChar(buf, FALSE, NULL, status);
/* EOF reached */
if (c == U_EOF) {
ucbuf_ungetc(c, buf);
return TOK_STRING;
}
if (U_FAILURE(*status)) {
return TOK_STRING;
}
if (c == QUOTE
|| c == OPENBRACE
|| c == CLOSEBRACE
|| c == COMMA
|| c == COLON) {
ucbuf_ungetc(c, buf);
break;
}
if (isWhitespace(c)) {
break;
}
if (c == ESCAPE) {
pTarget = target;
c = unescape(buf, status);
if (c == U_ERR) {
return TOK_ERROR;
}
}
U_APPEND_CHAR32(c, pTarget,len);
pTarget = target;
ustr_uscat(token, pTarget,len, status);
len=0;
if (U_FAILURE(*status)) {
return TOK_ERROR;
}
}
}
/* DO skip whitespace */
c = getNextChar(buf, TRUE, NULL, status);
if (U_FAILURE(*status)) {
return TOK_STRING;
}
if (c == OPENBRACE || c == CLOSEBRACE || c == COMMA || c == COLON) {
ucbuf_ungetc(c, buf);
return TOK_STRING;
}
}
}
