/* Parse a long string or long comment (tv set to NULL). */
static void lex_longstring(LexState *ls, TValue *tv, int sep)
{
lex_savenext(ls); /* Skip second '['. */
if (lex_iseol(ls)) /* Skip initial newline. */
lex_newline(ls);
for (;;) {
switch (ls->c) {
case LEX_EOF:
lj_lex_error(ls, TK_eof, tv ? LJ_ERR_XLSTR : LJ_ERR_XLCOM);
break;
case ']':
if (lex_skipeq(ls) == sep) {
lex_savenext(ls); /* Skip second ']'. */
goto endloop;
}
break;
case '\n':
case '\r':
lex_save(ls, '\n');
lex_newline(ls);
if (!tv) lj_buf_reset(&ls->sb); /* Don't waste space for comments. */
break;
default:
lex_savenext(ls);
break;
}
} endloop:
if (tv) {
GCstr *str = lj_parse_keepstr(ls, sbufB(&ls->sb) + (2 + (MSize)sep),
sbuflen(&ls->sb) - 2*(2 + (MSize)sep));
setstrV(ls->L, tv, str);
}
}
/**
* Test that we correctly identify a CRLF sequence split between two buffers
*/
TEST_F(Lexer,LexNewLineCRLFSplit)
{
webvtt_uint pos = 0;
EXPECT_EQ( UNFINISHED, lex_newline( "\r", pos, false ) );
EXPECT_EQ( 1, pos );
EXPECT_EQ( L_NEWLINE0, lexerState() );
pos = 0;
EXPECT_EQ( NEWLINE, lex_newline( "\n", pos, true ) );
EXPECT_EQ( 1, pos );
EXPECT_EQ( L_START, lexerState() );
}
/**
* Test that we correctly identify a CR that is not followed by an LF, and
* backup to the correct position.
*/
TEST_F(Lexer,LexNewLineCRXX)
{
webvtt_uint pos = 0;
EXPECT_EQ( NEWLINE, lex_newline( "\rx", pos ) );
EXPECT_EQ( 1, pos );
EXPECT_EQ( L_START, lexerState() );
}
/**
* Test that we correctly return BADTOKEN and backup appropriately when we don't
* encounter a CR or LF
*/
TEST_F(Lexer,LexNewLineXX)
{
webvtt_uint pos = 0;
EXPECT_EQ( BADTOKEN, lex_newline( "xxx", pos ) );
EXPECT_EQ( 0, pos );
EXPECT_EQ( L_START, lexerState() );
}
/**
* Test that we return end-of-buffer as a newline.
*/
TEST_F(Lexer,LexNewLineEOL)
{
webvtt_uint pos = 0;
EXPECT_EQ( NEWLINE, lex_newline( "", pos ) );
EXPECT_EQ( 0, pos );
EXPECT_EQ( L_START, lexerState() );
}
/* Parse a string. */
static void lex_string(LexState *ls, TValue *tv)
{
LexChar delim = ls->c; /* Delimiter is '\'' or '"'. */
lex_savenext(ls);
while (ls->c != delim) {
switch (ls->c) {
case LEX_EOF:
lj_lex_error(ls, TK_eof, LJ_ERR_XSTR);
continue;
case '\n':
case '\r':
lj_lex_error(ls, TK_string, LJ_ERR_XSTR);
continue;
case '\\': {
LexChar c = lex_next(ls); /* Skip the '\\'. */
switch (c) {
case 'a': c = '\a'; break;
case 'b': c = '\b'; break;
case 'f': c = '\f'; break;
case 'n': c = '\n'; break;
case 'r': c = '\r'; break;
case 't': c = '\t'; break;
case 'v': c = '\v'; break;
case 'x': /* Hexadecimal escape '\xXX'. */
c = (lex_next(ls) & 15u) << 4;
if (!lj_char_isdigit(ls->c)) {
if (!lj_char_isxdigit(ls->c))
lj_lex_error(ls, TK_string, LJ_ERR_XHEX);
c += 9 << 4;
}
c += (lex_next(ls) & 15u);
if (!lj_char_isdigit(ls->c)) {
if (!lj_char_isxdigit(ls->c))
lj_lex_error(ls, TK_string, LJ_ERR_XHEX);
c += 9;
}
break;
case 'u': /* Unicode escape '\u{XX...}'. */
if (lex_next(ls) != '{') goto err_xesc;
lex_next(ls);
c = 0;
do {
c = (c << 4) | (ls->c & 15u);
if (!lj_char_isdigit(ls->c)) {
if (!lj_char_isxdigit(ls->c)) goto err_xesc;
c += 9;
}
if (c >= 0x110000) goto err_xesc; /* Out of Unicode range. */
} while (lex_next(ls) != '}');
if (c < 0x800) {
if (c < 0x80) break;
lex_save(ls, 0xc0 | (c >> 6));
} else {
if (c >= 0x10000) {
lex_save(ls, 0xf0 | (c >> 18));
lex_save(ls, 0x80 | ((c >> 12) & 0x3f));
} else {
if (c >= 0xd800 && c < 0xe000) goto err_xesc; /* No surrogates. */
lex_save(ls, 0xe0 | (c >> 12));
}
lex_save(ls, 0x80 | ((c >> 6) & 0x3f));
}
c = 0x80 | (c & 0x3f);
break;
case 'z': /* Skip whitespace. */
lex_next(ls);
while (lj_char_isspace(ls->c))
if (lex_iseol(ls)) lex_newline(ls); else lex_next(ls);
continue;
case '\n': case '\r': lex_save(ls, '\n'); lex_newline(ls); continue;
case '\\': case '\"': case '\'': break;
case LEX_EOF: continue;
default:
if (!lj_char_isdigit(c))
goto err_xesc;
c -= '0'; /* Decimal escape '\ddd'. */
if (lj_char_isdigit(lex_next(ls))) {
c = c*10 + (ls->c - '0');
if (lj_char_isdigit(lex_next(ls))) {
c = c*10 + (ls->c - '0');
if (c > 255) {
err_xesc:
lj_lex_error(ls, TK_string, LJ_ERR_XESC);
}
lex_next(ls);
}
}
lex_save(ls, c);
continue;
}
lex_save(ls, c);
lex_next(ls);
continue;
}