/* toke.c
*
- * Copyright (c) 1991-2000, Larry Wall
+ * Copyright (c) 1991-2001, Larry Wall
*
* You may distribute under the terms of either the GNU General Public
* License or the Artistic License, as specified in the README file.
# endif
YYSTYPE* yylval_pointer[YYMAXLEVEL];
int* yychar_pointer[YYMAXLEVEL];
-int yyactlevel = 0;
+int yyactlevel = -1;
# undef yylval
# undef yychar
# define yylval (*yylval_pointer[yyactlevel])
S_sublex_done(pTHX)
{
if (!PL_lex_starts++) {
+ SV *sv = newSVpvn("",0);
+ if (SvUTF8(PL_linestr))
+ SvUTF8_on(sv);
PL_expect = XOPERATOR;
- yylval.opval = (OP*)newSVOP(OP_CONST, 0, newSVpvn("",0));
+ yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
return THING;
}
register char *d = SvPVX(sv); /* destination for copies */
bool dorange = FALSE; /* are we in a translit range? */
bool didrange = FALSE; /* did we just finish a range? */
- bool has_utf8 = FALSE; /* embedded \x{} */
+ bool has_utf8 = (PL_linestr && SvUTF8(PL_linestr));
+ /* the constant is UTF8 */
UV uv;
I32 utf = (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op)
break; /* in regexp, $ might be tail anchor */
}
- /* (now in tr/// code again) */
-
- if (*s & 0x80 && (this_utf8 || has_utf8)) {
- STRLEN len = (STRLEN) -1;
- UV uv;
- if (this_utf8) {
- uv = utf8_to_uv((U8*)s, send - s, &len, UTF8_CHECK_ONLY);
- }
- if (len == (STRLEN)-1) {
- /* Illegal UTF8 (a high-bit byte), make it valid. */
- char *old_pvx = SvPVX(sv);
- /* need space for one extra char (NOTE: SvCUR() not set here) */
- d = SvGROW(sv, SvLEN(sv) + 1) + (d - old_pvx);
- d = (char*)uv_to_utf8((U8*)d, (U8)*s++);
- }
- else {
- while (len--)
- *d++ = *s++;
- }
- has_utf8 = TRUE;
- continue;
- }
-
/* backslashes */
if (*s == '\\' && s+1 < send) {
- bool to_be_utf8 = FALSE;
-
s++;
/* some backslashes we leave behind */
else {
STRLEN len = 1; /* allow underscores */
uv = (UV)scan_hex(s + 1, e - s - 1, &len);
- to_be_utf8 = TRUE;
}
s = e + 1;
}
NUM_ESCAPE_INSERT:
/* Insert oct or hex escaped character.
- * There will always enough room in sv since such escapes will
- * be longer than any utf8 sequence they can end up as
- */
+ * There will always enough room in sv since such
+ * escapes will be longer than any UT-F8 sequence
+ * they can end up as. */
+
+ /* This spot is wrong for EBCDIC. Characters like
+ * the lowercase letters and digits are >127 in EBCDIC,
+ * so here they would need to be mapped to the Unicode
+ * repertoire. --jhi */
+
if (uv > 127) {
- if (!has_utf8 && (to_be_utf8 || uv > 255)) {
- /* might need to recode whatever we have accumulated so far
- * if it contains any hibit chars
+ if (!has_utf8 && uv > 255) {
+ /* Might need to recode whatever we have
+ * accumulated so far if it contains any
+ * hibit chars.
+ *
+ * (Can't we keep track of that and avoid
+ * this rescan? --jhi)
*/
int hicount = 0;
char *c;
+
for (c = SvPVX(sv); c < d; c++) {
- if (*c & 0x80)
+ if (UTF8_IS_CONTINUED(*c))
hicount++;
}
if (hicount) {
char *old_pvx = SvPVX(sv);
char *src, *dst;
- d = SvGROW(sv, SvCUR(sv) + hicount + 1) + (d - old_pvx);
+
+ d = SvGROW(sv,
+ SvCUR(sv) + hicount + 1) +
+ (d - old_pvx);
src = d - 1;
d += hicount;
dst = d - 1;
while (src < dst) {
- if (*src & 0x80) {
- dst--;
- uv_to_utf8((U8*)dst, (U8)*src--);
- dst--;
+ if (UTF8_IS_CONTINUED(*src)) {
+ *dst-- = UTF8_EIGHT_BIT_LO(*src);
+ *dst-- = UTF8_EIGHT_BIT_HI(*src--);
}
else {
*dst-- = *src--;
}
}
- if (to_be_utf8 || uv > 255) {
+ if (has_utf8 || uv > 255) {
d = (char*)uv_to_utf8((U8*)d, uv);
has_utf8 = TRUE;
}
continue;
} /* end if (backslash) */
+ /* (now in tr/// code again) */
+
+ if (UTF8_IS_CONTINUED(*s) && (this_utf8 || has_utf8)) {
+ STRLEN len = (STRLEN) -1;
+ UV uv;
+ if (this_utf8) {
+ uv = utf8_to_uv((U8*)s, send - s, &len, 0);
+ }
+ if (len == (STRLEN)-1) {
+ /* Illegal UTF8 (a high-bit byte), make it valid. */
+ char *old_pvx = SvPVX(sv);
+ /* need space for one extra char (NOTE: SvCUR() not set here) */
+ d = SvGROW(sv, SvLEN(sv) + 1) + (d - old_pvx);
+ d = (char*)uv_to_utf8((U8*)d, (U8)*s++);
+ }
+ else {
+ while (len--)
+ *d++ = *s++;
+ }
+ has_utf8 = TRUE;
+ continue;
+ }
+
*d++ = *s++;
} /* while loop to process each character */
{
int r;
+ yyactlevel++;
yylval_pointer[yyactlevel] = lvalp;
yychar_pointer[yyactlevel] = lcharp;
- yyactlevel++;
if (yyactlevel >= YYMAXLEVEL)
Perl_croak(aTHX_ "panic: YYMAXLEVEL");
r = Perl_yylex(aTHX);
- yyactlevel--;
+ if (yyactlevel > 0)
+ yyactlevel--;
return r;
}
#ifdef __SC__
#pragma segment Perl_yylex
#endif
-
int
-#ifdef USE_PURE_BISON
-Perl_yylex(pTHX_ YYSTYPE *lvalp, int *lcharp)
-#else
Perl_yylex(pTHX)
-#endif
{
register char *s;
register char *d;
STRLEN len;
GV *gv = Nullgv;
GV **gvp = 0;
+ bool bof = FALSE;
/* check if there's an identifier for us to look at */
if (PL_pending_ident) {
goto retry;
}
do {
- bool bof = PL_rsfp ? TRUE : FALSE;
+ bof = PL_rsfp ? TRUE : FALSE;
if (bof) {
#ifdef PERLIO_IS_STDIO
# ifdef __GNU_LIBRARY__
if (ftst) {
PL_last_lop_op = ftst;
DEBUG_T( { PerlIO_printf(Perl_debug_log,
- "### Saw file test %c\n", ftst);
+ "### Saw file test %c\n", (int)ftst);
} )
FTST(ftst);
}
/* Assume it was a minus followed by a one-letter named
* subroutine call (or a -bareword), then. */
DEBUG_T( { PerlIO_printf(Perl_debug_log,
- "### %c looked like a file test but was not\n", ftst);
+ "### %c looked like a file test but was not\n",
+ (int)ftst);
} )
s -= 2;
}
PL_lex_stuff = Nullsv;
}
else {
- attrs = append_elem(OP_LIST, attrs,
- newSVOP(OP_CONST, 0,
- newSVpvn(s, len)));
+ if (!PL_in_my && len == 6 && strnEQ(s, "lvalue", len))
+ CvLVALUE_on(PL_compcv);
+ else if (!PL_in_my && len == 6 && strnEQ(s, "locked", len))
+ CvLOCKED_on(PL_compcv);
+ else if (!PL_in_my && len == 6 && strnEQ(s, "method", len))
+ CvMETHOD_on(PL_compcv);
+ /* After we've set the flags, it could be argued that
+ we don't need to do the attributes.pm-based setting
+ process, and shouldn't bother appending recognized
+ flags. To experiment with that, uncomment the
+ following "else": */
+ /* else */
+ attrs = append_elem(OP_LIST, attrs,
+ newSVOP(OP_CONST, 0,
+ newSVpvn(s, len)));
}
s = skipspace(d);
if (*s == ':' && s[1] != ':')
missingterm((char*)0);
yylval.ival = OP_CONST;
for (d = SvPV(PL_lex_stuff, len); len; len--, d++) {
- if (*d == '$' || *d == '@' || *d == '\\' || *d & 0x80) {
+ if (*d == '$' || *d == '@' || *d == '\\' || UTF8_IS_CONTINUED(*d)) {
yylval.ival = OP_STRINGIFY;
break;
}
TOKEN('(');
case KEY_qq:
+ case KEY_qu:
s = scan_str(s,FALSE,FALSE);
+ if (tmp == KEY_qu && is_utf8_string((U8*)s, SvCUR(PL_lex_stuff)))
+ SvUTF8_on(PL_lex_stuff);
if (!s)
missingterm((char*)0);
yylval.ival = OP_STRINGIFY;
if (strEQ(d,"q")) return KEY_q;
if (strEQ(d,"qr")) return KEY_qr;
if (strEQ(d,"qq")) return KEY_qq;
+ if (strEQ(d,"qu")) return KEY_qu;
if (strEQ(d,"qw")) return KEY_qw;
if (strEQ(d,"qx")) return KEY_qx;
}
*d++ = *s++;
*d++ = *s++;
}
- else if (UTF && *(U8*)s >= 0xc0 && isALNUM_utf8((U8*)s)) {
+ else if (UTF && UTF8_IS_START(*s) && isALNUM_utf8((U8*)s)) {
char *t = s + UTF8SKIP(s);
- while (*t & 0x80 && is_utf8_mark((U8*)t))
+ while (UTF8_IS_CONTINUED(*t) && is_utf8_mark((U8*)t))
t += UTF8SKIP(t);
if (d + (t - s) > e)
Perl_croak(aTHX_ ident_too_long);
*d++ = *s++;
*d++ = *s++;
}
- else if (UTF && *(U8*)s >= 0xc0 && isALNUM_utf8((U8*)s)) {
+ else if (UTF && UTF8_IS_START(*s) && isALNUM_utf8((U8*)s)) {
char *t = s + UTF8SKIP(s);
- while (*t & 0x80 && is_utf8_mark((U8*)t))
+ while (UTF8_IS_CONTINUED(*t) && is_utf8_mark((U8*)t))
t += UTF8SKIP(t);
if (d + (t - s) > e)
Perl_croak(aTHX_ ident_too_long);
e = s;
while ((e < send && isALNUM_lazy_if(e,UTF)) || *e == ':') {
e += UTF8SKIP(e);
- while (e < send && *e & 0x80 && is_utf8_mark((U8*)e))
+ while (e < send && UTF8_IS_CONTINUED(*e) && is_utf8_mark((U8*)e))
e += UTF8SKIP(e);
}
Copy(s, d, e - s, char);
/* after skipping whitespace, the next character is the terminator */
term = *s;
- if ((term & 0x80) && UTF)
+ if (UTF8_IS_CONTINUED(term) && UTF)
has_utf8 = TRUE;
/* mark where we are */
have found the terminator */
else if (*s == term)
break;
- else if (!has_utf8 && (*s & 0x80) && UTF)
+ else if (!has_utf8 && UTF8_IS_CONTINUED(*s) && UTF)
has_utf8 = TRUE;
*to = *s;
}
break;
else if (*s == PL_multi_open)
brackets++;
- else if (!has_utf8 && (*s & 0x80) && UTF)
+ else if (!has_utf8 && UTF8_IS_CONTINUED(*s) && UTF)
has_utf8 = TRUE;
*to = *s;
}
while (isDIGIT(*pos) || *pos == '_')
pos++;
if (!isALPHA(*pos)) {
- UV rev;
+ UV rev, revmax = 0;
U8 tmpbuf[UTF8_MAXLEN+1];
U8 *tmpend;
- bool utf8 = FALSE;
s++; /* get past 'v' */
sv = NEWSV(92,5);
}
}
tmpend = uv_to_utf8(tmpbuf, rev);
- utf8 = utf8 || rev > 127;
+ if (rev > revmax)
+ revmax = rev;
sv_catpvn(sv, (const char*)tmpbuf, tmpend - tmpbuf);
if (*pos == '.' && isDIGIT(pos[1]))
s = ++pos;
SvPOK_on(sv);
SvREADONLY_on(sv);
- if (utf8) {
+ if (revmax > 127) {
SvUTF8_on(sv);
- if (!UTF||IN_BYTE)
+ if (revmax < 256)
sv_utf8_downgrade(sv, TRUE);
}
}
projects / p5sagit/p5-mst-13.2.git / blobdiff