summaryrefslogtreecommitdiffstats
path: root/parser.l
diff options
context:
space:
mode:
Diffstat (limited to 'parser.l')
-rw-r--r--parser.l954
1 files changed, 490 insertions, 464 deletions
diff --git a/parser.l b/parser.l
index a53f8514..732e2ce0 100644
--- a/parser.l
+++ b/parser.l
@@ -150,10 +150,13 @@ static wchar_t num_esc(char *num)
SYM [a-zA-Z0-9_]+
NUM [+\-]?[0-9]+
-NSCHR [a-zA-Z0-9!$%&*+\-<=>?\\^_~]
+BSCHR [a-zA-Z0-9!$%&*+\-<=>?\\^_~]
+BSYM {BSCHR}({BSCHR}|#)*
+NSCHR [a-zA-Z0-9!$%&*+\-<=>?\\^_~/]
NSYM {NSCHR}({NSCHR}|#)*
TOK :?{SYM}
ATNUM @{NUM}
+BTOK [:@]?{BSYM}
NTOK [:@]?{NSYM}
ID_END [^a-zA-Z0-9_]
WS [\t ]*
@@ -171,497 +174,520 @@ UANY {ASC}|{U2}{U}|{U3}{U}{U}|{U4}{U}{U}{U}
UANYN {ASCN}|{U2}{U}|{U3}{U}{U}|{U4}{U}{U}{U}
UONLY {U2}{U}|{U3}{U}{U}|{U4}{U}{U}{U}
-%x SPECIAL NESTED REGEX STRLIT CHRLIT QSILIT
+%x SPECIAL BRACED NESTED REGEX STRLIT CHRLIT QSILIT
%%
-<SPECIAL,NESTED>{NUM} {
- val str = string_own(utf8_dup_from(yytext));
-
- if (yy_top_state() == INITIAL
- || yy_top_state() == QSILIT)
- yy_pop_state();
-
- yylval.num = int_str(str, num(10));
- return NUMBER;
- }
-
-<NESTED,QSILIT>{ATNUM} {
- val str = string_own(utf8_dup_from(yytext + 1));
-
- if (yy_top_state() == INITIAL
- || yy_top_state() == QSILIT)
- yy_pop_state();
- yylval.num = int_str(str, num(10));
- return METANUM;
- }
-
-<SPECIAL>{TOK} |
-<NESTED>{NTOK} {
- if (yy_top_state() == INITIAL
- || yy_top_state() == QSILIT)
- yy_pop_state();
-
- switch (yytext[0]) {
- case ':':
- yylval.lexeme = utf8_dup_from(yytext + 1);
- return KEYWORD;
- case '@':
- yylval.lexeme = utf8_dup_from(yytext + 1);
- return METAVAR;
- default:
- yylval.lexeme = utf8_dup_from(yytext);
- return IDENT;
- }
- }
-<NESTED>: {
- if (yy_top_state() == INITIAL
- || yy_top_state() == QSILIT)
- yy_pop_state();
- yylval.lexeme = utf8_dup_from("");
- return KEYWORD;
- }
-
-<SPECIAL>\({WS}all{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return ALL;
- }
-
-<SPECIAL>\({WS}some/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return SOME;
- }
-
-<SPECIAL>\({WS}none{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return NONE;
- }
-
-<SPECIAL>\({WS}maybe{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return MAYBE;
- }
-
-<SPECIAL>\({WS}cases{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return CASES;
- }
-
-<SPECIAL>\({WS}choose/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return CHOOSE;
- }
-
-<SPECIAL>\({WS}gather/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return GATHER;
- }
-
-<SPECIAL>\({WS}and{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return AND;
- }
-
-<SPECIAL>\({WS}or{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return OR;
- }
-
-<SPECIAL>\({WS}end{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return END;
- }
+<SPECIAL,NESTED,BRACED>{NUM} {
+ val str = string_own(utf8_dup_from(yytext));
+
+ if (yy_top_state() == INITIAL
+ || yy_top_state() == QSILIT)
+ yy_pop_state();
+
+ yylval.num = int_str(str, num(10));
+ return NUMBER;
+}
+
+<NESTED,QSILIT>{ATNUM} {
+ val str = string_own(utf8_dup_from(yytext + 1));
+
+ if (yy_top_state() == INITIAL
+ || yy_top_state() == QSILIT)
+ yy_pop_state();
+ yylval.num = int_str(str, num(10));
+ return METANUM;
+}
+
+<SPECIAL>{TOK} |
+<BRACED>{BTOK} |
+<NESTED>{NTOK} {
+ if (yy_top_state() == INITIAL
+ || yy_top_state() == QSILIT)
+ yy_pop_state();
+
+ switch (yytext[0]) {
+ case ':':
+ yylval.lexeme = utf8_dup_from(yytext + 1);
+ return KEYWORD;
+ case '@':
+ yylval.lexeme = utf8_dup_from(yytext + 1);
+ return METAVAR;
+ default:
+ yylval.lexeme = utf8_dup_from(yytext);
+ return IDENT;
+ }
+}
+
+<BRACED,NESTED>: {
+ if (yy_top_state() == INITIAL
+ || yy_top_state() == QSILIT)
+ yy_pop_state();
+ yylval.lexeme = utf8_dup_from("");
+ return KEYWORD;
+}
+
+<SPECIAL>\({WS}all{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return ALL;
+}
+
+<SPECIAL>\({WS}some/{ID_END} {
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return SOME;
+}
+
+<SPECIAL>\({WS}none{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return NONE;
+}
+
+<SPECIAL>\({WS}maybe{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return MAYBE;
+}
+
+<SPECIAL>\({WS}cases{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return CASES;
+}
+
+<SPECIAL>\({WS}choose/{ID_END} {
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return CHOOSE;
+}
+
+<SPECIAL>\({WS}gather/{ID_END} {
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return GATHER;
+}
+
+<SPECIAL>\({WS}and{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return AND;
+}
+
+<SPECIAL>\({WS}or{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return OR;
+}
+
+<SPECIAL>\({WS}end{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return END;
+}
<SPECIAL>\({WS}collect/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return COLLECT;
- }
-
-<SPECIAL>\({WS}coll/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return COLL;
- }
-
-<SPECIAL>\({WS}until{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return UNTIL;
- }
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return COLLECT;
+}
+
+<SPECIAL>\({WS}coll/{ID_END} {
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return COLL;
+}
+
+<SPECIAL>\({WS}until{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return UNTIL;
+}
<SPECIAL>\({WS}output/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return OUTPUT;
- }
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return OUTPUT;
+}
<SPECIAL>\({WS}repeat/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return REPEAT;
- }
-
-
-<SPECIAL>\({WS}rep/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return REP;
- }
-
-<SPECIAL>\({WS}single{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return SINGLE;
- }
-
-<SPECIAL>\({WS}first{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return FIRST;
- }
-
-<SPECIAL>\({WS}last{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return LAST;
- }
-
-<SPECIAL>\({WS}empty{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return EMPTY;
- }
-
-<SPECIAL>\({WS}mod/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return MOD;
- }
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return REPEAT;
+}
+
+
+<SPECIAL>\({WS}rep/{ID_END} {
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return REP;
+}
+
+<SPECIAL>\({WS}single{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return SINGLE;
+}
+
+<SPECIAL>\({WS}first{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return FIRST;
+}
+
+<SPECIAL>\({WS}last{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return LAST;
+}
+
+<SPECIAL>\({WS}empty{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return EMPTY;
+}
+
+<SPECIAL>\({WS}mod/{ID_END} {
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return MOD;
+}
<SPECIAL>\({WS}modlast/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return MODLAST;
- }
-
-<SPECIAL>\({WS}define/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return DEFINE;
- }
-
-<SPECIAL>\({WS}try{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return TRY;
- }
-
-<SPECIAL>\({WS}catch/{ID_END} {
- yy_push_state(NESTED);
- yylval.lineno = lineno;
- return CATCH;
- }
-
-<SPECIAL>\({WS}finally{WS}\) {
- yy_pop_state();
- yylval.lineno = lineno;
- return FINALLY;
- }
-
-<NESTED>@[\(\[] |
-<SPECIAL,NESTED>[{(\[] {
- yy_push_state(NESTED);
- if (yytext[0] == '@') {
- yylval.chr = yytext[1];
- return yytext[1] == '(' ? METAPAR : METABKT;
- }
- yylval.lineno = lineno;
- return yytext[0];
- }
-
-<SPECIAL,NESTED>,[*] {
- yylval.chr = '*';
- return SPLICE;
- }
-
-<SPECIAL,NESTED>[,'] {
- yylval.chr = yytext[0];
- return yytext[0];
- }
-
-<SPECIAL,NESTED>[})\]] {
- yy_pop_state();
- if (yy_top_state() == INITIAL
- || yy_top_state() == QSILIT)
- yy_pop_state();
- return yytext[0];
- }
-
-<SPECIAL,NESTED>{WS} { /* Eat whitespace in directive */ }
-
-<SPECIAL,NESTED>\" {
- yy_push_state(STRLIT);
- return '"';
- }
-
-<SPECIAL,NESTED>#\\ {
- yy_push_state(CHRLIT);
- return HASH_BACKSLASH;
- }
-
-<SPECIAL,NESTED>` {
- yy_push_state(QSILIT);
- return '`';
- }
-
-<NESTED># {
- return '#';
- }
-
-<NESTED>\.\. {
- yylval.lineno = lineno;
- return DOTDOT;
- }
-
-<SPECIAL>@ {
- yy_pop_state();
- yylval.lexeme = chk_strdup(L"@");
- return TEXT;
- }
-
-<SPECIAL,NESTED>\n {
- lineno++;
- }
-
-<SPECIAL,NESTED>[/] {
- yy_push_state(REGEX);
- return '/';
- }
-
-<SPECIAL,NESTED>\. {
- yylval.chr = '.';
- return '.';
- }
-
-<SPECIAL,NESTED>[\\]\n{WS} {
- yy_pop_state();
- lineno++;
- }
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return MODLAST;
+}
+
+<SPECIAL>\({WS}define/{ID_END} {
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return DEFINE;
+}
+
+<SPECIAL>\({WS}try{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return TRY;
+}
+
+<SPECIAL>\({WS}catch/{ID_END} {
+ yy_push_state(NESTED);
+ yylval.lineno = lineno;
+ return CATCH;
+}
+
+<SPECIAL>\({WS}finally{WS}\) {
+ yy_pop_state();
+ yylval.lineno = lineno;
+ return FINALLY;
+}
+
+<SPECIAL>[{] {
+ yy_push_state(BRACED);
+ yylval.lineno = lineno;
+ return yytext[0];
+}
+
+<SPECIAL>[(\[] |
+<NESTED,BRACED>@?[(\[] {
+ yy_push_state(NESTED);
+ if (yytext[0] == '@') {
+ yylval.chr = yytext[1];
+ return yytext[1] == '(' ? METAPAR : METABKT;
+ }
+ yylval.lineno = lineno;
+ return yytext[0];
+}
+
+<NESTED>,[*] {
+ yylval.chr = '*';
+ return SPLICE;
+}
+
+<NESTED>[,'] {
+ yylval.chr = yytext[0];
+ return yytext[0];
+}
+
+<BRACED>[}] {
+ yy_pop_state();
+ if (yy_top_state() == INITIAL
+ || yy_top_state() == QSILIT)
+ yy_pop_state();
+ return yytext[0];
+}
+
+<SPECIAL,NESTED>[)\]] {
+ yy_pop_state();
+ if (yy_top_state() == INITIAL
+ || yy_top_state() == QSILIT)
+ yy_pop_state();
+ return yytext[0];
+}
+
+<SPECIAL,NESTED,BRACED>{WS} {
+ /* Eat whitespace in directive */
+}
+
+<SPECIAL,NESTED,BRACED>\" {
+ yy_push_state(STRLIT);
+ return '"';
+}
+
+<SPECIAL,NESTED,BRACED>#\\ {
+ yy_push_state(CHRLIT);
+ return HASH_BACKSLASH;
+}
+
+<SPECIAL,NESTED,BRACED>#[/] {
+ yy_push_state(REGEX);
+ return HASH_SLASH;
+}
+
+<SPECIAL,NESTED,BRACED>` {
+ yy_push_state(QSILIT);
+ return '`';
+}
+
+<NESTED,BRACED># {
+ return '#';
+}
+
+<NESTED>\.\. {
+ yylval.lineno = lineno;
+ return DOTDOT;
+}
+
+<SPECIAL>@ {
+ yy_pop_state();
+ yylval.lexeme = chk_strdup(L"@");
+ return TEXT;
+}
+
+<SPECIAL,NESTED,BRACED>\n {
+ lineno++;
+}
+
+<SPECIAL,BRACED>[/] {
+ yy_push_state(REGEX);
+ return '/';
+}
+
+<SPECIAL,NESTED>\. {
+ yylval.chr = '.';
+ return '.';
+}
+
+<SPECIAL,NESTED,BRACED>[\\]\n{WS} {
+ yy_pop_state();
+ lineno++;
+}
<SPECIAL>[\\][abtnvfre ] {
- wchar_t lexeme[2];
- lexeme[0] = char_esc(yytext[1]);
- lexeme[1] = 0;
- yylval.lexeme = chk_strdup(lexeme);
- yy_pop_state();
- return TEXT;
- }
-
-<SPECIAL>[\\](x{HEX}+|{OCT}+) {
- wchar_t lexeme[2];
- lexeme[0] = num_esc(yytext + 1);
- lexeme[1] = 0;
- yylval.lexeme = chk_strdup(lexeme);
- yy_pop_state();
- return TEXT;
- }
-
-<SPECIAL,NESTED>[;].* {
- /* comment */
- }
-<SPECIAL,NESTED>{UANYN} {
- yyerrprepf(lit("bad character in directive: '~a'"),
- string_utf8(yytext), nao);
- return ERRTOK;
- }
-
-<SPECIAL,NESTED>. {
- yyerrprepf(lit("non-UTF-8 byte in directive: "
- "'\\x~02x'"),
- num((unsigned char) yytext[0]), nao);
- return ERRTOK;
- }
-
-<REGEX>[/] {
- yylval.chr = '/';
- return '/';
- }
-
-
-<REGEX>[\\][abtnvfre\\ ] {
- yylval.chr = char_esc(yytext[1]);
- return REGCHAR;
- }
-
-<REGEX>[\\](x{HEX}+|{OCT}+);? {
- yylval.chr = num_esc(yytext + 1);
- return REGCHAR;
- }
-
-<REGEX>{WS}[\\]\n{WS} {
- lineno++;
- }
-
-<REGEX>\n {
- lineno++;
- yyerrprepf(lit("newline in regex"), nao);
- return ERRTOK;
- }
-
-<REGEX>[.*?+~&%] {
- yylval.chr = yytext[0];
- return yytext[0];
- }
+ wchar_t lexeme[2];
+ lexeme[0] = char_esc(yytext[1]);
+ lexeme[1] = 0;
+ yylval.lexeme = chk_strdup(lexeme);
+ yy_pop_state();
+ return TEXT;
+}
+
+<SPECIAL>[\\](x{HEX}+|{OCT}+) {
+ wchar_t lexeme[2];
+ lexeme[0] = num_esc(yytext + 1);
+ lexeme[1] = 0;
+ yylval.lexeme = chk_strdup(lexeme);
+ yy_pop_state();
+ return TEXT;
+}
+
+<SPECIAL,NESTED,BRACED>[;].* {
+ /* comment */
+}
+<SPECIAL,NESTED,BRACED>{UANYN} {
+ yyerrprepf(lit("bad character in directive: '~a'"),
+ string_utf8(yytext), nao);
+ return ERRTOK;
+}
+
+<SPECIAL,NESTED,BRACED>. {
+ yyerrprepf(lit("non-UTF-8 byte in directive: "
+ "'\\x~02x'"),
+ num((unsigned char) yytext[0]), nao);
+ return ERRTOK;
+}
+
+<REGEX>[/] {
+ yylval.chr = '/';
+ return '/';
+}
+
+
+<REGEX>[\\][abtnvfre\\ ] {
+ yylval.chr = char_esc(yytext[1]);
+ return REGCHAR;
+}
+
+<REGEX>[\\](x{HEX}+|{OCT}+);? {
+ yylval.chr = num_esc(yytext + 1);
+ return REGCHAR;
+}
+
+<REGEX>{WS}[\\]\n{WS} {
+ lineno++;
+}
+
+<REGEX>\n {
+ lineno++;
+ yyerrprepf(lit("newline in regex"), nao);
+ return ERRTOK;
+}
+
+<REGEX>[.*?+~&%] {
+ yylval.chr = yytext[0];
+ return yytext[0];
+}
<REGEX>[\[\]\-] {
- yylval.chr = yytext[0];
- return yytext[0];
- }
+ yylval.chr = yytext[0];
+ return yytext[0];
+}
-<REGEX>[()|] {
- yylval.chr = yytext[0];
- return yytext[0];
- }
+<REGEX>[()|] {
+ yylval.chr = yytext[0];
+ return yytext[0];
+}
-<REGEX>[\\]. {
- yylval.chr = yytext[1];
- return REGCHAR;
- }
+<REGEX>[\\]. {
+ yylval.chr = yytext[1];
+ return REGCHAR;
+}
<REGEX>{UANYN} {
- wchar_t buf[8];
- utf8_from(buf, yytext);
- yylval.chr = buf[0];
- return REGCHAR;
- }
-
-<REGEX>. {
- yyerrprepf(lit("non-UTF-8 byte in regex: '\\x~02x'"),
- num((unsigned char) yytext[0]), nao);
- return ERRTOK;
- }
-
-<INITIAL>[ ]+ {
- yylval.lexeme = utf8_dup_from(yytext);
- return SPACE;
- }
-
-<INITIAL>({UONLY}|[^@\n ])+ {
- yylval.lexeme = utf8_dup_from(yytext);
- return TEXT;
- }
-
-<INITIAL>\n {
- lineno++;
- return '\n';
- }
-
-<INITIAL>@{WS}\* {
- yy_push_state(SPECIAL);
- return '*';
- }
-
-<INITIAL>@ {
- yy_push_state(SPECIAL);
- }
-
-<INITIAL>^@[#;].*\n {
- /* eat whole line comment */
- lineno++;
- }
-
-<INITIAL>@[#;].* {
- /* comment to end of line */
- }
-
-<STRLIT>\" {
- yy_pop_state();
- return yytext[0];
- }
-
-<QSILIT>` {
- yy_pop_state();
- return yytext[0];
- }
-
-<STRLIT,QSILIT>[\\][abtnvfre"`'\\] {
- yylval.chr = char_esc(yytext[1]);
- return LITCHAR;
- }
-
-<STRLIT,QSILIT>{WS}[\\]\n{WS} {
- lineno++;
- }
+ wchar_t buf[8];
+ utf8_from(buf, yytext);
+ yylval.chr = buf[0];
+ return REGCHAR;
+}
+
+<REGEX>. {
+ yyerrprepf(lit("non-UTF-8 byte in regex: '\\x~02x'"),
+ num((unsigned char) yytext[0]), nao);
+ return ERRTOK;
+}
+
+<INITIAL>[ ]+ {
+ yylval.lexeme = utf8_dup_from(yytext);
+ return SPACE;
+}
+
+<INITIAL>({UONLY}|[^@\n ])+ {
+ yylval.lexeme = utf8_dup_from(yytext);
+ return TEXT;
+}
+
+<INITIAL>\n {
+ lineno++;
+ return '\n';
+}
+
+<INITIAL>@{WS}\* {
+ yy_push_state(SPECIAL);
+ return '*';
+}
+
+<INITIAL>@ {
+ yy_push_state(SPECIAL);
+}
+
+<INITIAL>^@[#;].*\n {
+ /* eat whole line comment */
+ lineno++;
+}
+
+<INITIAL>@[#;].* {
+ /* comment to end of line */
+}
+
+<STRLIT>\" {
+ yy_pop_state();
+ return yytext[0];
+}
+
+<QSILIT>` {
+ yy_pop_state();
+ return yytext[0];
+}
+
+<STRLIT,QSILIT>[\\][abtnvfre"`'\\] {
+ yylval.chr = char_esc(yytext[1]);
+ return LITCHAR;
+}
+
+<STRLIT,QSILIT>{WS}[\\]\n{WS} {
+ lineno++;
+}
<STRLIT,QSILIT>[\\](x{HEX}+|{OCT}+);? {
- yylval.chr = num_esc(yytext+1);
- return LITCHAR;
- }
+ yylval.chr = num_esc(yytext+1);
+ return LITCHAR;
+}
<CHRLIT>(x{HEX}+|o{OCT}+) {
- yylval.chr = num_esc(yytext);
- return LITCHAR;
- }
-
-<CHRLIT>{SYM} {
- yylval.lexeme = utf8_dup_from(yytext);
- return IDENT;
- }
-
-<CHRLIT>[^ \t\n] {
- yylval.lexeme = utf8_dup_from(yytext);
- return IDENT; /* hack */
- }
-
-<STRLIT>\n {
- yyerrprepf(lit("newline in string literal"), nao);
- lineno++;
- yylval.chr = yytext[0];
- return ERRTOK;
- }
-
-<CHRLIT>\n {
- yyerrprepf(lit("newline in character literal"), nao);
- lineno++;
- yylval.chr = yytext[0];
- return ERRTOK;
- }
-
-<QSILIT>\n {
- yyerrprepf(lit("newline in string quasiliteral"), nao);
- lineno++;
- yylval.chr = yytext[0];
- return ERRTOK;
- }
-
-<QSILIT>@ {
- yy_push_state(SPECIAL);
- }
+ yylval.chr = num_esc(yytext);
+ return LITCHAR;
+}
+
+<CHRLIT>{SYM} {
+ yylval.lexeme = utf8_dup_from(yytext);
+ return IDENT;
+}
+
+<CHRLIT>[^ \t\n] {
+ yylval.lexeme = utf8_dup_from(yytext);
+ return IDENT; /* hack */
+}
+
+<STRLIT>\n {
+ yyerrprepf(lit("newline in string literal"), nao);
+ lineno++;
+ yylval.chr = yytext[0];
+ return ERRTOK;
+}
+
+<CHRLIT>\n {
+ yyerrprepf(lit("newline in character literal"), nao);
+ lineno++;
+ yylval.chr = yytext[0];
+ return ERRTOK;
+}
+
+<QSILIT>\n {
+ yyerrprepf(lit("newline in string quasiliteral"), nao);
+ lineno++;
+ yylval.chr = yytext[0];
+ return ERRTOK;
+}
+
+<QSILIT>@ {
+ yy_push_state(SPECIAL);
+}
<STRLIT,CHRLIT,QSILIT>{UANYN} {
- wchar_t buf[8];
- utf8_from(buf, yytext);
- yylval.chr = buf[0];
- return LITCHAR;
- }
+ wchar_t buf[8];
+ utf8_from(buf, yytext);
+ yylval.chr = buf[0];
+ return LITCHAR;
+}
<STRLIT,CHRLIT,QSILIT>. {
- yyerrprepf(lit("non-UTF-8 byte in literal: '\\x~02x'"),
- num((unsigned char) yytext[0]), nao);
- return ERRTOK;
- }
+ yyerrprepf(lit("non-UTF-8 byte in literal: '\\x~02x'"),
+ num((unsigned char) yytext[0]), nao);
+ return ERRTOK;
+}
%%
generated by cgit v1.2.3 (git 2.25.1) at 2025-03-15 20:10:54 +0000