From: Theophile Ranquet Date: Thu, 15 Nov 2012 17:10:35 +0000 (+0000) Subject: scanner: reintroduce unput for missing end tokens X-Git-Tag: v2.7.90~208 X-Git-Url: https://git.saurik.com/bison.git/commitdiff_plain/c1b2677ad0a0a507010fa7dde5ff07dbe4596a10?ds=inline scanner: reintroduce unput for missing end tokens Unput was no longer used since a POSIX-compatiblity issue with Flex 2.5.31, which has been adressed in newer versions of Flex. See this discussion: This partially reverts commit aa4180418fff518198e1b0f2c43fec6432210dc7. * src/scan-gram.l (unexpected_end): Here. * tests/input.at: Adjust for new order of error reports. --- diff --git a/src/scan-gram.l b/src/scan-gram.l index 4b7591bb..0ea06c3f 100644 --- a/src/scan-gram.l +++ b/src/scan-gram.l @@ -517,23 +517,15 @@ eqopt ([[:space:]]*=)? { - "\""|"\n" { - if (yytext[0] == '\n') - unexpected_newline (token_start, "\""); - STRING_FINISH; - loc->start = token_start; - val->chars = last_string; - BEGIN INITIAL; - return STRING; - } - <> { - unexpected_eof (token_start, "\""); + "\"" { STRING_FINISH; loc->start = token_start; val->chars = last_string; BEGIN INITIAL; return STRING; } + <> unexpected_eof (token_start, "\""); + "\n" unexpected_newline (token_start, "\""); } /*----------------------------------------------------------. @@ -543,49 +535,27 @@ eqopt ([[:space:]]*=)? { - "'"|"\n" { - STRING_FINISH; - loc->start = token_start; - val->character = last_string[0]; - { - /* FIXME: Eventually, make these errors. */ - if (last_string[0] == '\0') - { - complain (loc, Wother, _("empty character literal")); - /* '\0' seems dangerous even if we are about to complain. */ - val->character = '\''; - } - else if (last_string[1] != '\0') - complain (loc, Wother, - _("extra characters in character literal")); - } - if (yytext[0] == '\n') - unexpected_newline (token_start, "'"); - STRING_FREE; - BEGIN INITIAL; - return CHAR; - } - <> { + "'" { STRING_FINISH; loc->start = token_start; val->character = last_string[0]; + + /* FIXME: Eventually, make these errors. */ + if (last_string[0] == '\0') { - /* FIXME: Eventually, make these errors. */ - if (last_string[0] == '\0') - { - complain (loc, Wother, _("empty character literal")); - /* '\0' seems dangerous even if we are about to complain. */ - val->character = '\''; - } - else if (last_string[1] != '\0') - complain (loc, Wother, - _("extra characters in character literal")); + complain (loc, Wother, _("empty character literal")); + /* '\0' seems dangerous even if we are about to complain. */ + val->character = '\''; } - unexpected_eof (token_start, "'"); + else if (last_string[1] != '\0') + complain (loc, Wother, + _("extra characters in character literal")); STRING_FREE; BEGIN INITIAL; return CHAR; } + "\n" unexpected_newline (token_start, "'"); + <> unexpected_eof (token_start, "'"); } /*-----------------------------------------------------------. @@ -612,15 +582,7 @@ eqopt ([[:space:]]*=)? [^<>]+ STRING_GROW; "<"+ STRING_GROW; nesting += yyleng; - <> { - unexpected_eof (token_start, ">"); - STRING_FINISH; - loc->start = token_start; - val->uniqstr = uniqstr_new (last_string); - STRING_FREE; - BEGIN INITIAL; - return TAG; - } + <> unexpected_eof (token_start, ">"); } /*----------------------------. @@ -691,15 +653,15 @@ eqopt ([[:space:]]*=)? { "'" STRING_GROW; BEGIN context_state; - \n unexpected_newline (token_start, "'"); BEGIN context_state; - <> unexpected_eof (token_start, "'"); BEGIN context_state; + \n unexpected_newline (token_start, "'"); + <> unexpected_eof (token_start, "'"); } { "\"" STRING_GROW; BEGIN context_state; - \n unexpected_newline (token_start, "\""); BEGIN context_state; - <> unexpected_eof (token_start, "\""); BEGIN context_state; + \n unexpected_newline (token_start, "\""); + <> unexpected_eof (token_start, "\""); } @@ -750,15 +712,7 @@ eqopt ([[:space:]]*=)? (as '<' '<%'). */ "<"{splice}"<" STRING_GROW; - <> { - int token = (YY_START == SC_BRACED_CODE) ? BRACED_CODE : BRACED_PREDICATE; - unexpected_eof (code_start, "}"); - STRING_FINISH; - loc->start = code_start; - val->code = last_string; - BEGIN INITIAL; - return token; - } + <> unexpected_eof (code_start, "}"); } @@ -809,14 +763,7 @@ eqopt ([[:space:]]*=)? return PROLOGUE; } - <> { - unexpected_eof (code_start, "%}"); - STRING_FINISH; - loc->start = code_start; - val->chars = last_string; - BEGIN INITIAL; - return PROLOGUE; - } + <> unexpected_eof (code_start, "%}"); } @@ -999,8 +946,8 @@ handle_syncline (char *args, location loc) /*----------------------------------------------------------------. | For a token or comment starting at START, report message MSGID, | -| which should say that an end marker was found before | -| the expected TOKEN_END. | +| which should say that an end marker was found before the | +| expected TOKEN_END. Then, pretend that TOKEN_END was found. | `----------------------------------------------------------------*/ static void @@ -1009,6 +956,15 @@ unexpected_end (boundary start, char const *msgid, char const *token_end) location loc; loc.start = start; loc.end = scanner_cursor; + size_t i = strlen (token_end); + +/* Adjust scanner cursor so that any later message does not count + the characters about to be inserted. */ + scanner_cursor.column -= i; + + while (i != 0) + unput (token_end[--i]); + token_end = quote (token_end); /* Instead of '\'', display "'". */ if (STREQ (token_end, "'\\''")) @@ -1020,6 +976,7 @@ unexpected_end (boundary start, char const *msgid, char const *token_end) /*------------------------------------------------------------------------. | Report an unexpected EOF in a token or comment starting at START. | | An end of file was encountered and the expected TOKEN_END was missing. | +| After reporting the problem, pretend that TOKEN_END was found. | `------------------------------------------------------------------------*/ static void diff --git a/tests/input.at b/tests/input.at index f60eaec7..385d0f3a 100644 --- a/tests/input.at +++ b/tests/input.at @@ -1469,12 +1469,22 @@ start: ' ]]) AT_CHECK([[$PERL -e "print 'start: \'';" >> empty.y || exit 77]]) -AT_BISON_CHECK([empty.y], [1], [], +AT_BISON_CHECK([-fcaret empty.y], [1], [], [[empty.y:2.8-9: warning: empty character literal [-Wother] -empty.y:3.8-4.0: warning: empty character literal [-Wother] + start: ''; + ^^ empty.y:3.8-4.0: error: missing "'" at end of line -empty.y:4.8: warning: empty character literal [-Wother] + start: ' + ^ +empty.y:3.8-4.0: warning: empty character literal [-Wother] + start: ' + ^ empty.y:4.8: error: missing "'" at end of file + start: ' + ^ +empty.y:4.8: warning: empty character literal [-Wother] + start: ' + ^ ]]) AT_DATA([two.y], @@ -1486,10 +1496,10 @@ AT_CHECK([[$PERL -e "print 'start: \'ab';" >> two.y || exit 77]]) AT_BISON_CHECK([two.y], [1], [], [[two.y:2.8-11: warning: extra characters in character literal [-Wother] -two.y:3.8-4.0: warning: extra characters in character literal [-Wother] two.y:3.8-4.0: error: missing "'" at end of line -two.y:4.8-10: warning: extra characters in character literal [-Wother] +two.y:3.8-4.0: warning: extra characters in character literal [-Wother] two.y:4.8-10: error: missing "'" at end of file +two.y:4.8-10: warning: extra characters in character literal [-Wother] ]]) AT_DATA([three.y], @@ -1501,10 +1511,10 @@ AT_CHECK([[$PERL -e "print 'start: \'abc';" >> three.y || exit 77]]) AT_BISON_CHECK([three.y], [1], [], [[three.y:2.8-12: warning: extra characters in character literal [-Wother] -three.y:3.8-4.0: warning: extra characters in character literal [-Wother] three.y:3.8-4.0: error: missing "'" at end of line -three.y:4.8-11: warning: extra characters in character literal [-Wother] +three.y:3.8-4.0: warning: extra characters in character literal [-Wother] three.y:4.8-11: error: missing "'" at end of file +three.y:4.8-11: warning: extra characters in character literal [-Wother] ]]) AT_CLEANUP