X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/ac9b0e954b1d3aed514a3bbd363da1514202af0f..026816664ff8283a55f91915843a8ff0ac5cf86c:/src/scan-gram.l diff --git a/src/scan-gram.l b/src/scan-gram.l index 0c12cbaa..555e6952 100644 --- a/src/scan-gram.l +++ b/src/scan-gram.l @@ -1,7 +1,6 @@ /* Bison Grammar Scanner -*- C -*- - Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 Free - Software Foundation, Inc. + Copyright (C) 2002-2013 Free Software Foundation, Inc. This file is part of Bison, the GNU Compiler Compiler. @@ -18,7 +17,7 @@ You should have received a copy of the GNU General Public License along with this program. If not, see . */ -%option debug nodefault nounput noyywrap never-interactive +%option debug nodefault noinput noyywrap never-interactive %option prefix="gram_" outfile="lex.yy.c" %{ @@ -28,24 +27,26 @@ #define gram_wrap() 1 #define FLEX_PREFIX(Id) gram_ ## Id -#include "flex-scanner.h" +#include -#include "complain.h" -#include "files.h" -#include "gram.h" -#include "quotearg.h" -#include "reader.h" -#include "uniqstr.h" +#include +#include +#include +#include +#include +#include +#include +#include #include #include -#include "scan-gram.h" +#include #define YY_DECL GRAM_LEX_DECL -#define YY_USER_INIT \ - code_start = scanner_cursor = loc->start; \ +#define YY_USER_INIT \ + code_start = scanner_cursor = loc->start; \ /* Location of scanner cursor. */ static boundary scanner_cursor; @@ -55,12 +56,33 @@ static boundary scanner_cursor; static size_t no_cr_read (FILE *, char *, size_t); #define YY_INPUT(buf, result, size) ((result) = no_cr_read (yyin, buf, size)) +#define RETURN_PERCENT_PARAM(Value) \ + RETURN_VALUE(PERCENT_PARAM, param, param_ ## Value) + +#define RETURN_PERCENT_FLAG(Value) \ + RETURN_VALUE(PERCENT_FLAG, uniqstr, uniqstr_new (Value)) + +#define RETURN_VALUE(Token, Field, Value) \ + do { \ + val->Field = Value; \ + return Token; \ + } while (0) + #define ROLLBACK_CURRENT_TOKEN \ do { \ - scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0); \ + scanner_cursor.column -= mbsnwidth (yytext, yyleng, 0); \ yyless (0); \ } while (0) +#define DEPRECATED(Msg) \ + do { \ + size_t i; \ + deprecated_directive (loc, yytext, Msg); \ + scanner_cursor.column -= mbsnwidth (Msg, strlen (Msg), 0); \ + for (i = strlen (Msg); i != 0; --i) \ + unput (Msg[i - 1]); \ + } while (0) + /* A string representing the most recently saved token. */ static char *last_string; @@ -88,14 +110,22 @@ static void unexpected_newline (boundary, char const *); /* Strings and characters in directives/rules. */ %x SC_ESCAPED_STRING SC_ESCAPED_CHARACTER /* A identifier was just read in directives/rules. Special state - to capture the sequence `identifier :'. */ + to capture the sequence 'identifier :'. */ %x SC_AFTER_IDENTIFIER - /* Three types of user code: - - prologue (code between `%{' `%}' in the first section, before %%); + /* POSIX says that a tag must be both an id and a C union member, but + historically almost any character is allowed in a tag. We + disallow NUL, as this simplifies our implementation. We match + angle brackets in nested pairs: several languages use them for + generics/template types. */ +%x SC_TAG + + /* Four types of user code: + - prologue (code between '%{' '%}' in the first section, before %%); - actions, printers, union, etc, (between braced in the middle section); - - epilogue (everything after the second %%). */ -%x SC_PROLOGUE SC_BRACED_CODE SC_EPILOGUE + - epilogue (everything after the second %%). + - predicate (code between '%?{' and '{' in middle section); */ +%x SC_PROLOGUE SC_BRACED_CODE SC_EPILOGUE SC_PREDICATE /* C and C++ comments in code. */ %x SC_COMMENT SC_LINE_COMMENT /* Strings and characters in code. */ @@ -103,30 +133,30 @@ static void unexpected_newline (boundary, char const *); /* Bracketed identifiers support. */ %x SC_BRACKETED_ID SC_RETURN_BRACKETED_ID -letter [-.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_] -id {letter}({letter}|[0-9])* -directive %{id} -int [0-9]+ - -/* POSIX says that a tag must be both an id and a C union member, but - historically almost any character is allowed in a tag. We disallow - NUL and newline, as this simplifies our implementation. */ -tag [^\0\n>]+ +letter [.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_] +notletter [^.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_]{-}[%\{] +id {letter}({letter}|[-0-9])* +int [0-9]+ /* Zero or more instances of backslash-newline. Following GCC, allow white space between the backslash and the newline. */ -splice (\\[ \f\t\v]*\n)* +splice (\\[ \f\t\v]*\n)* + +/* An equal sign, with optional leading whitespaces. This is used in some + deprecated constructs. */ +eqopt ([[:space:]]*=)? %% %{ - /* Nesting level of the current code in braces. */ - int braces_level IF_LINT (= 0); + /* Nesting level. Either for nested braces, or nested angle brackets + (but not mixed). */ + int nesting PACIFY_CC (= 0); /* Parent context state, when applicable. */ - int context_state IF_LINT (= 0); + int context_state PACIFY_CC (= 0); /* Location of most recent identifier, when applicable. */ - location id_loc IF_LINT (= empty_location); + location id_loc PACIFY_CC (= empty_location); /* Where containing code started, when applicable. Its initial value is relevant only when yylex is invoked in the SC_EPILOGUE @@ -135,7 +165,7 @@ splice (\\[ \f\t\v]*\n)* /* Where containing comment or string or character literal started, when applicable. */ - boundary token_start IF_LINT (= scanner_cursor); + boundary token_start PACIFY_CC (= scanner_cursor); %} @@ -146,9 +176,11 @@ splice (\\[ \f\t\v]*\n)* { /* Comments and white space. */ - "," warn_at (*loc, _("stray `,' treated as white space")); + "," { + complain (loc, Wother, _("stray ',' treated as white space")); + } [ \f\n\t\v] | - "//".* ; + "//".* continue; "/*" { token_start = loc->start; context_state = YY_START; @@ -157,7 +189,7 @@ splice (\\[ \f\t\v]*\n)* /* #line directives are not documented, and may be withdrawn or modified in future versions of Bison. */ - ^"#line "{int}" \"".*"\"\n" { + ^"#line "{int}(" \"".*"\"")?"\n" { handle_syncline (yytext + sizeof "#line " - 1, *loc); } } @@ -168,7 +200,7 @@ splice (\\[ \f\t\v]*\n)* `----------------------------*/ /* For directives that are also command line options, the regex must be - "%..." + "%..." after "[-_]"s are removed, and the directive must match the --long option name, with a single string argument. Otherwise, add exceptions to ../build-aux/cross-options.pl. */ @@ -177,56 +209,70 @@ splice (\\[ \f\t\v]*\n)* { "%binary" return PERCENT_NONASSOC; "%code" return PERCENT_CODE; - "%debug" return PERCENT_DEBUG; - "%default"[-_]"prec" return PERCENT_DEFAULT_PREC; + "%debug" RETURN_PERCENT_FLAG("parse.trace"); + "%default-prec" return PERCENT_DEFAULT_PREC; "%define" return PERCENT_DEFINE; "%defines" return PERCENT_DEFINES; "%destructor" return PERCENT_DESTRUCTOR; "%dprec" return PERCENT_DPREC; - "%error"[-_]"verbose" return PERCENT_ERROR_VERBOSE; + "%empty" return PERCENT_EMPTY; + "%error-verbose" return PERCENT_ERROR_VERBOSE; "%expect" return PERCENT_EXPECT; - "%expect"[-_]"rr" return PERCENT_EXPECT_RR; + "%expect-rr" return PERCENT_EXPECT_RR; "%file-prefix" return PERCENT_FILE_PREFIX; - "%fixed"[-_]"output"[-_]"files" return PERCENT_YACC; + "%fixed-output-files" return PERCENT_YACC; "%initial-action" return PERCENT_INITIAL_ACTION; "%glr-parser" return PERCENT_GLR_PARSER; "%language" return PERCENT_LANGUAGE; "%left" return PERCENT_LEFT; - "%lex-param" return PERCENT_LEX_PARAM; - "%locations" return PERCENT_LOCATIONS; + "%lex-param" RETURN_PERCENT_PARAM(lex); + "%locations" RETURN_PERCENT_FLAG("locations"); "%merge" return PERCENT_MERGE; - "%name"[-_]"prefix" return PERCENT_NAME_PREFIX; - "%no"[-_]"default"[-_]"prec" return PERCENT_NO_DEFAULT_PREC; - "%no"[-_]"lines" return PERCENT_NO_LINES; + "%name-prefix" return PERCENT_NAME_PREFIX; + "%no-default-prec" return PERCENT_NO_DEFAULT_PREC; + "%no-lines" return PERCENT_NO_LINES; "%nonassoc" return PERCENT_NONASSOC; "%nondeterministic-parser" return PERCENT_NONDETERMINISTIC_PARSER; "%nterm" return PERCENT_NTERM; "%output" return PERCENT_OUTPUT; - "%parse-param" return PERCENT_PARSE_PARAM; + "%param" RETURN_PERCENT_PARAM(both); + "%parse-param" RETURN_PERCENT_PARAM(parse); "%prec" return PERCENT_PREC; + "%precedence" return PERCENT_PRECEDENCE; "%printer" return PERCENT_PRINTER; - "%pure"[-_]"parser" return PERCENT_PURE_PARSER; + "%pure-parser" RETURN_PERCENT_FLAG("api.pure"); "%require" return PERCENT_REQUIRE; "%right" return PERCENT_RIGHT; "%skeleton" return PERCENT_SKELETON; "%start" return PERCENT_START; "%term" return PERCENT_TOKEN; "%token" return PERCENT_TOKEN; - "%token"[-_]"table" return PERCENT_TOKEN_TABLE; + "%token-table" return PERCENT_TOKEN_TABLE; "%type" return PERCENT_TYPE; "%union" return PERCENT_UNION; "%verbose" return PERCENT_VERBOSE; "%yacc" return PERCENT_YACC; - {directive} { - complain_at (*loc, _("invalid directive: %s"), quote (yytext)); + /* deprecated */ + "%default"[-_]"prec" DEPRECATED("%default-prec"); + "%error"[-_]"verbose" DEPRECATED("%define parse.error verbose"); + "%expect"[-_]"rr" DEPRECATED("%expect-rr"); + "%file-prefix"{eqopt} DEPRECATED("%file-prefix"); + "%fixed"[-_]"output"[-_]"files" DEPRECATED("%fixed-output-files"); + "%name"[-_]"prefix"{eqopt} DEPRECATED("%name-prefix"); + "%no"[-_]"default"[-_]"prec" DEPRECATED("%no-default-prec"); + "%no"[-_]"lines" DEPRECATED("%no-lines"); + "%output"{eqopt} DEPRECATED("%output"); + "%pure"[-_]"parser" DEPRECATED("%pure-parser"); + "%token"[-_]"table" DEPRECATED("%token-table"); + + "%"{id} { + complain (loc, complaint, _("invalid directive: %s"), quote (yytext)); } "=" return EQUAL; "|" return PIPE; ";" return SEMICOLON; - "<*>" return TYPE_TAG_ANY; - "<>" return TYPE_TAG_NONE; {id} { val->uniqstr = uniqstr_new (yytext); @@ -247,14 +293,14 @@ splice (\\[ \f\t\v]*\n)* /* Identifiers may not start with a digit. Yet, don't silently accept "1FOO" as "1 FOO". */ {int}{id} { - complain_at (*loc, _("invalid identifier: %s"), quote (yytext)); + complain (loc, complaint, _("invalid identifier: %s"), quote (yytext)); } /* Characters. */ - "'" STRING_GROW; token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER; + "'" token_start = loc->start; BEGIN SC_ESCAPED_CHARACTER; /* Strings. */ - "\"" token_start = loc->start; BEGIN SC_ESCAPED_STRING; + "\"" token_start = loc->start; BEGIN SC_ESCAPED_STRING; /* Prologue. */ "%{" code_start = loc->start; BEGIN SC_PROLOGUE; @@ -262,18 +308,25 @@ splice (\\[ \f\t\v]*\n)* /* Code in between braces. */ "{" { STRING_GROW; - braces_level = 0; + nesting = 0; code_start = loc->start; BEGIN SC_BRACED_CODE; } + /* Semantic predicate. */ + "%?"[ \f\n\t\v]*"{" { + nesting = 0; + code_start = loc->start; + BEGIN SC_PREDICATE; + } + /* A type. */ - "<"{tag}">" { - obstack_grow (&obstack_for_string, yytext + 1, yyleng - 2); - STRING_FINISH; - val->uniqstr = uniqstr_new (last_string); - STRING_FREE; - return TYPE; + "<*>" return TAG_ANY; + "<>" return TAG_NONE; + "<" { + nesting = 0; + token_start = loc->start; + BEGIN SC_TAG; } "%%" { @@ -290,8 +343,10 @@ splice (\\[ \f\t\v]*\n)* BEGIN SC_BRACKETED_ID; } - . { - complain_at (*loc, _("invalid character: %s"), quote (yytext)); + [^\[%A-Za-z0-9_<>{}\"\'*;|=/, \f\n\t\v]+|. { + complain (loc, complaint, "%s: %s", + ngettext ("invalid character", "invalid characters", yyleng), + quote_mem (yytext, yyleng)); } <> { @@ -301,6 +356,17 @@ splice (\\[ \f\t\v]*\n)* } + /*--------------------------------------------------------------. + | Supporting \0 complexifies our implementation for no expected | + | added value. | + `--------------------------------------------------------------*/ + + +{ + \0 complain (loc, complaint, _("invalid null character")); +} + + /*-----------------------------------------------------------------. | Scanning after an identifier, checking whether a colon is next. | `-----------------------------------------------------------------*/ @@ -310,16 +376,16 @@ splice (\\[ \f\t\v]*\n)* "[" { if (bracketed_id_str) { - ROLLBACK_CURRENT_TOKEN; - BEGIN SC_RETURN_BRACKETED_ID; - *loc = id_loc; - return ID; + ROLLBACK_CURRENT_TOKEN; + BEGIN SC_RETURN_BRACKETED_ID; + *loc = id_loc; + return ID; } else { - bracketed_id_start = loc->start; - bracketed_id_context_state = YY_START; - BEGIN SC_BRACKETED_ID; + bracketed_id_start = loc->start; + bracketed_id_context_state = YY_START; + BEGIN SC_BRACKETED_ID; } } ":" { @@ -349,34 +415,39 @@ splice (\\[ \f\t\v]*\n)* {id} { if (bracketed_id_str) { - complain_at (*loc, _("unexpected identifier in bracketed name: %s"), - quote (yytext)); + complain (loc, complaint, + _("unexpected identifier in bracketed name: %s"), + quote (yytext)); } else { - bracketed_id_str = uniqstr_new (yytext); - bracketed_id_loc = *loc; + bracketed_id_str = uniqstr_new (yytext); + bracketed_id_loc = *loc; } } "]" { BEGIN bracketed_id_context_state; if (bracketed_id_str) { - if (INITIAL == bracketed_id_context_state) - { - val->uniqstr = bracketed_id_str; - bracketed_id_str = 0; - *loc = bracketed_id_loc; - return BRACKETED_ID; - } + if (INITIAL == bracketed_id_context_state) + { + val->uniqstr = bracketed_id_str; + bracketed_id_str = 0; + *loc = bracketed_id_loc; + return BRACKETED_ID; + } } else - complain_at (*loc, _("an identifier expected")); + complain (loc, complaint, _("an identifier expected")); } - . { - complain_at (*loc, _("invalid character in bracketed name: %s"), - quote (yytext)); + + [^\].A-Za-z0-9_/ \f\n\t\v]+|. { + complain (loc, complaint, "%s: %s", + ngettext ("invalid character in bracketed name", + "invalid characters in bracketed name", yyleng), + quote_mem (yytext, yyleng)); } + <> { BEGIN bracketed_id_context_state; unexpected_eof (bracketed_id_start, "]"); @@ -397,37 +468,37 @@ splice (\\[ \f\t\v]*\n)* /*---------------------------------------------------------------. - | Scanning a Yacc comment. The initial `/ *' is already eaten. | + | Scanning a Yacc comment. The initial '/ *' is already eaten. | `---------------------------------------------------------------*/ { "*/" BEGIN context_state; - .|\n ; + .|\n continue; <> unexpected_eof (token_start, "*/"); BEGIN context_state; } /*------------------------------------------------------------. - | Scanning a C comment. The initial `/ *' is already eaten. | + | Scanning a C comment. The initial '/ *' is already eaten. | `------------------------------------------------------------*/ { "*"{splice}"/" STRING_GROW; BEGIN context_state; - <> unexpected_eof (token_start, "*/"); BEGIN context_state; + <> unexpected_eof (token_start, "*/"); BEGIN context_state; } /*--------------------------------------------------------------. - | Scanning a line comment. The initial `//' is already eaten. | + | Scanning a line comment. The initial '//' is already eaten. | `--------------------------------------------------------------*/ { - "\n" STRING_GROW; BEGIN context_state; - {splice} STRING_GROW; - <> BEGIN context_state; + "\n" STRING_GROW; BEGIN context_state; + {splice} STRING_GROW; + <> BEGIN context_state; } @@ -438,78 +509,74 @@ splice (\\[ \f\t\v]*\n)* { - "\""|"\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; + val->code = last_string; BEGIN INITIAL; return STRING; } + <> unexpected_eof (token_start, "\""); + "\n" unexpected_newline (token_start, "\""); } /*----------------------------------------------------------. | Scanning a Bison character literal, decoding its escapes. | - | The initial quote is already eaten. | + | The initial quote is already eaten. | `----------------------------------------------------------*/ { - "'"|"\n" { - STRING_GROW; - STRING_FINISH; - loc->start = token_start; - val->character = last_string[1]; - { - /* FIXME: Eventually, make these errors. */ - size_t length = strlen (last_string); - if (strlen (last_string) < 3) - warn_at (*loc, _("empty character literal")); - else if (strlen (last_string) > 3) - warn_at (*loc, _("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') { - size_t length = strlen (last_string); - /* FIXME: Eventually, make these errors. */ - if (length < 2) - warn_at (*loc, _("empty character literal")); - else if (length > 2) - warn_at (*loc, _("extra characters in character literal")); - if (length > 1) - val->character = last_string[1]; - else - val->character = last_string[0]; + 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, "'"); } - + + + /*--------------------------------------------------------------. + | Scanning a tag. The initial angle bracket is already eaten. | + `--------------------------------------------------------------*/ + + { - \0 complain_at (*loc, _("invalid null character")); -} + ">" { + --nesting; + if (nesting < 0) + { + STRING_FINISH; + loc->start = token_start; + val->uniqstr = uniqstr_new (last_string); + STRING_FREE; + BEGIN INITIAL; + return TAG; + } + STRING_GROW; + } + + ([^<>]|->)+ STRING_GROW; + "<"+ STRING_GROW; nesting += yyleng; + <> unexpected_eof (token_start, ">"); +} /*----------------------------. | Decode escaped characters. | @@ -519,10 +586,9 @@ splice (\\[ \f\t\v]*\n)* { \\[0-7]{1,3} { unsigned long int c = strtoul (yytext + 1, NULL, 8); - if (UCHAR_MAX < c) - complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext)); - else if (! c) - complain_at (*loc, _("invalid null character: %s"), quote (yytext)); + if (!c || UCHAR_MAX < c) + complain (loc, complaint, _("invalid number after \\-escape: %s"), + yytext+1); else obstack_1grow (&obstack_for_string, c); } @@ -530,37 +596,41 @@ splice (\\[ \f\t\v]*\n)* \\x[0-9abcdefABCDEF]+ { verify (UCHAR_MAX < ULONG_MAX); unsigned long int c = strtoul (yytext + 2, NULL, 16); - if (UCHAR_MAX < c) - complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext)); - else if (! c) - complain_at (*loc, _("invalid null character: %s"), quote (yytext)); + if (!c || UCHAR_MAX < c) + complain (loc, complaint, _("invalid number after \\-escape: %s"), + yytext+1); else obstack_1grow (&obstack_for_string, c); } - \\a obstack_1grow (&obstack_for_string, '\a'); - \\b obstack_1grow (&obstack_for_string, '\b'); - \\f obstack_1grow (&obstack_for_string, '\f'); - \\n obstack_1grow (&obstack_for_string, '\n'); - \\r obstack_1grow (&obstack_for_string, '\r'); - \\t obstack_1grow (&obstack_for_string, '\t'); - \\v obstack_1grow (&obstack_for_string, '\v'); + \\a obstack_1grow (&obstack_for_string, '\a'); + \\b obstack_1grow (&obstack_for_string, '\b'); + \\f obstack_1grow (&obstack_for_string, '\f'); + \\n obstack_1grow (&obstack_for_string, '\n'); + \\r obstack_1grow (&obstack_for_string, '\r'); + \\t obstack_1grow (&obstack_for_string, '\t'); + \\v obstack_1grow (&obstack_for_string, '\v'); /* \\[\"\'?\\] would be shorter, but it confuses xgettext. */ \\("\""|"'"|"?"|"\\") obstack_1grow (&obstack_for_string, yytext[1]); \\(u|U[0-9abcdefABCDEF]{4})[0-9abcdefABCDEF]{4} { int c = convert_ucn_to_byte (yytext); - if (c < 0) - complain_at (*loc, _("invalid escape sequence: %s"), quote (yytext)); - else if (! c) - complain_at (*loc, _("invalid null character: %s"), quote (yytext)); + if (c <= 0) + complain (loc, complaint, _("invalid number after \\-escape: %s"), + yytext+1); else obstack_1grow (&obstack_for_string, c); } - \\(.|\n) { - complain_at (*loc, _("unrecognized escape sequence: %s"), quote (yytext)); - STRING_GROW; + \\(.|\n) { + char const *p = yytext + 1; + /* Quote only if escaping won't make the character visible. */ + if (c_isspace ((unsigned char) *p) && c_isprint ((unsigned char) *p)) + p = quote (p); + else + p = quotearg_style_mem (escape_quoting_style, p, 1); + complain (loc, complaint, _("invalid character after \\-escape: %s"), + p); } } @@ -570,21 +640,21 @@ splice (\\[ \f\t\v]*\n)* { - {splice}|\\{splice}[^\n\[\]] STRING_GROW; + {splice}|\\{splice}[^\n\[\]] STRING_GROW; } { - "'" STRING_GROW; BEGIN context_state; - \n unexpected_newline (token_start, "'"); BEGIN context_state; - <> unexpected_eof (token_start, "'"); BEGIN context_state; + "'" STRING_GROW; 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; + "\"" STRING_GROW; BEGIN context_state; + \n unexpected_newline (token_start, "\""); + <> unexpected_eof (token_start, "\""); } @@ -592,7 +662,7 @@ splice (\\[ \f\t\v]*\n)* | Strings, comments etc. can be found in user code. | `---------------------------------------------------*/ - + { "'" { STRING_GROW; @@ -622,43 +692,56 @@ splice (\\[ \f\t\v]*\n)* /*-----------------------------------------------------------. - | Scanning some code in braces (actions). The initial "{" is | - | already eaten. | + | Scanning some code in braces (actions, predicates). The | + | initial "{" is already eaten. | `-----------------------------------------------------------*/ + +{ + "{"|"<"{splice}"%" STRING_GROW; nesting++; + "%"{splice}">" STRING_GROW; nesting--; + + /* Tokenize '<<%' correctly (as '<<' '%') rather than incorrrectly + (as '<' '<%'). */ + "<"{splice}"<" STRING_GROW; + + <> unexpected_eof (code_start, "}"); +} + { - "{"|"<"{splice}"%" STRING_GROW; braces_level++; - "%"{splice}">" STRING_GROW; braces_level--; "}" { obstack_1grow (&obstack_for_string, '}'); - --braces_level; - if (braces_level < 0) + --nesting; + if (nesting < 0) { - STRING_FINISH; - loc->start = code_start; - val->code = last_string; - BEGIN INITIAL; - return BRACED_CODE; + STRING_FINISH; + loc->start = code_start; + val->code = last_string; + BEGIN INITIAL; + return BRACED_CODE; } } +} - /* Tokenize `<<%' correctly (as `<<' `%') rather than incorrrectly - (as `<' `<%'). */ - "<"{splice}"<" STRING_GROW; - - <> { - unexpected_eof (code_start, "}"); - STRING_FINISH; - loc->start = code_start; - val->code = last_string; - BEGIN INITIAL; - return BRACED_CODE; + +{ + "}" { + --nesting; + if (nesting < 0) + { + STRING_FINISH; + loc->start = code_start; + val->code = last_string; + BEGIN INITIAL; + return BRACED_PREDICATE; + } + else + obstack_1grow (&obstack_for_string, '}'); } } - /*--------------------------------------------------------------. | Scanning some prologue: from "%{" (already scanned) to "%}". | `--------------------------------------------------------------*/ @@ -668,19 +751,12 @@ splice (\\[ \f\t\v]*\n)* "%}" { STRING_FINISH; loc->start = code_start; - val->chars = last_string; + val->code = last_string; BEGIN INITIAL; return PROLOGUE; } - <> { - unexpected_eof (code_start, "%}"); - STRING_FINISH; - loc->start = code_start; - val->chars = last_string; - BEGIN INITIAL; - return PROLOGUE; - } + <> unexpected_eof (code_start, "%}"); } @@ -694,7 +770,7 @@ splice (\\[ \f\t\v]*\n)* <> { STRING_FINISH; loc->start = code_start; - val->chars = last_string; + val->code = last_string; BEGIN INITIAL; return EPILOGUE; } @@ -705,8 +781,8 @@ splice (\\[ \f\t\v]*\n)* | By default, grow the string obstack with the input. | `-----------------------------------------------------*/ -. | -\n STRING_GROW; +. | + \n STRING_GROW; %% @@ -722,35 +798,35 @@ no_cr_read (FILE *fp, char *buf, size_t size) { char *w = memchr (buf, '\r', bytes_read); if (w) - { - char const *r = ++w; - char const *lim = buf + bytes_read; - - for (;;) - { - /* Found an '\r'. Treat it like '\n', but ignore any - '\n' that immediately follows. */ - w[-1] = '\n'; - if (r == lim) - { - int ch = getc (fp); - if (ch != '\n' && ungetc (ch, fp) != ch) - break; - } - else if (*r == '\n') - r++; - - /* Copy until the next '\r'. */ - do - { - if (r == lim) - return w - buf; - } - while ((*w++ = *r++) != '\r'); - } - - return w - buf; - } + { + char const *r = ++w; + char const *lim = buf + bytes_read; + + for (;;) + { + /* Found an '\r'. Treat it like '\n', but ignore any + '\n' that immediately follows. */ + w[-1] = '\n'; + if (r == lim) + { + int ch = getc (fp); + if (ch != '\n' && ungetc (ch, fp) != ch) + break; + } + else if (*r == '\n') + r++; + + /* Copy until the next '\r'. */ + do + { + if (r == lim) + return w - buf; + } + while ((*w++ = *r++) != '\r'); + } + + return w - buf; + } } return bytes_read; @@ -770,7 +846,8 @@ scan_integer (char const *number, int base, location loc) if (INT_MAX < num) { - complain_at (loc, _("integer out of range: %s"), quote (number)); + complain (&loc, complaint, _("integer out of range: %s"), + quote (number)); num = INT_MAX; } @@ -781,7 +858,7 @@ scan_integer (char const *number, int base, location loc) /*------------------------------------------------------------------. | Convert universal character name UCN to a single-byte character, | | and return that character. Return -1 if UCN does not correspond | -| to a single-byte character. | +| to a single-byte character. | `------------------------------------------------------------------*/ static int @@ -809,22 +886,22 @@ convert_ucn_to_byte (char const *ucn) about. */ static signed char const table[] = { - '\0', -1, -1, -1, -1, -1, -1, '\a', - '\b', '\t', '\n', '\v', '\f', '\r', -1, -1, - -1, -1, -1, -1, -1, -1, -1, -1, - -1, -1, -1, -1, -1, -1, -1, -1, - ' ', '!', '"', '#', '$', '%', '&', '\'', - '(', ')', '*', '+', ',', '-', '.', '/', - '0', '1', '2', '3', '4', '5', '6', '7', - '8', '9', ':', ';', '<', '=', '>', '?', - '@', 'A', 'B', 'C', 'D', 'E', 'F', 'G', - 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', - 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', - 'X', 'Y', 'Z', '[', '\\', ']', '^', '_', - '`', 'a', 'b', 'c', 'd', 'e', 'f', 'g', - 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', - 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', - 'x', 'y', 'z', '{', '|', '}', '~' + '\0', -1, -1, -1, -1, -1, -1, '\a', + '\b', '\t', '\n', '\v', '\f', '\r', -1, -1, + -1, -1, -1, -1, -1, -1, -1, -1, + -1, -1, -1, -1, -1, -1, -1, -1, + ' ', '!', '"', '#', '$', '%', '&', '\'', + '(', ')', '*', '+', ',', '-', '.', '/', + '0', '1', '2', '3', '4', '5', '6', '7', + '8', '9', ':', ';', '<', '=', '>', '?', + '@', 'A', 'B', 'C', 'D', 'E', 'F', 'G', + 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', + 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', + 'X', 'Y', 'Z', '[', '\\', ']', '^', '_', + '`', 'a', 'b', 'c', 'd', 'e', 'f', 'g', + 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', + 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', + 'x', 'y', 'z', '{', '|', '}', '~' }; code = code < sizeof table ? table[code] : -1; @@ -835,31 +912,35 @@ convert_ucn_to_byte (char const *ucn) } -/*----------------------------------------------------------------. -| Handle `#line INT "FILE"'. ARGS has already skipped `#line '. | -`----------------------------------------------------------------*/ +/*---------------------------------------------------------------------. +| Handle '#line INT( "FILE")?\n'. ARGS has already skipped '#line '. | +`---------------------------------------------------------------------*/ static void handle_syncline (char *args, location loc) { - char *after_num; - unsigned long int lineno = strtoul (args, &after_num, 10); - char *file = strchr (after_num, '"') + 1; - *strchr (file, '"') = '\0'; + char *file; + unsigned long int lineno = strtoul (args, &file, 10); if (INT_MAX <= lineno) { - warn_at (loc, _("line number overflow")); + complain (&loc, Wother, _("line number overflow")); lineno = INT_MAX; } - current_file = uniqstr_new (file); + + file = strchr (file, '"'); + if (file) + { + *strchr (file + 1, '"') = '\0'; + current_file = uniqstr_new (file + 1); + } boundary_set (&scanner_cursor, current_file, lineno, 1); } /*----------------------------------------------------------------. | 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 @@ -868,19 +949,33 @@ unexpected_end (boundary start, char const *msgid, char const *token_end) location loc; loc.start = start; loc.end = scanner_cursor; - complain_at (loc, _(msgid), token_end); + 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, "'\\''")) + token_end = "\"'\""; + complain (&loc, complaint, _(msgid), 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 unexpected_eof (boundary start, char const *token_end) { - unexpected_end (start, N_("missing `%s' at end of file"), token_end); + unexpected_end (start, N_("missing %s at end of file"), token_end); } @@ -891,7 +986,7 @@ unexpected_eof (boundary start, char const *token_end) static void unexpected_newline (boundary start, char const *token_end) { - unexpected_end (start, N_("missing `%s' at end of line"), token_end); + unexpected_end (start, N_("missing %s at end of line"), token_end); }