X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/457bf91968a87a01dfb5317b008497ac4dcde8a4..06ec0105b12cdfa2994283e345154e6720354808:/src/scan-gram.l?ds=sidebyside diff --git a/src/scan-gram.l b/src/scan-gram.l index dbea2a1d..98124556 100644 --- a/src/scan-gram.l +++ b/src/scan-gram.l @@ -17,7 +17,7 @@ You should have received a copy of the GNU General Public License along with this program. If not, see . */ -%option debug nodefault noinput nounput noyywrap never-interactive +%option debug nodefault noinput noyywrap never-interactive %option prefix="gram_" outfile="lex.yy.c" %{ @@ -27,25 +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 +56,35 @@ 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; \ + complain (loc, Wdeprecated, \ + _("deprecated directive: %s, use %s"), \ + quote (yytext), quote_n (1, 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; @@ -90,12 +114,15 @@ static void unexpected_newline (boundary, char const *); /* A identifier was just read in directives/rules. Special state to capture the sequence 'identifier :'. */ %x SC_AFTER_IDENTIFIER + /* A complex tag, with nested angles brackets. */ +%x SC_TAG - /* Three types of user code: + /* 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. */ @@ -103,24 +130,31 @@ 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])* +letter [.abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_] +id {letter}({letter}|[-0-9])* directive %{id} -int [0-9]+ +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>]+ + NUL, as this simplifies our implementation. We disallow angle + bracket to match them in nested pairs: several languages use them + for generics/template types. */ +tag [^\0<>]+ /* 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 PACIFY_CC (= 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 PACIFY_CC (= 0); @@ -146,7 +180,9 @@ 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] | "//".* ; "/*" { @@ -168,7 +204,7 @@ 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 +213,69 @@ 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; + "%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; + /* 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"); + {directive} { - complain_at (*loc, _("invalid directive: %s"), quote (yytext)); + 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 +296,14 @@ 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. */ - "'" 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 +311,32 @@ 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. */ + "<*>" return TAG_ANY; + "<>" return TAG_NONE; "<"{tag}">" { obstack_grow (&obstack_for_string, yytext + 1, yyleng - 2); STRING_FINISH; val->uniqstr = uniqstr_new (last_string); STRING_FREE; - return TYPE; + return TAG; + } + "<" { + nesting = 0; + token_start = loc->start; + BEGIN SC_TAG; } "%%" { @@ -291,7 +354,8 @@ splice (\\[ \f\t\v]*\n)* } . { - complain_at (*loc, _("invalid character: %s"), quote (yytext)); + complain (loc, complaint, _("invalid character: %s"), + quote_mem (yytext, yyleng)); } <> { @@ -301,6 +365,17 @@ 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 +385,16 @@ 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; } } ":" { @@ -349,33 +424,34 @@ 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)); + complain (loc, complaint, _("invalid character in bracketed name: %s"), + quote_mem (yytext, yyleng)); } <> { BEGIN bracketed_id_context_state; @@ -403,7 +479,7 @@ splice (\\[ \f\t\v]*\n)* { "*/" BEGIN context_state; - .|\n ; + .|\n ; <> unexpected_eof (token_start, "*/"); BEGIN context_state; } @@ -415,7 +491,7 @@ splice (\\[ \f\t\v]*\n)* { "*"{splice}"/" STRING_GROW; BEGIN context_state; - <> unexpected_eof (token_start, "*/"); BEGIN context_state; + <> unexpected_eof (token_start, "*/"); BEGIN context_state; } @@ -425,9 +501,9 @@ splice (\\[ \f\t\v]*\n)* { - "\n" STRING_GROW; BEGIN context_state; - {splice} STRING_GROW; - <> BEGIN context_state; + "\n" STRING_GROW; BEGIN context_state; + {splice} STRING_GROW; + <> BEGIN context_state; } @@ -459,7 +535,7 @@ splice (\\[ \f\t\v]*\n)* /*----------------------------------------------------------. | Scanning a Bison character literal, decoding its escapes. | - | The initial quote is already eaten. | + | The initial quote is already eaten. | `----------------------------------------------------------*/ @@ -472,12 +548,13 @@ splice (\\[ \f\t\v]*\n)* /* FIXME: Eventually, make these errors. */ if (last_string[0] == '\0') { - warn_at (*loc, _("empty character literal")); + complain (loc, Wother, _("empty character literal")); /* '\0' seems dangerous even if we are about to complain. */ val->character = '\''; } else if (last_string[1] != '\0') - warn_at (*loc, _("extra characters in character literal")); + complain (loc, Wother, + _("extra characters in character literal")); } if (yytext[0] == '\n') unexpected_newline (token_start, "'"); @@ -493,12 +570,13 @@ splice (\\[ \f\t\v]*\n)* /* FIXME: Eventually, make these errors. */ if (last_string[0] == '\0') { - warn_at (*loc, _("empty character literal")); + complain (loc, Wother, _("empty character literal")); /* '\0' seems dangerous even if we are about to complain. */ val->character = '\''; } else if (last_string[1] != '\0') - warn_at (*loc, _("extra characters in character literal")); + complain (loc, Wother, + _("extra characters in character literal")); } unexpected_eof (token_start, "'"); STRING_FREE; @@ -507,11 +585,40 @@ splice (\\[ \f\t\v]*\n)* } } - + /*-----------------------------------------------------------. + | Scanning a Bison nested 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, ">"); + STRING_FINISH; + loc->start = token_start; + val->uniqstr = uniqstr_new (last_string); + STRING_FREE; + BEGIN INITIAL; + return TAG; + } +} /*----------------------------. | Decode escaped characters. | @@ -522,7 +629,7 @@ splice (\\[ \f\t\v]*\n)* \\[0-7]{1,3} { unsigned long int c = strtoul (yytext + 1, NULL, 8); if (!c || UCHAR_MAX < c) - complain_at (*loc, _("invalid number after \\-escape: %s"), + complain (loc, complaint, _("invalid number after \\-escape: %s"), yytext+1); else obstack_1grow (&obstack_for_string, c); @@ -532,19 +639,19 @@ splice (\\[ \f\t\v]*\n)* verify (UCHAR_MAX < ULONG_MAX); unsigned long int c = strtoul (yytext + 2, NULL, 16); if (!c || UCHAR_MAX < c) - complain_at (*loc, _("invalid number after \\-escape: %s"), + 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]); @@ -552,19 +659,20 @@ splice (\\[ \f\t\v]*\n)* \\(u|U[0-9abcdefABCDEF]{4})[0-9abcdefABCDEF]{4} { int c = convert_ucn_to_byte (yytext); if (c <= 0) - complain_at (*loc, _("invalid number after \\-escape: %s"), + complain (loc, complaint, _("invalid number after \\-escape: %s"), yytext+1); else obstack_1grow (&obstack_for_string, c); } - \\(.|\n) { + \\(.|\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_at (*loc, _("invalid character after \\-escape: %s"), p); + complain (loc, complaint, _("invalid character after \\-escape: %s"), + p); } } @@ -574,21 +682,21 @@ 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, "'"); BEGIN context_state; + <> unexpected_eof (token_start, "'"); BEGIN context_state; } { - "\"" 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, "\""); BEGIN context_state; + <> unexpected_eof (token_start, "\""); BEGIN context_state; } @@ -596,7 +704,7 @@ splice (\\[ \f\t\v]*\n)* | Strings, comments etc. can be found in user code. | `---------------------------------------------------*/ - + { "'" { STRING_GROW; @@ -626,42 +734,63 @@ 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; braces_level++; - "%"{splice}">" STRING_GROW; braces_level--; - "}" { - obstack_1grow (&obstack_for_string, '}'); - - --braces_level; - if (braces_level < 0) - { - STRING_FINISH; - loc->start = code_start; - val->code = last_string; - BEGIN INITIAL; - return BRACED_CODE; - } - } + "{"|"<"{splice}"%" STRING_GROW; nesting++; + "%"{splice}">" STRING_GROW; nesting--; /* Tokenize '<<%' correctly (as '<<' '%') rather than incorrrectly (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 BRACED_CODE; + return token; } } + +{ + "}" { + obstack_1grow (&obstack_for_string, '}'); + + --nesting; + if (nesting < 0) + { + 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 "%}". | @@ -709,8 +838,8 @@ splice (\\[ \f\t\v]*\n)* | By default, grow the string obstack with the input. | `-----------------------------------------------------*/ -. | -\n STRING_GROW; +. | + \n STRING_GROW; %% @@ -726,35 +855,35 @@ 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; @@ -774,7 +903,8 @@ 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; } @@ -785,7 +915,7 @@ 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 @@ -813,22 +943,22 @@ 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; @@ -848,11 +978,11 @@ handle_syncline (char *args, location loc) { char *after_num; unsigned long int lineno = strtoul (args, &after_num, 10); - char *file = mbschr (after_num, '"') + 1; - *mbschr (file, '"') = '\0'; + char *file = strchr (after_num, '"') + 1; + *strchr (file, '"') = '\0'; if (INT_MAX <= lineno) { - warn_at (loc, _("line number overflow")); + complain (&loc, Wother, _("line number overflow")); lineno = INT_MAX; } current_file = uniqstr_new (file); @@ -862,8 +992,8 @@ 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. | `----------------------------------------------------------------*/ static void @@ -874,9 +1004,9 @@ unexpected_end (boundary start, char const *msgid, char const *token_end) loc.end = scanner_cursor; token_end = quote (token_end); // Instead of '\'', display "'". - if (!strcmp (token_end, "'\\''")) + if (STREQ (token_end, "'\\''")) token_end = "\"'\""; - complain_at (loc, _(msgid), token_end); + complain (&loc, complaint, _(msgid), token_end); }