%{
-/* Copyright (C) 1991 Free Software Foundation, Inc.
+
+/* Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
+ 2000, 2001, 2002 Free Software Foundation, Inc.
This file is part of GLD, the Gnu Linker.
GLD is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 1, or (at your option)
+the Free Software Foundation; either version 2, or (at your option)
any later version.
GLD is distributed in the hope that it will be useful,
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
-along with GLD; see the file COPYING. If not, write to
-the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */
+along with GLD; see the file COPYING. If not, write to the Free
+Software Foundation, 59 Temple Place - Suite 330, Boston, MA
+02111-1307, USA. */
/*
- * $Id$
-
- *
+This was written by steve chamberlain
+ sac@cygnus.com
*/
+#include <ansidecl.h>
+#include <stdio.h>
-/*SUPPRESS 529*/
-/*SUPPRESS 26*/
-/*SUPPRESS 29*/
-#define LEXDEBUG 0
-#include "sysdep.h"
-#include "bfd.h"
-
-#include <ctype.h>
-#include "ldlex.h"
+#ifdef MPW
+/* Prevent enum redefinition problems. */
+#define TRUE_FALSE_ALREADY_DEFINED
+#endif /* MPW */
+#include "bfd.h"
+#include "sysdep.h"
+#include "safe-ctype.h"
#include "ld.h"
-#include "ldexp.h"
-#include "ldgram.tab.h"
+#include "ldgram.h"
#include "ldmisc.h"
+#include "ldexp.h"
+#include "ldlang.h"
+#include "ldfile.h"
+#include "ldlex.h"
+#include "ldmain.h"
+#include "libiberty.h"
+
+/* The type of top-level parser input.
+ yylex and yyparse (indirectly) both check this. */
+input_type parser_input;
+
+/* Line number in the current input file.
+ (FIXME Actually, it doesn't appear to get reset for each file?) */
+unsigned int lineno = 1;
+
+/* The string we are currently lexing, or NULL if we are reading a
+ file. */
+const char *lex_string = NULL;
+
+/* Support for flex reading from more than one input file (stream).
+ `include_stack' is flex's input state for each open file;
+ `file_name_stack' is the file names. `lineno_stack' is the current
+ line numbers.
+
+ If `include_stack_ptr' is 0, we haven't started reading anything yet.
+ Otherwise, stack elements 0 through `include_stack_ptr - 1' are valid. */
+
+#undef YY_INPUT
+#define YY_INPUT(buf,result,max_size) yy_input(buf, &result, max_size)
+
+#define MAX_INCLUDE_DEPTH 10
+static YY_BUFFER_STATE include_stack[MAX_INCLUDE_DEPTH];
+static const char *file_name_stack[MAX_INCLUDE_DEPTH];
+static unsigned int lineno_stack[MAX_INCLUDE_DEPTH];
+static unsigned int include_stack_ptr = 0;
+static int vers_node_nesting = 0;
+
+static YY_BUFFER_STATE yy_create_string_buffer PARAMS ((const char *string,
+ size_t size));
+static void yy_input PARAMS ((char *, int *result, int max_size));
+
+static void comment PARAMS ((void));
+static void lex_warn_invalid PARAMS ((char *where, char *what));
+
+/* STATES
+ EXPRESSION definitely in an expression
+ SCRIPT definitely in a script
+ BOTH either EXPRESSION or SCRIPT
+ DEFSYMEXP in an argument to -defsym
+ MRI in an MRI script
+ VERS_START starting a Sun style mapfile
+ VERS_SCRIPT a Sun style mapfile
+ VERS_NODE a node within a Sun style mapfile
+*/
+#define RTOKEN(x) { yylval.token = x; return x; }
-#undef input
-#undef unput
-#define input lex_input
-#define unput lex_unput
-int debug;
-
-extern boolean ldgram_in_expression;
-extern boolean ldgram_in_defsym;
-extern boolean ldgram_in_script;
-static char *command_line;
+/* Some versions of flex want this. */
+#ifndef yywrap
+int yywrap () { return 1; }
+#endif
+%}
-extern int fgetc();
-extern int yyparse();
+%a 4000
+%o 5000
-typedef struct {
- char *name;
-int value;
-} keyword_type;
-#define RTOKEN(x) { yylval.token = x; return x; }
-keyword_type keywords[] =
-{
-"MEMORY",MEMORY,
-"ORIGIN",ORIGIN,
-"BLOCK",BLOCK,
-"LENGTH",LENGTH,
-"ALIGN",ALIGN_K,
-"SUBSECTION_ALIGN",SUBSECTION_ALIGN,
-"ADDR",ADDR,
-"ENTRY",ENTRY,
-"SCRIPT", SCRIPT,
-"ENDSCRIPT", ENDSCRIPT,
-"NEXT",NEXT,
-"MAP",MAP,
-"SIZEOF",SIZEOF,
-"TARGET",TARGET_K,
-"SEARCH_DIR",SEARCH_DIR,
-"OUTPUT",OUTPUT,
-"INPUT",INPUT,
-"DEFINED",DEFINED,
-"CREATE_OBJECT_SYMBOLS",CREATE_OBJECT_SYMBOLS,
-"FORCE_COMMON_ALLOCATION",FORCE_COMMON_ALLOCATION,
-"SECTIONS",SECTIONS,
-"FILL",FILL,
-"STARTUP",STARTUP,
-"OUTPUT_FORMAT",OUTPUT_FORMAT,
-"HLL",HLL,
-"SYSLIB",SYSLIB,
-"FLOAT",FLOAT,
-"LONG", LONG,
-"SHORT", SHORT,
-"BYTE", BYTE,
-"NOFLOAT",NOFLOAT,
-"o",ORIGIN,
-"org",ORIGIN,
-"l", LENGTH,
-"len", LENGTH,
-0,0};
-unsigned int lineno;
-extern boolean hex_mode;
-FILE *ldlex_input_stack;
-static unsigned int have_pushback;
-#define NPUSHBACK 10
-int pushback[NPUSHBACK];
-int thischar;
-extern char *ldfile_input_filename;
-int donehash = 0;
-int
-lex_input()
-{
- if (have_pushback > 0)
- {
- have_pushback --;
- return thischar = pushback[have_pushback];
- }
- if (ldlex_input_stack) {
- thischar = fgetc(ldlex_input_stack);
-
- if (thischar == EOF) {
- fclose(ldlex_input_stack);
- ldlex_input_stack = (FILE *)NULL;
- ldfile_input_filename = (char *)NULL;
- /* First char after script eof is a @ so that we can tell the grammer
- that we've eft */
- thischar = '@';
+CMDFILENAMECHAR [_a-zA-Z0-9\/\.\\_\+\$\:\[\]\\\,\=\&\!\<\>\-\~]
+CMDFILENAMECHAR1 [_a-zA-Z0-9\/\.\\_\+\$\:\[\]\\\,\=\&\!\<\>\~]
+FILENAMECHAR1 [_a-zA-Z\/\.\\\$\_\~]
+SYMBOLCHARN [_a-zA-Z\/\.\\\$\_\~0-9]
+FILENAMECHAR [_a-zA-Z0-9\/\.\-\_\+\=\$\:\[\]\\\,\~]
+WILDCHAR [_a-zA-Z0-9\/\.\-\_\+\=\$\:\[\]\\\,\~\?\*]
+WHITE [ \t\n\r]+
+
+NOCFILENAMECHAR [_a-zA-Z0-9\/\.\-\_\+\$\:\[\]\\\~]
+
+V_TAG [.$_a-zA-Z][._a-zA-Z0-9]*
+V_IDENTIFIER [*?.$_a-zA-Z]([*?.$_a-zA-Z0-9]|::)*
+
+%s SCRIPT
+%s EXPRESSION
+%s BOTH
+%s DEFSYMEXP
+%s MRI
+%s VERS_START
+%s VERS_SCRIPT
+%s VERS_NODE
+%%
+ if (parser_input != input_selected)
+ {
+ /* The first token of the input determines the initial parser state. */
+ input_type t = parser_input;
+ parser_input = input_selected;
+ switch (t)
+ {
+ case input_script: return INPUT_SCRIPT; break;
+ case input_mri_script: return INPUT_MRI_SCRIPT; break;
+ case input_version_script: return INPUT_VERSION_SCRIPT; break;
+ case input_defsym: return INPUT_DEFSYM; break;
+ default: abort ();
+ }
}
- }
- else if (command_line && *command_line) {
- thischar = *(command_line++);
- }
- else {
- thischar = 0;
- }
- if(thischar == '\t') thischar = ' ';
- if (thischar == '\n') { thischar = ' '; lineno++; }
- return thischar ;
-}
-void
-lex_unput(c)
-int c;
-{
- if (have_pushback > NPUSHBACK) {
- info("%F%P Too many pushbacks\n");
- }
+<BOTH,SCRIPT,EXPRESSION>"/*" { comment(); }
+
+
+<DEFSYMEXP>"-" { RTOKEN('-');}
+<DEFSYMEXP>"+" { RTOKEN('+');}
+<DEFSYMEXP>{FILENAMECHAR1}{SYMBOLCHARN}* { yylval.name = xstrdup(yytext); return NAME; }
+<DEFSYMEXP>"=" { RTOKEN('='); }
+
+<MRI,EXPRESSION>"$"([0-9A-Fa-f])+ {
+ yylval.integer = bfd_scan_vma (yytext+1, 0,16);
+ return INT;
+ }
+
+<MRI,EXPRESSION>([0-9A-Fa-f])+(H|h|X|x|B|b|O|o|D|d) {
+ int ibase ;
+ switch (yytext[yyleng-1]) {
+ case 'X':
+ case 'x':
+ case 'H':
+ case 'h':
+ ibase = 16;
+ break;
+ case 'O':
+ case 'o':
+ ibase = 8;
+ break;
+ case 'B':
+ case 'b':
+ ibase = 2;
+ break;
+ default:
+ ibase = 10;
+ }
+ yylval.integer = bfd_scan_vma (yytext, 0,
+ ibase);
+ return INT;
+ }
+<SCRIPT,DEFSYMEXP,MRI,BOTH,EXPRESSION>((("$"|"0x")([0-9A-Fa-f])+)|(([0-9])+))(M|K|m|k)? {
+ char *s = yytext;
+
+ if (*s == '$')
+ ++s;
+ yylval.integer = bfd_scan_vma (s, 0, 0);
+ if (yytext[yyleng-1] == 'M'
+ || yytext[yyleng-1] == 'm')
+ yylval.integer *= 1024 * 1024;
+ if (yytext[yyleng-1] == 'K'
+ || yytext[yyleng-1]=='k')
+ yylval.integer *= 1024;
+ return INT;
+ }
+<BOTH,SCRIPT,EXPRESSION,MRI>"]" { RTOKEN(']');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"[" { RTOKEN('[');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"<<=" { RTOKEN(LSHIFTEQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>">>=" { RTOKEN(RSHIFTEQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"||" { RTOKEN(OROR);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"==" { RTOKEN(EQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"!=" { RTOKEN(NE);}
+<BOTH,SCRIPT,EXPRESSION,MRI>">=" { RTOKEN(GE);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"<=" { RTOKEN(LE);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"<<" { RTOKEN(LSHIFT);}
+<BOTH,SCRIPT,EXPRESSION,MRI>">>" { RTOKEN(RSHIFT);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"+=" { RTOKEN(PLUSEQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"-=" { RTOKEN(MINUSEQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"*=" { RTOKEN(MULTEQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"/=" { RTOKEN(DIVEQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"&=" { RTOKEN(ANDEQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"|=" { RTOKEN(OREQ);}
+<BOTH,SCRIPT,EXPRESSION,MRI>"&&" { RTOKEN(ANDAND);}
+<BOTH,SCRIPT,EXPRESSION,MRI>">" { RTOKEN('>');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"," { RTOKEN(',');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"&" { RTOKEN('&');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"|" { RTOKEN('|');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"~" { RTOKEN('~');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"!" { RTOKEN('!');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"?" { RTOKEN('?');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"*" { RTOKEN('*');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"+" { RTOKEN('+');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"-" { RTOKEN('-');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"/" { RTOKEN('/');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"%" { RTOKEN('%');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"<" { RTOKEN('<');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"=" { RTOKEN('=');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"}" { RTOKEN('}') ; }
+<BOTH,SCRIPT,EXPRESSION,MRI>"{" { RTOKEN('{'); }
+<BOTH,SCRIPT,EXPRESSION,MRI>")" { RTOKEN(')');}
+<BOTH,SCRIPT,EXPRESSION,MRI>"(" { RTOKEN('(');}
+<BOTH,SCRIPT,EXPRESSION,MRI>":" { RTOKEN(':'); }
+<BOTH,SCRIPT,EXPRESSION,MRI>";" { RTOKEN(';');}
+<BOTH,SCRIPT>"MEMORY" { RTOKEN(MEMORY);}
+<BOTH,SCRIPT>"ORIGIN" { RTOKEN(ORIGIN);}
+<BOTH,SCRIPT>"VERSION" { RTOKEN(VERSIONK);}
+<EXPRESSION,BOTH,SCRIPT>"BLOCK" { RTOKEN(BLOCK);}
+<EXPRESSION,BOTH,SCRIPT>"BIND" { RTOKEN(BIND);}
+<BOTH,SCRIPT>"LENGTH" { RTOKEN(LENGTH);}
+<EXPRESSION,BOTH,SCRIPT>"ALIGN" { RTOKEN(ALIGN_K);}
+<EXPRESSION,BOTH,SCRIPT>"ADDR" { RTOKEN(ADDR);}
+<EXPRESSION,BOTH,SCRIPT>"LOADADDR" { RTOKEN(LOADADDR);}
+<EXPRESSION,BOTH>"MAX" { RTOKEN(MAX_K); }
+<EXPRESSION,BOTH>"MIN" { RTOKEN(MIN_K); }
+<EXPRESSION,BOTH>"ASSERT" { RTOKEN(ASSERT_K); }
+<BOTH,SCRIPT>"ENTRY" { RTOKEN(ENTRY);}
+<BOTH,SCRIPT,MRI>"EXTERN" { RTOKEN(EXTERN);}
+<EXPRESSION,BOTH,SCRIPT>"NEXT" { RTOKEN(NEXT);}
+<EXPRESSION,BOTH,SCRIPT>"sizeof_headers" { RTOKEN(SIZEOF_HEADERS);}
+<EXPRESSION,BOTH,SCRIPT>"SIZEOF_HEADERS" { RTOKEN(SIZEOF_HEADERS);}
+<BOTH,SCRIPT>"MAP" { RTOKEN(MAP);}
+<EXPRESSION,BOTH,SCRIPT>"SIZEOF" { RTOKEN(SIZEOF);}
+<BOTH,SCRIPT>"TARGET" { RTOKEN(TARGET_K);}
+<BOTH,SCRIPT>"SEARCH_DIR" { RTOKEN(SEARCH_DIR);}
+<BOTH,SCRIPT>"OUTPUT" { RTOKEN(OUTPUT);}
+<BOTH,SCRIPT>"INPUT" { RTOKEN(INPUT);}
+<EXPRESSION,BOTH,SCRIPT>"GROUP" { RTOKEN(GROUP);}
+<EXPRESSION,BOTH,SCRIPT>"DEFINED" { RTOKEN(DEFINED);}
+<BOTH,SCRIPT>"CREATE_OBJECT_SYMBOLS" { RTOKEN(CREATE_OBJECT_SYMBOLS);}
+<BOTH,SCRIPT>"CONSTRUCTORS" { RTOKEN( CONSTRUCTORS);}
+<BOTH,SCRIPT>"FORCE_COMMON_ALLOCATION" { RTOKEN(FORCE_COMMON_ALLOCATION);}
+<BOTH,SCRIPT>"INHIBIT_COMMON_ALLOCATION" { RTOKEN(INHIBIT_COMMON_ALLOCATION);}
+<BOTH,SCRIPT>"SECTIONS" { RTOKEN(SECTIONS);}
+<BOTH,SCRIPT>"FILL" { RTOKEN(FILL);}
+<BOTH,SCRIPT>"STARTUP" { RTOKEN(STARTUP);}
+<BOTH,SCRIPT>"OUTPUT_FORMAT" { RTOKEN(OUTPUT_FORMAT);}
+<BOTH,SCRIPT>"OUTPUT_ARCH" { RTOKEN( OUTPUT_ARCH);}
+<BOTH,SCRIPT>"HLL" { RTOKEN(HLL);}
+<BOTH,SCRIPT>"SYSLIB" { RTOKEN(SYSLIB);}
+<BOTH,SCRIPT>"FLOAT" { RTOKEN(FLOAT);}
+<BOTH,SCRIPT>"QUAD" { RTOKEN( QUAD);}
+<BOTH,SCRIPT>"SQUAD" { RTOKEN( SQUAD);}
+<BOTH,SCRIPT>"LONG" { RTOKEN( LONG);}
+<BOTH,SCRIPT>"SHORT" { RTOKEN( SHORT);}
+<BOTH,SCRIPT>"BYTE" { RTOKEN( BYTE);}
+<BOTH,SCRIPT>"NOFLOAT" { RTOKEN(NOFLOAT);}
+<EXPRESSION,BOTH,SCRIPT>"NOCROSSREFS" { RTOKEN(NOCROSSREFS);}
+<BOTH,SCRIPT>"OVERLAY" { RTOKEN(OVERLAY); }
+<BOTH,SCRIPT>"SORT" { RTOKEN(SORT); }
+<EXPRESSION,BOTH,SCRIPT>"NOLOAD" { RTOKEN(NOLOAD);}
+<EXPRESSION,BOTH,SCRIPT>"DSECT" { RTOKEN(DSECT);}
+<EXPRESSION,BOTH,SCRIPT>"COPY" { RTOKEN(COPY);}
+<EXPRESSION,BOTH,SCRIPT>"INFO" { RTOKEN(INFO);}
+<EXPRESSION,BOTH,SCRIPT>"OVERLAY" { RTOKEN(OVERLAY);}
+<BOTH,SCRIPT>"o" { RTOKEN(ORIGIN);}
+<BOTH,SCRIPT>"org" { RTOKEN(ORIGIN);}
+<BOTH,SCRIPT>"l" { RTOKEN( LENGTH);}
+<BOTH,SCRIPT>"len" { RTOKEN( LENGTH);}
+<BOTH,SCRIPT>"INCLUDE" { RTOKEN(INCLUDE);}
+<BOTH,SCRIPT>"PHDRS" { RTOKEN (PHDRS); }
+<EXPRESSION,BOTH,SCRIPT>"AT" { RTOKEN(AT);}
+<EXPRESSION,BOTH,SCRIPT>"PROVIDE" { RTOKEN(PROVIDE); }
+<EXPRESSION,BOTH,SCRIPT>"KEEP" { RTOKEN(KEEP); }
+<EXPRESSION,BOTH,SCRIPT>"EXCLUDE_FILE" { RTOKEN(EXCLUDE_FILE); }
+<MRI>"#".*\n? { ++ lineno; }
+<MRI>"\n" { ++ lineno; RTOKEN(NEWLINE); }
+<MRI>"*".* { /* Mri comment line */ }
+<MRI>";".* { /* Mri comment line */ }
+<MRI>"END" { RTOKEN(ENDWORD); }
+<MRI>"ALIGNMOD" { RTOKEN(ALIGNMOD);}
+<MRI>"ALIGN" { RTOKEN(ALIGN_K);}
+<MRI>"CHIP" { RTOKEN(CHIP); }
+<MRI>"BASE" { RTOKEN(BASE); }
+<MRI>"ALIAS" { RTOKEN(ALIAS); }
+<MRI>"TRUNCATE" { RTOKEN(TRUNCATE); }
+<MRI>"LOAD" { RTOKEN(LOAD); }
+<MRI>"PUBLIC" { RTOKEN(PUBLIC); }
+<MRI>"ORDER" { RTOKEN(ORDER); }
+<MRI>"NAME" { RTOKEN(NAMEWORD); }
+<MRI>"FORMAT" { RTOKEN(FORMAT); }
+<MRI>"CASE" { RTOKEN(CASE); }
+<MRI>"START" { RTOKEN(START); }
+<MRI>"LIST".* { RTOKEN(LIST); /* LIST and ignore to end of line */ }
+<MRI>"SECT" { RTOKEN(SECT); }
+<EXPRESSION,BOTH,SCRIPT,MRI>"ABSOLUTE" { RTOKEN(ABSOLUTE); }
+<MRI>"end" { RTOKEN(ENDWORD); }
+<MRI>"alignmod" { RTOKEN(ALIGNMOD);}
+<MRI>"align" { RTOKEN(ALIGN_K);}
+<MRI>"chip" { RTOKEN(CHIP); }
+<MRI>"base" { RTOKEN(BASE); }
+<MRI>"alias" { RTOKEN(ALIAS); }
+<MRI>"truncate" { RTOKEN(TRUNCATE); }
+<MRI>"load" { RTOKEN(LOAD); }
+<MRI>"public" { RTOKEN(PUBLIC); }
+<MRI>"order" { RTOKEN(ORDER); }
+<MRI>"name" { RTOKEN(NAMEWORD); }
+<MRI>"format" { RTOKEN(FORMAT); }
+<MRI>"case" { RTOKEN(CASE); }
+<MRI>"extern" { RTOKEN(EXTERN); }
+<MRI>"start" { RTOKEN(START); }
+<MRI>"list".* { RTOKEN(LIST); /* LIST and ignore to end of line */ }
+<MRI>"sect" { RTOKEN(SECT); }
+<EXPRESSION,BOTH,SCRIPT,MRI>"absolute" { RTOKEN(ABSOLUTE); }
+
+<MRI>{FILENAMECHAR1}{NOCFILENAMECHAR}* {
+/* Filename without commas, needed to parse mri stuff */
+ yylval.name = xstrdup(yytext);
+ return NAME;
+ }
+
+
+<BOTH,EXPRESSION>{FILENAMECHAR1}{FILENAMECHAR}* {
+ yylval.name = xstrdup(yytext);
+ return NAME;
+ }
+<BOTH,EXPRESSION>"-l"{FILENAMECHAR}+ {
+ yylval.name = xstrdup (yytext + 2);
+ return LNAME;
+ }
+<SCRIPT>{WILDCHAR}* {
+ /* Annoyingly, this pattern can match comments, and we have
+ longest match issues to consider. So if the first two
+ characters are a comment opening, put the input back and
+ try again. */
+ if (yytext[0] == '/' && yytext[1] == '*')
+ {
+ yyless(2);
+ comment ();
+ }
+ else
+ {
+ yylval.name = xstrdup(yytext);
+ return NAME;
+ }
+ }
- pushback[have_pushback] = c;
- have_pushback ++;
-}
+<EXPRESSION,BOTH,SCRIPT,VERS_NODE>"\""[^\"]*"\"" {
+ /* No matter the state, quotes
+ give what's inside */
+ yylval.name = xstrdup(yytext+1);
+ yylval.name[yyleng-2] = 0;
+ return NAME;
+ }
+<BOTH,SCRIPT,EXPRESSION>"\n" { lineno++;}
+<MRI,BOTH,SCRIPT,EXPRESSION>[ \t\r]+ { }
+<VERS_NODE,VERS_SCRIPT>[:,;] { return *yytext; }
- int
-yywrap()
- { return 1; }
-/*VARARGS*/
+<VERS_NODE>global { RTOKEN(GLOBAL); }
-void
-allprint(x)
-int x;
-{
-fprintf(yyout,"%d",x);
-}
+<VERS_NODE>local { RTOKEN(LOCAL); }
-void
-sprint(x)
-char *x;
-{
-fprintf(yyout,"%s",x);
-}
+<VERS_NODE>extern { RTOKEN(EXTERN); }
-int thischar;
+<VERS_NODE>{V_IDENTIFIER} { yylval.name = xstrdup (yytext);
+ return VERS_IDENTIFIER; }
-void parse_line(arg)
-char *arg;
-{
- command_line = arg;
- have_pushback = 0;
- yyparse();
-}
+<VERS_SCRIPT>{V_TAG} { yylval.name = xstrdup (yytext);
+ return VERS_TAG; }
+<VERS_START>"{" { BEGIN(VERS_SCRIPT); return *yytext; }
+<VERS_SCRIPT>"{" { BEGIN(VERS_NODE);
+ vers_node_nesting = 0;
+ return *yytext;
+ }
+<VERS_SCRIPT>"}" { return *yytext; }
+<VERS_NODE>"{" { vers_node_nesting++; return *yytext; }
+<VERS_NODE>"}" { if (--vers_node_nesting < 0)
+ BEGIN(VERS_SCRIPT);
+ return *yytext;
+ }
-void
-parse_args(ac, av)
-int ac;
-char **av;
-{
- char *p;
- int i;
- size_t size = 0;
- char *dst;
- debug = 1;
- for (i= 1; i < ac; i++) {
- size += strlen(av[i]) + 2;
- }
- dst = p = (char *)ldmalloc(size + 2);
-/* Put a space arount each option */
+<VERS_START,VERS_NODE,VERS_SCRIPT>[\n] { lineno++; }
+<VERS_START,VERS_NODE,VERS_SCRIPT>#.* { /* Eat up comments */ }
- for (i =1; i < ac; i++) {
+<VERS_START,VERS_NODE,VERS_SCRIPT>[ \t\r]+ { /* Eat up whitespace */ }
- unsigned int s = strlen(av[i]);
- *dst++ = ' ';
- memcpy(dst, av[i], s);
- dst[s] = ' ';
- dst += s + 1;
+<<EOF>> {
+ include_stack_ptr--;
+
+ if (include_stack_ptr == 0)
+ {
+ yyterminate();
+ }
+ else
+ {
+ yy_switch_to_buffer(include_stack[include_stack_ptr]);
}
- *dst = 0;
- parse_line(p);
-
- free(p);
+ ldfile_input_filename = file_name_stack[include_stack_ptr - 1];
+ lineno = lineno_stack[include_stack_ptr];
+ return END;
}
-long number(text, base)
-char *text;
-int base;
+<SCRIPT,MRI,VERS_START,VERS_SCRIPT,VERS_NODE>. lex_warn_invalid(" in script", yytext);
+<EXPRESSION,DEFSYMEXP,BOTH>. lex_warn_invalid(" in expression", yytext);
+
+%%
+\f
+
+/* Switch flex to reading script file NAME, open on FILE,
+ saving the current input info on the include stack. */
+
+void
+lex_push_file (file, name)
+ FILE *file;
+ const char *name;
{
- unsigned long l = 0;
- char *p;
- for (p = text; *p != 0; p++) {
- if (*p == 'K') {
- l =l * 1024;
- }
- else if(*p== 'M') {
- l =l * 1024 * 1024;
+ if (include_stack_ptr >= MAX_INCLUDE_DEPTH)
+ {
+ einfo("%F:includes nested too deeply\n");
}
- else {
- l =l * base;
- if (isdigit(*p)) {
- l += *p - '0';
- }
- else if (islower(*p)) {
- l += *p - 'a' + 10;
- }
- else {
- l += *p - 'A' + 10;
- }
- }
- }
- return l;
+ file_name_stack[include_stack_ptr] = name;
+ lineno_stack[include_stack_ptr] = lineno;
+ include_stack[include_stack_ptr] = YY_CURRENT_BUFFER;
+
+ include_stack_ptr++;
+ lineno = 1;
+ yyin = file;
+ yy_switch_to_buffer(yy_create_buffer(yyin, YY_BUF_SIZE));
}
-%}
-%a 4000
-%o 5000
-FILENAMECHAR [a-zA-Z0-9\/\.\-\_\+\=]
-FILENAME {FILENAMECHAR}+
+/* Return a newly created flex input buffer containing STRING,
+ which is SIZE bytes long. */
+static YY_BUFFER_STATE
+yy_create_string_buffer (string, size)
+ CONST char *string;
+ size_t size;
+{
+ YY_BUFFER_STATE b;
+
+ /* Calls to m-alloc get turned by sed into xm-alloc. */
+ b = (YY_BUFFER_STATE) malloc (sizeof (struct yy_buffer_state));
+ b->yy_input_file = 0;
+ b->yy_buf_size = size;
+
+ /* yy_ch_buf has to be 2 characters longer than the size given because
+ we need to put in 2 end-of-buffer characters. */
+ b->yy_ch_buf = (char *) malloc ((unsigned) (b->yy_buf_size + 3));
+
+ b->yy_ch_buf[0] = '\n';
+ strcpy (b->yy_ch_buf+1, string);
+ b->yy_ch_buf[size+1] = YY_END_OF_BUFFER_CHAR;
+ b->yy_ch_buf[size+2] = YY_END_OF_BUFFER_CHAR;
+ b->yy_n_chars = size+1;
+ b->yy_buf_pos = &b->yy_ch_buf[1];
+
+ b->yy_is_our_buffer = 1;
+ b->yy_is_interactive = 0;
+ b->yy_at_bol = 1;
+ b->yy_fill_buffer = 0;
+
+ /* flex 2.4.7 changed the interface. FIXME: We should not be using
+ a flex internal interface in the first place! */
+#ifdef YY_BUFFER_NEW
+ b->yy_buffer_status = YY_BUFFER_NEW;
+#else
+ b->yy_eof_status = EOF_NOT_SEEN;
+#endif
+
+ return b;
+}
-WHITE [ \t]+
-
-%%
+/* Switch flex to reading from STRING, saving the current input info
+ on the include stack. */
-"@" { return ENDSCRIPT; }
-"\ -defsym\ " { return OPTION_defsym; }
-"\ -noinhibit_exec\ " { return OPTION_noinhibit_exec; }
-"\ -format\ " { return OPTION_format; }
-"\ -n\ " { return OPTION_n; }
-"\ -r\ " { return OPTION_r; }
-"\ -Ur\ " { return OPTION_Ur; }
-"\ -o\ " { return OPTION_o; }
-"\ -g\ " { return OPTION_g; }
-"\ -e\ " { return OPTION_e; }
-"\ -b\ " { return OPTION_b; }
-"\ -dc\ " { return OPTION_dc; }
-"\ -dp\ " { return OPTION_dp; }
-"\ -d\ " { return OPTION_d; }
-"\ -v\ " { return OPTION_v; }
-"\ -M\ " { return OPTION_M; }
-"\ -t\ " { return OPTION_t; }
-"\ -X\ " { return OPTION_X; }
-"\ -x\ " { return OPTION_x; }
-"\ -c\ " { return OPTION_c; }
-"\ -R\ " { return OPTION_R; }
-"\ -u\ " { return OPTION_u; }
-"\ -s\ " { return OPTION_s; }
-"\ -S\ " { return OPTION_S; }
-"\ -l"{FILENAME} {
- yylval.name = buystring(yytext+3);
- return OPTION_l;
- }
+void
+lex_redirect (string)
+ CONST char *string;
+{
+ YY_BUFFER_STATE tmp;
-"\ -L"{FILENAME} {
- yylval.name = buystring(yytext+3);
- return OPTION_L;
- }
-"\ -Ttext\ " {
- yylval.name = ".text";
- return OPTION_Texp;
- }
-"\ -Tdata\ " {
- yylval.name = ".data";
- return OPTION_Texp;
- }
-"\ -Tbss\ " {
- yylval.name = ".bss";
- return OPTION_Texp;
- }
-
-"\ -T"{FILENAME} {
- yylval.name = buystring(yytext+3);
- return OPTION_Tfile;
- }
-"\ -T\ " {
- return OPTION_T;
- }
-
-"\ -F"{FILENAME} {
- return OPTION_F;
- }
-"\ -F\ " {
- return OPTION_F;
- }
-
-"\ -A"{FILENAME} {
- yylval.name = buystring(yytext+3);
- return OPTION_Aarch;
- }
-" " { }
-"<<=" { RTOKEN(LSHIFTEQ);}
-">>=" { RTOKEN(RSHIFTEQ);}
-"||" { RTOKEN(OROR);}
-"==" { RTOKEN(EQ);}
-"!=" { RTOKEN(NE);}
-">=" { RTOKEN(GE);}
-"<=" { RTOKEN(LE);}
-"<<" { RTOKEN(LSHIFT);}
-">>" { RTOKEN(RSHIFT);}
-"+=" { RTOKEN(PLUSEQ);}
-"-=" { RTOKEN(MINUSEQ);}
-"*=" { RTOKEN(MULTEQ);}
-"/=" { RTOKEN(DIVEQ);}
-"&=" { RTOKEN(ANDEQ);}
-"|=" { RTOKEN(OREQ);}
-"&&" { RTOKEN(ANDAND);}
-">" { RTOKEN('>');}
-"," { RTOKEN(',');}
-"&" { RTOKEN('&');}
-"|" { RTOKEN('|');}
-"~" { RTOKEN('~');}
-"!" { RTOKEN('!');}
-"?" { RTOKEN('?');}
-"*" { RTOKEN('*');}
-"%" { RTOKEN('%');}
-"<" { RTOKEN('<');}
-">" { RTOKEN('>');}
-"}" { RTOKEN('}') ; }
-"{" { RTOKEN('{'); }
-")" { RTOKEN(')');}
-"(" { RTOKEN('(');}
-"]" { RTOKEN(']');}
-"[" { RTOKEN('[');}
-":" { RTOKEN(':'); }
-";" { RTOKEN('\;');}
-"-" { RTOKEN('-');}
-
-
-
-"/*" {
- while (1) {
- int ch;
- ch = input();
- while (ch != '*') {
- ch = input();
+ yy_init = 0;
+ if (include_stack_ptr >= MAX_INCLUDE_DEPTH)
+ {
+ einfo("%F: macros nested too deeply\n");
}
+ file_name_stack[include_stack_ptr] = "redirect";
+ lineno_stack[include_stack_ptr] = lineno;
+ include_stack[include_stack_ptr] = YY_CURRENT_BUFFER;
+ include_stack_ptr++;
+ lineno = 1;
+ tmp = yy_create_string_buffer (string, strlen (string));
+ yy_switch_to_buffer (tmp);
+}
+\f
+/* Functions to switch to a different flex start condition,
+ saving the current start condition on `state_stack'. */
+static int state_stack[MAX_INCLUDE_DEPTH * 2];
+static int *state_stack_p = state_stack;
-
- if (input() == '/') {
- break;
- }
- unput(yytext[yyleng-1]);
- }
+void
+ldlex_script ()
+{
+ *(state_stack_p)++ = yy_start;
+ BEGIN (SCRIPT);
}
-"\""[^\"]*"\"" {
+void
+ldlex_mri_script ()
+{
+ *(state_stack_p)++ = yy_start;
+ BEGIN (MRI);
+}
- yylval.name = buystring(yytext+1);
- yylval.name[yyleng-2] = 0; /* Fry final quote */
- return NAME;
+void
+ldlex_version_script ()
+{
+ *(state_stack_p)++ = yy_start;
+ BEGIN (VERS_START);
}
-[0][0-7KM]* {
- yylval.integer = number(yytext+1, 8);
- return INT;
+void
+ldlex_version_file ()
+{
+ *(state_stack_p)++ = yy_start;
+ BEGIN (VERS_SCRIPT);
}
-[0-9]+[KM]? {
- if (hex_mode == true || ldgram_in_defsym == true) {
- yylval.integer = number(yytext, 16);
- }
- else {
- yylval.integer = number(yytext, 10);
- }
- return INT;
+void
+ldlex_defsym ()
+{
+ *(state_stack_p)++ = yy_start;
+ BEGIN (DEFSYMEXP);
+}
+
+void
+ldlex_expression ()
+{
+ *(state_stack_p)++ = yy_start;
+ BEGIN (EXPRESSION);
}
-0[Xx][0-9a-fA-FKM]+ {
-
- yylval.integer = number(yytext+2,16);
- return INT;
+void
+ldlex_both ()
+{
+ *(state_stack_p)++ = yy_start;
+ BEGIN (BOTH);
}
-"\#"{WHITE}*{FILENAMECHAR}+ {
- char *p = yytext+1;
- while(*p ==' ' || *p == '\t') p++;
- yylval.name = buystring(p);
- return NAME;
+void
+ldlex_popstate ()
+{
+ yy_start = *(--state_stack_p);
}
-{FILENAMECHAR} {
-
- boolean loop = false;
- /*
- Tokenize a name, this is really pain, since a name can be a
- filename or a symbol name. filenames have slashes and stuff whist
- in an expression those things are seperate tokens. We hack this by
- setting ldlang_in_script when we are expecting a symbol, so that
- [/+-] get taken to be seperate tokens. An extra gotcha is
- expressions after defsyms, we only allow +s and -s in a defsym
- expression, so -defsym foo=bar+9 /file.o is parsed ok.
-
- The more I think about this the more I hate it. I've got a problem
- now with the = sign, what should I do ? imagine:
- __start=.;
- You'd think that was pretty unambiguous wouldn't you. Well it's
- not since __start=. is (at the moment) a perfectly valid
- filename. And in some cases we don't know what's going on. I'm
- going to have to hack this. If we see a '/' before the = sign then
- we say we've got an = in a filename, otherwise it's an operator.
- (later)
- That's it, I've had enough. From now on, an =s on a command line
- will be taken to be part of a file name unless its in a defsym,
- and an = in a file will be taken to be an operator.
- */
- int ch;
- keyword_type *k;
-
- if (hex_mode) {
- ch = yytext[0];
- /* Then always read a number */
- while (isxdigit(ch)) {
- yytext[yyleng++] = ch;
- ch = input();
- }
- yytext[yyleng] = 0;
- unput(ch);
+\f
- yylval.integer = number(yytext,16);
- return INT;
- }
+/* Place up to MAX_SIZE characters in BUF and return in *RESULT
+ either the number of characters read, or 0 to indicate EOF. */
- if (ldfile_input_filename) {
- /* We're inside a file */
- if (yytext[0]== '=') {
- RTOKEN('=');
+static void
+yy_input (buf, result, max_size)
+ char *buf;
+ int *result;
+ int max_size;
+{
+ *result = 0;
+ if (yy_current_buffer->yy_input_file)
+ {
+ if (yyin)
+ {
+ *result = fread ((char *) buf, 1, max_size, yyin);
+ if (*result < 0)
+ einfo ("%F%P: read in flex scanner failed\n");
+ }
}
- }
+}
-
+/* Eat the rest of a C-style comment. */
- /* Otherwise we only notice special things if were in an
- expression */
-
- if (ldgram_in_expression) {
- if (yytext[0] != '/' || ldgram_in_defsym == false) {
- switch (yytext[0]) {
- case '/': RTOKEN('/');
- case '=': RTOKEN('=');
- case '+': RTOKEN('+');
- case '-': RTOKEN('-');
- }
+static void
+comment ()
+{
+ int c;
+
+ while (1)
+ {
+ c = input();
+ while (c != '*' && c != EOF)
+ {
+ if (c == '\n')
+ lineno++;
+ c = input();
}
- }
- ch = input();
- while (true)
- {
- if (isalpha(ch) || isdigit(ch) || ch == '.' || ch == '_' ) {
- yytext[yyleng++] = ch;
- }
-else if (ch == '=' && ldgram_in_script) {
-/* An = within a script is always taken to be an operator */
-break;
-}
- else if (ch == '+' || ch == '-' || ch == '/' || ch == '=') {
- if (ldgram_in_expression) break;
- yytext[yyleng++] = ch;
- }
- else
- break;
- ch = input();
- }
-
- yytext[yyleng] = 0;
- unput(ch);
- /* Filenames of just =signs are tokens */
- if (yyleng == 1 && yytext[0] == '=') {
- RTOKEN('=');
- }
- for(k = keywords; k ->name != (char *)NULL; k++) {
+ if (c == '*')
+ {
+ c = input();
+ while (c == '*')
+ c = input();
+ if (c == '/')
+ break; /* found the end */
+ }
+
+ if (c == '\n')
+ lineno++;
- if (strcmp(k->name, yytext)==0) {
- yylval.token = k->value;
- return k->value;
+ if (c == EOF)
+ {
+ einfo( "%F%P: EOF in comment\n");
+ break;
}
}
- yylval.name = buystring(yytext);
- return NAME;
}
+/* Warn the user about a garbage character WHAT in the input
+ in context WHERE. */
+static void
+lex_warn_invalid (where, what)
+ char *where, *what;
+{
+ char buf[5];
+
+ /* If we have found an input file whose format we do not recognize,
+ and we are therefore treating it as a linker script, and we find
+ an invalid character, then most likely this is a real object file
+ of some different format. Treat it as such. */
+ if (ldfile_assumed_script)
+ {
+ bfd_set_error (bfd_error_file_not_recognized);
+ einfo ("%F%s: file not recognized: %E\n", ldfile_input_filename);
+ }
+ if (! ISPRINT (*what))
+ {
+ sprintf (buf, "\\%03o", (unsigned int) *what);
+ what = buf;
+ }
-
-%%
+ einfo ("%P:%S: ignoring invalid character `%s'%s\n", what, where);
+}