projects / unix-history / blob
? search:
summary | tags | clone url | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
BSD 4_3_Net_2 release
[unix-history] / usr / src / usr.bin / pascal / pc3 / pc3.c
/*-
* Copyright (c) 1980, 1982, 1983 The Regents of the University of California.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#ifndef lint
char copyright[] =
"@(#) Copyright (c) 1980, 1982, 1983 The Regents of the University of California.\n\
All rights reserved.\n";
#endif /* not lint */
#ifndef lint
static char sccsid[] = "@(#)pc3.c 5.2 (Berkeley) 4/16/91";
#endif /* not lint */
/*
* Pc3 is a pass in the Berkeley Pascal compilation
* process that is performed just prior to linking Pascal
* object files. Its purpose is to enforce the rules of
* separate compilation for Berkeley Pascal. Pc3 is called
* with the same argument list of object files that is sent to
* the loader. These checks are performed by pc3 by examining
* the symbol tables of the object files:
* (1) All .o files must be up to date with respect to the
* runtime libraries.
* (2) Each global Pascal symbol (label, constant, type,
* variable, procedure, or function name) must be uniquely
* declared, i.e. declared in only one included file or
* source file.
* (3) Each external function (or procedure) may be resolved
* at most once in a source file which included the
* external declaration of the function.
*
* The symbol table of each object file is scanned and
* each global Pascal symbol is placed in a hashed symbol
* table. The Pascal compiler has been modified to emit all
* Pascal global symbols to the object file symbol table. The
* information stored in the symbol table for each such symbol
* is:
*
* - the name of the symbol;
* - a subtype descriptor;
* - the file which logically contains the declaration of
* the symbol or which caused the inclusion of an include file.
* - for included files:
* - a checksum;
* - for symbols:
* - the file which textually contains the declaration of
* the symbol (possibly an include file);
* - the line number at which the symbol is declared;
* - the file which contains the resolution of the symbol.
* - the line number at which the symbol is resolved;
*
* If a symbol has been previously entered into the symbol
* table, a check is made that the current declaration is of
* the same type and from the same include file as the previous
* one. Except for files and functions and procedures, it is
* an error for a symbol declaration to be encountered more
* than once, unless the re-declarations come from the same
* included file as the original.
*
* As an include file symbol is encountered in a source
* file, the symbol table entry of each symbol declared in that
* include file is modified to reflect its new logical
* inclusion in the source file. File symbols are also
* encountered as an included file ends, signaling the
* continuation of the enclosing file.
*
* Functions and procedures which have been declared
* external may be resolved by declarations from source files
* which included the external declaration of the function.
* Functions and procedures may be resolved at most once across
* a set of object files. The loader will complain if a
* function is not resolved at least once.
*/
\f
char program[] = "pc";
#include <sys/types.h>
#include <sys/stat.h>
#include <ar.h>
#include <stdio.h>
#include <ctype.h>
#include <a.out.h>
#include <stab.h>
#include "pstab.h"
#include "pc3.h"
int errors = NONE;
BOOL wflag = FALSE;
/*
* check each of the argument .o files (or archives of .o files).
*/
main( argc , argv )
int argc;
char **argv;
{
struct fileinfo ofile;
for ( argv++ ; *argv != 0 && **argv == '-' ; argv++ ) {
(*argv)++;
switch ( **argv ) {
default:
error( FATAL , "pc3: bad flag -%c\n" , **argv );
case 'w':
wflag = TRUE;
break;
}
}
for ( /* void */ ; *argv != 0 ; argv++ ) {
# ifdef DEBUG
fprintf( stderr , "[main] *argv = %s\n" , *argv );
# endif DEBUG
ofile.name = *argv;
checkfile( &ofile );
}
exit( errors );
}
/*
* check the namelist of a file, or all namelists of an archive.
*/
checkfile( ofilep )
struct fileinfo *ofilep;
{
union {
char mag_armag[ SARMAG + 1 ];
struct exec mag_exec;
} mag_un;
int red;
struct stat filestat;
ofilep -> file = fopen( ofilep -> name , "r" );
if ( ofilep -> file == NULL ) {
error( ERROR , "cannot open: %s" , ofilep -> name );
return;
}
fstat( fileno( ofilep -> file ) , &filestat );
red = fread( (char *) &mag_un , 1 , sizeof mag_un , ofilep -> file );
if ( red != sizeof mag_un ) {
error( ERROR , "cannot read header: %s" , ofilep -> name );
return;
}
if ( mag_un.mag_exec.a_magic == OARMAG ) {
error( WARNING , "old archive: %s" , ofilep -> name );
return;
}
if ( strncmp( mag_un.mag_armag , ARMAG , SARMAG ) == 0 ) {
/* archive, iterate through elements */
# ifdef DEBUG
fprintf( stderr , "[checkfile] archive %s\n" , ofilep -> name );
# endif DEBUG
ofilep -> nextoffset = SARMAG;
while ( nextelement( ofilep ) ) {
checknl( ofilep );
}
} else if ( N_BADMAG( mag_un.mag_exec ) ) {
/* not a file.o */
error( ERROR , "bad format: %s" , ofilep -> name );
return;
} else {
/* a file.o */
# ifdef DEBUG
fprintf( stderr , "[checkfile] .o file %s\n" , ofilep -> name );
# endif DEBUG
fseek( ofilep -> file , 0L , 0 );
ofilep -> nextoffset = filestat.st_size;
checknl( ofilep );
}
fclose( ofilep -> file );
}
/*
* check the namelist of this file for conflicts with
* previously entered symbols.
*/
checknl( ofilep )
register struct fileinfo *ofilep;
{
long red;
struct exec oexec;
off_t symoff;
long numsyms;
register struct nlist *nlp;
register char *stringp;
long strsize;
long sym;
red = fread( (char *) &oexec , 1 , sizeof oexec , ofilep -> file );
if ( red != sizeof oexec ) {
error( ERROR , "error reading struct exec: %s"
, ofilep -> name );
return;
}
if ( N_BADMAG( oexec ) ) {
return;
}
symoff = N_SYMOFF( oexec ) - sizeof oexec;
fseek( ofilep -> file , symoff , 1 );
numsyms = oexec.a_syms / sizeof ( struct nlist );
if ( numsyms == 0 ) {
error( WARNING , "no name list: %s" , ofilep -> name );
return;
}
nlp = (struct nlist *) calloc( numsyms , sizeof ( struct nlist ) );
if ( nlp == 0 ) {
error( FATAL , "no room for %d nlists" , numsyms );
}
red = fread( ( char * ) nlp , numsyms , sizeof ( struct nlist )
, ofilep -> file );
if ( ftell( ofilep -> file ) + sizeof ( off_t )
>= ofilep -> nextoffset ) {
error( WARNING , "no string table (old format .o?)"
, ofilep -> name );
return;
}
red = fread( (char *) &strsize , sizeof strsize , 1
, ofilep -> file );
if ( red != 1 ) {
error( WARNING , "no string table (old format .o?)"
, ofilep -> name );
return;
}
stringp = ( char * ) malloc( strsize );
if ( stringp == 0 ) {
error( FATAL , "no room for %d bytes of strings" , strsize );
}
red = fread( stringp + sizeof strsize
, strsize - sizeof ( strsize ) , 1 , ofilep -> file );
if ( red != 1 ) {
error( WARNING , "error reading string table: %s"
, ofilep -> name );
}
# ifdef DEBUG
fprintf( stderr , "[checknl] %s: %d symbols\n"
, ofilep -> name , numsyms );
# endif DEBUG
for ( sym = 0 ; sym < numsyms ; sym++) {
if ( nlp[ sym ].n_un.n_strx ) {
nlp[ sym ].n_un.n_name = stringp + nlp[ sym ].n_un.n_strx;
} else {
nlp[ sym ].n_un.n_name = "";
}
checksymbol( &nlp[ sym ] , ofilep );
}
if ( nlp ) {
free( nlp );
}
if ( stringp ) {
free( stringp );
}
}
/*
* check a symbol.
* look it up in the hashed symbol table,
* entering it if necessary.
* this maintains a state of which .p and .i files
* it is currently in the midst from the nlist entries
* for source and included files.
* if we are inside a .p but not a .i, pfilep == ifilep.
*/
checksymbol( nlp , ofilep )
struct nlist *nlp;
struct fileinfo *ofilep;
{
static struct symbol *pfilep = NIL;
static struct symbol *ifilep = NIL;
register struct symbol *symbolp;
int errtype;
# ifdef DEBUG
if ( pfilep && ifilep ) {
fprintf( stderr , "[checksymbol] pfile %s ifile %s\n"
, pfilep -> name , ifilep -> name );
}
fprintf( stderr , "[checksymbol] ->name %s ->n_desc %x (%s)\n"
, nlp -> n_un.n_name , nlp -> n_desc
, classify( nlp -> n_desc ) );
# endif DEBUG
if ( nlp -> n_type != N_PC ) {
/* don't care about the others */
return;
}
symbolp = entersymbol( nlp -> n_un.n_name );
if ( symbolp -> lookup == NEW ) {
# ifdef DEBUG
fprintf( stderr , "[checksymbol] ->name %s is NEW\n"
, symbolp -> name );
# endif DEBUG
symbolp -> desc = nlp -> n_desc;
symbolp -> fromp = pfilep;
switch ( symbolp -> desc ) {
default:
error( FATAL , "panic: [checksymbol] NEW" );
case N_PGLABEL:
case N_PGCONST:
case N_PGTYPE:
case N_PGVAR:
case N_PGFUNC:
case N_PGPROC:
case N_PLDATA:
case N_PLTEXT:
symbolp -> sym_un.sym_str.rfilep = ifilep;
symbolp -> sym_un.sym_str.rline = nlp -> n_value;
symbolp -> sym_un.sym_str.fromi = ifilep;
symbolp -> sym_un.sym_str.iline = nlp -> n_value;
return;
case N_PEFUNC:
case N_PEPROC:
symbolp -> sym_un.sym_str.rfilep = NIL;
symbolp -> sym_un.sym_str.rline = 0;
/*
* functions can only be declared external
* in included files.
*/
if ( pfilep == ifilep ) {
error( WARNING
, "%s, line %d: %s %s must be declared in included file"
, pfilep -> name , nlp -> n_value
, classify( symbolp -> desc )
, symbolp -> name );
}
symbolp -> sym_un.sym_str.fromi = ifilep;
symbolp -> sym_un.sym_str.iline = nlp -> n_value;
return;
case N_PSO:
if ( nlp -> n_value < N_FLAGCHECKSUM ) {
error( WARNING,
"%s is out of date and should be recompiled",
ofilep -> name );
}
pfilep = symbolp;
ifilep = symbolp;
symbolp -> sym_un.checksum = N_FLAGCHECKSUM;
return;
case N_PSOL:
ifilep = symbolp;
symbolp -> sym_un.checksum = nlp -> n_value;
return;
}
} else {
# ifdef DEBUG
fprintf( stderr , "[checksymbol] ->name %s is OLD\n"
, symbolp -> name );
# endif DEBUG
errtype = ERROR;
switch ( symbolp -> desc ) {
default:
error( FATAL , "panic [checksymbol] OLD" );
return;
case N_PSO:
/*
* finding a file again means you are back
* in it after finishing an include file.
*/
if ( symbolp -> desc != nlp -> n_desc ) {
error( FATAL , "panic [checksymbol] PSO" );
return;
}
pfilep = symbolp;
ifilep = symbolp;
return;
case N_PSOL:
/*
* include files can be seen more than once,
* but their checksums are checked if they are
* greater than N_FLAGCHECKSUM.
* PSOL's are seen with checksums as the
* include file is entered, and with
* N_FLAGCHECKSUM as we are back in an
* included file from a nested include.
*/
if ( symbolp -> desc != nlp -> n_desc ) {
error( FATAL , "panic [checksymbol] PSOL" );
return;
}
if ((unsigned) symbolp->sym_un.checksum > N_FLAGCHECKSUM
&& (unsigned) nlp -> n_value > N_FLAGCHECKSUM
&& symbolp -> sym_un.checksum != nlp -> n_value ) {
error( ERROR,
"%s included in %s differs from %s included in %s",
symbolp -> name, pfilep -> name,
symbolp -> name, symbolp -> fromp -> name );
}
ifilep = symbolp;
return;
case N_PEFUNC:
case N_PEPROC:
/*
* this might be the resolution of the external
* has to match func/proc of external
* and has to have included external
* and has to not have been previously resolved.
*/
if ( ( ( symbolp -> desc == N_PEFUNC
&& nlp -> n_desc == N_PGFUNC )
|| ( symbolp -> desc == N_PEPROC
&& nlp -> n_desc == N_PGPROC ) )
&& ( symbolp -> fromp == pfilep )
&& ( symbolp -> sym_un.sym_str.rfilep == NIL ) ) {
/*
* resolve external
*/
# ifdef DEBUG
fprintf( stderr , "[checksymbol] resolving external\n" );
# endif DEBUG
symbolp -> sym_un.sym_str.rfilep = ifilep;
symbolp -> sym_un.sym_str.rline = nlp -> n_value;
return;
}
/*
* otherwise, it might be another external,
* which is okay if it's
* the same type and from the same include file
*/
if ( ( ( symbolp -> desc == N_PEFUNC
&& nlp -> n_desc == N_PEFUNC )
|| ( symbolp -> desc == N_PEPROC
&& nlp -> n_desc == N_PEPROC ) )
&& ( symbolp -> sym_un.sym_str.fromi == ifilep ) ) {
/*
* just another pretty external
* make it look like it comes from here.
*/
# ifdef DEBUG
fprintf( stderr , "[checksymbol] just another pretty external\n" );
# endif DEBUG
symbolp -> fromp = pfilep;
return;
}
/*
* something is wrong
* if it's not resolved, use the header file
* otherwise, it's just a regular error
*/
if ( symbolp -> sym_un.sym_str.rfilep == NIL ) {
error( ERROR ,
"%s, line %d: %s is already defined\n\t(%s, line %d)." ,
ifilep -> name , nlp -> n_value ,
nlp -> n_un.n_name ,
symbolp -> sym_un.sym_str.fromi -> name ,
symbolp -> sym_un.sym_str.iline );
return;
}
break;
case N_PGFUNC:
case N_PGPROC:
/*
* functions may not be seen more than once.
* the loader will complain about
* `multiply defined', but we can, too.
*/
break;
case N_PGLABEL:
case N_PGCONST:
case N_PGTYPE:
case N_PGVAR:
/*
* labels, constants, types, variables
* and external declarations
* may be seen as many times as they want,
* as long as they come from the same include file.
* make it look like they come from this .p file.
*/
included:
if ( nlp -> n_desc != symbolp -> desc
|| symbolp -> sym_un.sym_str.fromi != ifilep ) {
break;
}
symbolp -> fromp = pfilep;
return;
case N_PLDATA:
case N_PLTEXT:
switch ( nlp -> n_desc ) {
default:
error( FATAL , "pc3: unknown stab 0x%x"
, nlp -> n_desc );
return;
case N_PSO:
case N_PSOL:
case N_PGCONST:
case N_PGTYPE:
/* these won't conflict with library */
return;
case N_PGLABEL:
case N_PGVAR:
case N_PGFUNC:
case N_PGPROC:
case N_PEFUNC:
case N_PEPROC:
case N_PLDATA:
case N_PLTEXT:
errtype = WARNING;
break;
}
break;
}
/*
* this is the breaks
*/
error( errtype
, "%s, line %d: %s %s is already defined\n\t%s%s (%s, line %d)."
, ifilep -> name
, nlp -> n_value
, classify( nlp -> n_desc )
, nlp -> n_un.n_name
, ( symbolp -> desc == nlp -> n_desc ? "" : " as " )
, ( symbolp -> desc == nlp -> n_desc
? "" : article( symbolp -> desc ) )
, symbolp -> sym_un.sym_str.rfilep -> name
, symbolp -> sym_un.sym_str.rline );
}
}
/*
* quadratically hashed symbol table.
* things are never deleted from the hash symbol table.
* as more hash table is needed,
* a new one is alloc'ed and chained to the end.
* search is by rehashing within each table,
* traversing chains to next table if unsuccessful.
*/
struct symbol *
entersymbol( name )
char *name;
{
static struct symboltableinfo symboltable;
char *enteredname;
long hashindex;
register struct symboltableinfo *tablep;
register struct symbol **herep;
register struct symbol **limitp;
register long increment;
enteredname = enterstring( name );
hashindex = SHORT_ABS( ( long ) enteredname ) % SYMBOLPRIME;
for ( tablep = &symboltable ; /*return*/ ; tablep = tablep -> chain ) {
if ( tablep == NIL ) {
# ifdef SPACEDEBUG
fprintf( stderr ,
"[entersymbol] calloc'ing table for %d symbols\n" ,
SYMBOLPRIME );
# endif SPACEDEBUG
for ( tablep = &symboltable
; tablep->chain != NIL
; tablep = tablep->chain ) {
continue;
}
tablep->chain = ( struct symboltableinfo * )
calloc( 1 , sizeof ( struct symboltableinfo ) );
if ( tablep->chain == NIL ) {
error( FATAL , "ran out of memory (entersymbol)" );
}
tablep = tablep->chain;
}
herep = &( tablep -> entry[ hashindex ] );
limitp = &( tablep -> entry[ SYMBOLPRIME ] );
increment = 1;
do {
if ( *herep == NIL ) {
/* empty */
if ( tablep -> used > ( ( SYMBOLPRIME / 4 ) * 3 ) ) {
/* too full, break for next table */
break;
}
tablep -> used++;
*herep = symbolalloc();
( *herep ) -> name = enteredname;
( *herep ) -> lookup = NEW;
# ifdef HASHDEBUG
fprintf( stderr ,
"[entersymbol] name %s NEW after %d\n" ,
enteredname , increment / 2 );
# endif HASHDEBUG
return *herep;
}
/* a find? */
if ( ( *herep ) -> name == enteredname ) {
( *herep ) -> lookup = OLD;
# ifdef HASHDEBUG
fprintf( stderr , "[entersymbol] name %s OLD at %d\n" ,
enteredname , increment / 2 );
# endif HASHDEBUG
return *herep;
}
herep += increment;
if ( herep >= limitp ) {
herep -= SYMBOLPRIME;
}
increment += 2;
} while ( increment < SYMBOLPRIME );
# ifdef HASHDEBUG
fprintf( stderr , "[entersymbol] next symboltable\n" );
# endif HASHDEBUG
}
}
/*
* allocate a symbol from the dynamically allocated symbol table.
*/
struct symbol *
symbolalloc()
{
static struct symbol *nextsymbol = NIL;
static long symbolsleft = 0;
struct symbol *newsymbol;
if ( symbolsleft <= 0 ) {
# ifdef SPACEDEBUG
fprintf( stderr ,
"[symbolalloc] malloc space for %d symbols\n" ,
SYMBOLALLOC / sizeof( struct symbol ) );
# endif SPACEDEBUG
nextsymbol = ( struct symbol * ) malloc( SYMBOLALLOC );
if ( nextsymbol == 0 ) {
error( FATAL , "ran out of memory (symbolalloc)" );
}
symbolsleft = SYMBOLALLOC / sizeof( struct symbol );
}
newsymbol = nextsymbol;
nextsymbol++;
symbolsleft--;
return newsymbol;
}
/*
* hash a string based on all of its characters.
*/
long
hashstring( string )
char *string;
{
register char *cp;
register long value;
value = 0;
for ( cp = string ; *cp ; cp++ ) {
value = ( value * 2 ) + *cp;
}
return value;
}
/*
* quadratically hashed string table.
* things are never deleted from the hash string table.
* as more hash table is needed,
* a new one is alloc'ed and chained to the end.
* search is by rehashing within each table,
* traversing chains to next table if unsuccessful.
*/
char *
enterstring( string )
char *string;
{
static struct stringtableinfo stringtable;
long hashindex;
register struct stringtableinfo *tablep;
register char **herep;
register char **limitp;
register long increment;
hashindex = SHORT_ABS( hashstring( string ) ) % STRINGPRIME;
for ( tablep = &stringtable ; /*return*/ ; tablep = tablep -> chain ) {
if ( tablep == NIL ) {
# ifdef SPACEDEBUG
fprintf( stderr ,
"[enterstring] calloc space for %d strings\n" ,
STRINGPRIME );
# endif SPACEDEBUG
for ( tablep = &stringtable
; tablep->chain != NIL
; tablep = tablep->chain ) {
continue;
}
tablep->chain = ( struct stringtableinfo * )
calloc( 1 , sizeof ( struct stringtableinfo ) );
if ( tablep->chain == NIL ) {
error( FATAL , "ran out of memory (enterstring)" );
}
tablep = tablep->chain;
}
herep = &( tablep -> entry[ hashindex ] );
limitp = &( tablep -> entry[ STRINGPRIME ] );
increment = 1;
do {
if ( *herep == NIL ) {
/* empty */
if ( tablep -> used > ( ( STRINGPRIME / 4 ) * 3 ) ) {
/* too full, break for next table */
break;
}
tablep -> used++;
*herep = charalloc( strlen( string ) );
strcpy( *herep , string );
# ifdef HASHDEBUG
fprintf( stderr ,
"[enterstring] string %s copied after %d\n" ,
*herep , increment / 2 );
# endif HASHDEBUG
return *herep;
}
/* quick, check the first chars and then the rest */
if ( **herep == *string && strcmp( *herep , string ) == 0 ) {
# ifdef HASHDEBUG
fprintf( stderr ,
"[enterstring] string %s found after %d\n" ,
*herep , increment / 2 );
# endif HASHDEBUG
return *herep;
}
herep += increment;
if ( herep >= limitp ) {
herep -= STRINGPRIME;
}
increment += 2;
} while ( increment < STRINGPRIME );
# ifdef HASHDEBUG
fprintf( stderr , "[enterstring] next stringtable\n" );
# endif HASHDEBUG
}
}
/*
* copy a string to the dynamically allocated character table.
*/
char *
charalloc( length )
register long length;
{
static char *nextchar = NIL;
static long charsleft = 0;
register long lengthplus1 = length + 1;
register long askfor;
char *newstring;
if ( charsleft < lengthplus1 ) {
askfor = lengthplus1 > CHARALLOC ? lengthplus1 : CHARALLOC;
# ifdef SPACEDEBUG
fprintf( stderr , "[charalloc] malloc space for %d chars\n"
, askfor );
# endif SPACEDEBUG
nextchar = ( char * ) malloc( askfor );
if ( nextchar == 0 ) {
error( FATAL , "no room for %d characters" , askfor );
}
charsleft = askfor;
}
newstring = nextchar;
nextchar += lengthplus1;
charsleft -= lengthplus1;
return newstring;
}
/*
* read an archive header for the next element
* and find the offset of the one after this.
*/
BOOL
nextelement( ofilep )
struct fileinfo *ofilep;
{
register char *cp;
register long red;
register off_t arsize;
struct ar_hdr archdr;
fseek( ofilep -> file , ofilep -> nextoffset , 0 );
red = fread( (char *) &archdr , 1 , sizeof archdr , ofilep -> file );
if ( red != sizeof archdr ) {
return FALSE;
}
/* null terminate the blank-padded name */
cp = &archdr.ar_name[ ( sizeof archdr.ar_name ) - 1 ];
*cp = '\0';
while ( *--cp == ' ' ) {
*cp = '\0';
}
/* set up the address of the beginning of next element */
arsize = atol( archdr.ar_size );
/* archive elements are aligned on 0 mod 2 boundaries */
if ( arsize & 1 ) {
arsize += 1;
}
ofilep -> nextoffset = ftell( ofilep -> file ) + arsize;
/* say we had one */
return TRUE;
}
/*
* variable number of arguments to error, like printf.
*/
error( type , message , arg1 , arg2 , arg3 , arg4 , arg5 , arg6 , arg7 , arg8 )
int type;
char *message;
{
errors = type > errors ? type : errors;
if ( wflag && type == WARNING ) {
return;
}
fprintf( stderr , "%s: " , program );
switch ( type ) {
case WARNING:
fprintf( stderr , "Warning: " );
break;
case ERROR:
fprintf( stderr , "Error: " );
break;
case FATAL:
fprintf( stderr , "Fatal: " );
break;
default:
fprintf( stderr , "Ooops: " );
break;
}
fprintf( stderr , message , arg1,arg2,arg3,arg4,arg5,arg6,arg7,arg8 );
fprintf( stderr , "\n" );
if ( type == FATAL ) {
exit( FATAL );
}
}
char *
classify( type )
unsigned char type;
{
switch ( type ) {
case N_PSO:
return "source file";
case N_PSOL:
return "include file";
case N_PGLABEL:
return "label";
case N_PGCONST:
return "constant";
case N_PGTYPE:
return "type";
case N_PGVAR:
return "variable";
case N_PGFUNC:
return "function";
case N_PGPROC:
return "procedure";
case N_PEFUNC:
return "external function";
case N_PEPROC:
return "external procedure";
case N_PLDATA:
return "library variable";
case N_PLTEXT:
return "library routine";
default:
return "unknown symbol";
}
}
char *
article( type )
unsigned char type;
{
switch ( type ) {
case N_PSO:
return "a source file";
case N_PSOL:
return "an include file";
case N_PGLABEL:
return "a label";
case N_PGCONST:
return "a constant";
case N_PGTYPE:
return "a type";
case N_PGVAR:
return "a variable";
case N_PGFUNC:
return "a function";
case N_PGPROC:
return "a procedure";
case N_PEFUNC:
return "an external function";
case N_PEPROC:
return "an external procedure";
case N_PLDATA:
return "a library variable";
case N_PLTEXT:
return "a library routine";
default:
return "an unknown symbol";
}
}
Continuous source code history from original PDP-7 UNIX to FreeBSD 2, the first fully unencumbered FreeBSD release.