X-Git-Url: http://git.efficios.com/?a=blobdiff_plain;f=binutils%2Fstrings.c;h=dc42a6f2f904a75aa41313ab0e8fc71391101945;hb=5a2296ac1b3c26cab589bddfded07a560e434d03;hp=02ef67ba37ac890bb5d1cdfdf41a74b4ae461825;hpb=cef35d488209227385e0aa9ab92ebf5a861c3780;p=deliverable%2Fbinutils-gdb.git diff --git a/binutils/strings.c b/binutils/strings.c index 02ef67ba37..dc42a6f2f9 100644 --- a/binutils/strings.c +++ b/binutils/strings.c @@ -1,9 +1,9 @@ /* strings -- print the strings of printable characters in files - Copyright (C) 1993, 94 Free Software Foundation, Inc. + Copyright (C) 1993-2020 Free Software Foundation, Inc. This program 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 2, or (at your option) + the Free Software Foundation; either version 3, or (at your option) any later version. This program is distributed in the hope that it will be useful, @@ -13,14 +13,18 @@ You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ + Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston, MA + 02110-1301, USA. */ /* Usage: strings [options] file... Options: --all -a - - Do not scan only the initialized data section of object files. + - Scan each file in its entirety. + + --data + -d Scan only the initialized data section(s) of object files. --print-file-name -f Print the name of the file before each string. @@ -34,40 +38,57 @@ -t {o,x,d} Print the offset within the file before each string, in octal/hex/decimal. + --include-all-whitespace + -w By default tab and space are the only whitepace included in graphic + char sequences. This option considers all of isspace() valid. + -o Like -to. (Some other implementations have -o like -to, others like -td. We chose one arbitrarily.) + --encoding={s,S,b,l,B,L} + -e {s,S,b,l,B,L} + Select character encoding: 7-bit-character, 8-bit-character, + bigendian 16-bit, littleendian 16-bit, bigendian 32-bit, + littleendian 32-bit. + --target=BFDNAME + -T {bfdname} Specify a non-default object file format. + --output-separator=sep_string + -s sep_string String used to separate parsed strings in output. + Default is newline. + --help -h Print the usage message on the standard output. --version + -V -v Print the program version number. Written by Richard Stallman and David MacKenzie . */ -#include -#include -#include -#include +#include "sysdep.h" #include "bfd.h" +#include "getopt.h" +#include "libiberty.h" +#include "safe-ctype.h" #include "bucomm.h" -#ifdef isascii -#define isgraphic(c) (isascii (c) && isprint (c)) -#else -#define isgraphic(c) (isprint (c)) -#endif +#define STRING_ISGRAPHIC(c) \ + ( (c) >= 0 \ + && (c) <= 255 \ + && ((c) == '\t' || ISPRINT (c) || (encoding == 'S' && (c) > 127) \ + || (include_all_whitespace && ISSPACE (c))) \ + ) #ifndef errno extern int errno; #endif /* The BFD section flags that identify an initialized data section. */ -#define DATA_FLAGS (SEC_ALLOC | SEC_LOAD | SEC_DATA | SEC_HAS_CONTENTS) +#define DATA_FLAGS (SEC_ALLOC | SEC_LOAD | SEC_HAS_CONTENTS) /* Radix for printing addresses (must be 8, 10 or 16). */ static int address_radix; @@ -75,91 +96,121 @@ static int address_radix; /* Minimum length of sequence of graphic chars to trigger output. */ static int string_min; -/* true means print address within file for each string. */ -static boolean print_addresses; +/* Whether or not we include all whitespace as a graphic char. */ +static bfd_boolean include_all_whitespace; -/* true means print filename for each string. */ -static boolean print_filenames; +/* TRUE means print address within file for each string. */ +static bfd_boolean print_addresses; -/* true means for object files scan only the data section. */ -static boolean datasection_only; +/* TRUE means print filename for each string. */ +static bfd_boolean print_filenames; -/* true if we found an initialized data section in the current file. */ -static boolean got_a_section; +/* TRUE means for object files scan only the data section. */ +static bfd_boolean datasection_only; /* The BFD object file format. */ static char *target; -extern char *program_version; +/* The character encoding format. */ +static char encoding; +static int encoding_bytes; + +/* Output string used to separate parsed strings */ +static char *output_separator; static struct option long_options[] = { {"all", no_argument, NULL, 'a'}, + {"data", no_argument, NULL, 'd'}, {"print-file-name", no_argument, NULL, 'f'}, {"bytes", required_argument, NULL, 'n'}, {"radix", required_argument, NULL, 't'}, + {"include-all-whitespace", no_argument, NULL, 'w'}, + {"encoding", required_argument, NULL, 'e'}, {"target", required_argument, NULL, 'T'}, + {"output-separator", required_argument, NULL, 's'}, {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {NULL, 0, NULL, 0} }; -static boolean strings_file PARAMS ((char *file)); -static int integer_arg PARAMS ((char *s)); -static void print_strings PARAMS ((char *filename, FILE *stream, - file_ptr address, int stop_point, - int magiccount, char *magic)); -static void usage PARAMS ((FILE *stream, int status)); +static bfd_boolean strings_file (char *); +static void print_strings (const char *, FILE *, file_ptr, int, int, char *); +static void usage (FILE *, int) ATTRIBUTE_NORETURN; -void -main (argc, argv) - int argc; - char **argv; +int main (int, char **); + +int +main (int argc, char **argv) { int optc; int exit_status = 0; - boolean files_given = false; + bfd_boolean files_given = FALSE; + char *s; + int numeric_opt = 0; + +#if defined (HAVE_SETLOCALE) + setlocale (LC_ALL, ""); +#endif + bindtextdomain (PACKAGE, LOCALEDIR); + textdomain (PACKAGE); program_name = argv[0]; - string_min = -1; - print_addresses = false; - print_filenames = false; - datasection_only = true; + xmalloc_set_program_name (program_name); + bfd_set_error_program_name (program_name); + + expandargv (&argc, &argv); + + string_min = 4; + include_all_whitespace = FALSE; + print_addresses = FALSE; + print_filenames = FALSE; + if (DEFAULT_STRINGS_ALL) + datasection_only = FALSE; + else + datasection_only = TRUE; target = NULL; + encoding = 's'; + output_separator = NULL; - while ((optc = getopt_long (argc, argv, "afn:ot:v0123456789", + while ((optc = getopt_long (argc, argv, "adfhHn:wot:e:T:s:Vv0123456789", long_options, (int *) 0)) != EOF) { switch (optc) { case 'a': - datasection_only = false; + datasection_only = FALSE; + break; + + case 'd': + datasection_only = TRUE; break; case 'f': - print_filenames = true; + print_filenames = TRUE; break; + case 'H': case 'h': usage (stdout, 0); case 'n': - string_min = integer_arg (optarg); - if (string_min < 1) - { - fprintf (stderr, "%s: invalid number %s\n", - program_name, optarg); - exit (1); - } + string_min = (int) strtoul (optarg, &s, 0); + if (s != NULL && *s != 0) + fatal (_("invalid integer argument %s"), optarg); + break; + + case 'w': + include_all_whitespace = TRUE; break; case 'o': - print_addresses = true; + print_addresses = TRUE; address_radix = 8; break; case 't': - print_addresses = true; + print_addresses = TRUE; if (optarg[1] != '\0') usage (stderr, 1); switch (optarg[0]) @@ -185,112 +236,183 @@ main (argc, argv) target = optarg; break; + case 'e': + if (optarg[1] != '\0') + usage (stderr, 1); + encoding = optarg[0]; + break; + + case 's': + output_separator = optarg; + break; + + case 'V': case 'v': - printf ("GNU %s version %s\n", program_name, program_version); - exit (0); + print_version ("strings"); + break; case '?': usage (stderr, 1); default: - if (string_min < 0) - string_min = optc; - else - string_min = string_min * 10 + optc - '0'; + numeric_opt = optind; break; } } - if (string_min < 0) - string_min = 4; + if (numeric_opt != 0) + { + string_min = (int) strtoul (argv[numeric_opt - 1] + 1, &s, 0); + if (s != NULL && *s != 0) + fatal (_("invalid integer argument %s"), argv[numeric_opt - 1] + 1); + } + if (string_min < 1) + fatal (_("invalid minimum string length %d"), string_min); - bfd_init (); + switch (encoding) + { + case 'S': + case 's': + encoding_bytes = 1; + break; + case 'b': + case 'l': + encoding_bytes = 2; + break; + case 'B': + case 'L': + encoding_bytes = 4; + break; + default: + usage (stderr, 1); + } + + if (bfd_init () != BFD_INIT_MAGIC) + fatal (_("fatal error: libbfd ABI mismatch")); + set_default_bfd_target (); - for (; optind < argc; ++optind) + if (optind >= argc) { - if (!strcmp (argv[optind], "-")) - datasection_only = false; - else + datasection_only = FALSE; + SET_BINARY (fileno (stdin)); + print_strings ("{standard input}", stdin, 0, 0, 0, (char *) NULL); + files_given = TRUE; + } + else + { + for (; optind < argc; ++optind) { - files_given = true; - exit_status |= (strings_file (argv[optind]) == false); + if (strcmp (argv[optind], "-") == 0) + datasection_only = FALSE; + else + { + files_given = TRUE; + exit_status |= !strings_file (argv[optind]); + } } } - if (files_given == false) + if (!files_given) usage (stderr, 1); - exit (exit_status); + return (exit_status); } -/* Scan section SECT of the file ABFD, whose printable name is FILE. - If it contains initialized data, - set `got_a_section' and print the strings in it. */ +/* Scan section SECT of the file ABFD, whose printable name is + FILENAME. If it contains initialized data set GOT_A_SECTION and + print the strings in it. */ static void -strings_a_section (abfd, sect, file) - bfd *abfd; - asection *sect; - PTR file; +strings_a_section (bfd *abfd, asection *sect, const char *filename, + bfd_boolean *got_a_section) { - if ((sect->flags & DATA_FLAGS) == DATA_FLAGS) + bfd_size_type sectsize; + bfd_byte *mem; + + if ((sect->flags & DATA_FLAGS) != DATA_FLAGS) + return; + + sectsize = bfd_section_size (sect); + if (sectsize == 0) + return; + + if (!bfd_malloc_and_get_section (abfd, sect, &mem)) { - bfd_size_type sz = bfd_get_section_size_before_reloc (sect); - PTR mem = xmalloc (sz); - if (bfd_get_section_contents (abfd, sect, mem, (file_ptr) 0, sz)) - { - got_a_section = true; - print_strings (file, (FILE *) NULL, sect->filepos, 0, sz, mem); - } - free (mem); + non_fatal (_("%s: Reading section %s failed: %s"), + filename, sect->name, bfd_errmsg (bfd_get_error ())); + return; } + + *got_a_section = TRUE; + print_strings (filename, NULL, sect->filepos, 0, sectsize, (char *) mem); + free (mem); } /* Scan all of the sections in FILE, and print the strings in the initialized data section(s). - Return true if successful, - false if not (such as if FILE is not an object file). */ + Return TRUE if successful, + FALSE if not (such as if FILE is not an object file). */ -static boolean -strings_object_file (file) - char *file; +static bfd_boolean +strings_object_file (const char *file) { - bfd *abfd = bfd_openr (file, target); + bfd *abfd; + asection *s; + bfd_boolean got_a_section; + + abfd = bfd_openr (file, target); if (abfd == NULL) - { - /* Treat the file as a non-object file. */ - return false; - } + /* Treat the file as a non-object file. */ + return FALSE; /* This call is mainly for its side effect of reading in the sections. We follow the traditional behavior of `strings' in that we don't complain if we don't recognize a file to be an object file. */ - if (bfd_check_format (abfd, bfd_object) == false) + if (!bfd_check_format (abfd, bfd_object)) { bfd_close (abfd); - return false; + return FALSE; } - got_a_section = false; - bfd_map_over_sections (abfd, strings_a_section, file); + got_a_section = FALSE; + for (s = abfd->sections; s != NULL; s = s->next) + strings_a_section (abfd, s, file, &got_a_section); if (!bfd_close (abfd)) { bfd_nonfatal (file); - return false; + return FALSE; } return got_a_section; } -/* Print the strings in FILE. Return true if ok, false if an error occurs. */ +/* Print the strings in FILE. Return TRUE if ok, FALSE if an error occurs. */ -static boolean -strings_file (file) - char *file; +static bfd_boolean +strings_file (char *file) { + struct stat st; + + /* get_file_size does not support non-S_ISREG files. */ + + if (stat (file, &st) < 0) + { + if (errno == ENOENT) + non_fatal (_("'%s': No such file"), file); + else + non_fatal (_("Warning: could not locate '%s'. reason: %s"), + file, strerror (errno)); + return FALSE; + } + else if (S_ISDIR (st.st_mode)) + { + non_fatal (_("Warning: '%s' is a directory"), file); + return FALSE; + } + /* If we weren't told to scan the whole file, try to open it as an object file and only look at initialized data sections. If that fails, fall back to the @@ -299,12 +421,12 @@ strings_file (file) { FILE *stream; - stream = fopen (file, "r"); + stream = fopen (file, FOPEN_RB); if (stream == NULL) { fprintf (stderr, "%s: ", program_name); perror (file); - return false; + return FALSE; } print_strings (file, stream, (file_ptr) 0, 0, 0, (char *) 0); @@ -313,11 +435,123 @@ strings_file (file) { fprintf (stderr, "%s: ", program_name); perror (file); - return false; + return FALSE; + } + } + + return TRUE; +} + +/* Read the next character, return EOF if none available. + Assume that STREAM is positioned so that the next byte read + is at address ADDRESS in the file. + + If STREAM is NULL, do not read from it. + The caller can supply a buffer of characters + to be processed before the data in STREAM. + MAGIC is the address of the buffer and + MAGICCOUNT is how many characters are in it. */ + +static long +get_char (FILE *stream, file_ptr *address, int *magiccount, char **magic) +{ + int c, i; + long r = 0; + + for (i = 0; i < encoding_bytes; i++) + { + if (*magiccount) + { + (*magiccount)--; + c = *(*magic)++; + } + else + { + if (stream == NULL) + return EOF; + + /* Only use getc_unlocked if we found a declaration for it. + Otherwise, libc is not thread safe by default, and we + should not use it. */ + +#if defined(HAVE_GETC_UNLOCKED) && HAVE_DECL_GETC_UNLOCKED + c = getc_unlocked (stream); +#else + c = getc (stream); +#endif + if (c == EOF) + return EOF; } + + (*address)++; + r = (r << 8) | (c & 0xff); + } + + switch (encoding) + { + default: + break; + case 'l': + r = ((r & 0xff) << 8) | ((r & 0xff00) >> 8); + break; + case 'L': + r = (((r & 0xff) << 24) | ((r & 0xff00) << 8) + | ((r & 0xff0000) >> 8) | ((r & 0xff000000) >> 24)); + break; } - return true; + return r; +} + +/* Throw away one byte of a (possibly) multi-byte char C, updating + address and buffer to suit. */ + +static void +unget_part_char (long c, file_ptr *address, int *magiccount, char **magic) +{ + static char tmp[4]; + + if (encoding_bytes > 1) + { + *address -= encoding_bytes - 1; + + if (*magiccount == 0) + { + /* If no magic buffer exists, use temp buffer. */ + switch (encoding) + { + default: + break; + case 'b': + tmp[0] = c & 0xff; + *magiccount = 1; + break; + case 'l': + tmp[0] = (c >> 8) & 0xff; + *magiccount = 1; + break; + case 'B': + tmp[0] = (c >> 16) & 0xff; + tmp[1] = (c >> 8) & 0xff; + tmp[2] = c & 0xff; + *magiccount = 3; + break; + case 'L': + tmp[0] = (c >> 8) & 0xff; + tmp[1] = (c >> 16) & 0xff; + tmp[2] = (c >> 24) & 0xff; + *magiccount = 3; + break; + } + *magic = tmp; + } + else + { + /* If magic buffer exists, rewind. */ + *magic -= encoding_bytes - 1; + *magiccount += encoding_bytes - 1; + } + } } /* Find the strings in file FILENAME, read from STREAM. @@ -333,181 +567,163 @@ strings_file (file) Those characters come at address ADDRESS and the data in STREAM follow. */ static void -print_strings (filename, stream, address, stop_point, magiccount, magic) - char *filename; - FILE *stream; - file_ptr address; - int stop_point; - int magiccount; - char *magic; +print_strings (const char *filename, FILE *stream, file_ptr address, + int stop_point, int magiccount, char *magic) { - int bufsize = 100; - char *buf = (char *) xmalloc (bufsize); + char *buf = (char *) xmalloc (sizeof (char) * (string_min + 1)); while (1) { + file_ptr start; int i; - int c; + long c; /* See if the next `string_min' chars are all graphic chars. */ tryline: if (stop_point && address >= stop_point) break; + start = address; for (i = 0; i < string_min; i++) { - if (magiccount) + c = get_char (stream, &address, &magiccount, &magic); + if (c == EOF) { - magiccount--; - c = *magic++; + free (buf); + return; } - else + + if (! STRING_ISGRAPHIC (c)) { - if (stream == NULL) - return; - c = getc (stream); - if (c < 0) - return; + /* Found a non-graphic. Try again starting with next byte. */ + unget_part_char (c, &address, &magiccount, &magic); + goto tryline; } - address++; - if (!isgraphic (c)) - /* Found a non-graphic. Try again starting with next char. */ - goto tryline; buf[i] = c; } - /* We found a run of `string_min' graphic characters. - Now see if it is terminated with a NUL byte or a newline. */ - while (1) - { - if (i == bufsize) - { - bufsize *= 2; - buf = (char *) xrealloc (buf, bufsize); - } - if (magiccount) - { - magiccount--; - c = *magic++; - } - else - { - if (stream == NULL) - return; - c = getc (stream); - if (c < 0) - return; - } - address++; - if (c == '\0' || c == '\n') - break; /* It is; print this string. */ - if (!isgraphic (c)) - goto tryline; /* It isn't; give up on this string. */ - buf[i++] = c; /* The string continues; store it all. */ - } + /* We found a run of `string_min' graphic characters. Print up + to the next non-graphic character. */ - /* If we get here, the string is all graphics and properly terminated, - so print it. It is all in `buf' and `i' is its length. */ - buf[i] = '\0'; if (print_filenames) printf ("%s: ", filename); if (print_addresses) switch (address_radix) { case 8: - printf ("%7lo ", (unsigned long) (address - i - 1)); +#ifdef HAVE_LONG_LONG + if (sizeof (start) > sizeof (long)) + { +# ifndef __MSVCRT__ + printf ("%7llo ", (unsigned long long) start); +# else + printf ("%7I64o ", (unsigned long long) start); +# endif + } + else +#elif !BFD_HOST_64BIT_LONG + if (start != (unsigned long) start) + printf ("++%7lo ", (unsigned long) start); + else +#endif + printf ("%7lo ", (unsigned long) start); break; case 10: - printf ("%7ld ", (long) (address - i - 1)); +#ifdef HAVE_LONG_LONG + if (sizeof (start) > sizeof (long)) + { +# ifndef __MSVCRT__ + printf ("%7llu ", (unsigned long long) start); +# else + printf ("%7I64d ", (unsigned long long) start); +# endif + } + else +#elif !BFD_HOST_64BIT_LONG + if (start != (unsigned long) start) + printf ("++%7lu ", (unsigned long) start); + else +#endif + printf ("%7ld ", (long) start); break; case 16: - printf ("%7lx ", (unsigned long) (address - i - 1)); +#ifdef HAVE_LONG_LONG + if (sizeof (start) > sizeof (long)) + { +# ifndef __MSVCRT__ + printf ("%7llx ", (unsigned long long) start); +# else + printf ("%7I64x ", (unsigned long long) start); +# endif + } + else +#elif !BFD_HOST_64BIT_LONG + if (start != (unsigned long) start) + printf ("%lx%8.8lx ", (unsigned long) (start >> 32), + (unsigned long) (start & 0xffffffff)); + else +#endif + printf ("%7lx ", (unsigned long) start); break; } - for (i = 0; (c = buf[i]) != '\0'; i++) - switch (c) - { - case '\n': - printf ("\\n"); - break; - case '\t': - printf ("\\t"); - break; - case '\f': - printf ("\\f"); - break; - case '\b': - printf ("\\b"); - break; - case '\r': - printf ("\\r"); - break; - default: - putchar (c); - } - putchar ('\n'); - } -} - -/* Parse string S as an integer, using decimal radix by default, - but allowing octal and hex numbers as in C. */ + buf[i] = '\0'; + fputs (buf, stdout); -static int -integer_arg (s) - char *s; -{ - int value; - int radix = 10; - char *p = s; - int c; - - if (*p != '0') - radix = 10; - else if (*++p == 'x') - { - radix = 16; - p++; - } - else - radix = 8; + while (1) + { + c = get_char (stream, &address, &magiccount, &magic); + if (c == EOF) + break; + if (! STRING_ISGRAPHIC (c)) + { + unget_part_char (c, &address, &magiccount, &magic); + break; + } + putchar (c); + } - value = 0; - while (((c = *p++) >= '0' && c <= '9') - || (radix == 16 && (c & ~40) >= 'A' && (c & ~40) <= 'Z')) - { - value *= radix; - if (c >= '0' && c <= '9') - value += c - '0'; + if (output_separator) + fputs (output_separator, stdout); else - value += (c & ~40) - 'A'; + putchar ('\n'); } - - if (c == 'b') - value *= 512; - else if (c == 'B') - value *= 1024; - else - p--; - - if (*p) - { - fprintf (stderr, "%s: invalid integer argument %s\n", program_name, s); - exit (1); - } - return value; + free (buf); } - + static void -usage (stream, status) - FILE *stream; - int status; +usage (FILE *stream, int status) { - fprintf (stream, "\ -Usage: %s [-afov] [-n min-len] [-min-len] [-t {o,x,d}] [-]\n\ - [--all] [--print-file-name] [--bytes=min-len] [--radix={o,x,d}]\n\ - [--target=bfdname] [--help] [--version] file...\n", - program_name); + fprintf (stream, _("Usage: %s [option(s)] [file(s)]\n"), program_name); + fprintf (stream, _(" Display printable strings in [file(s)] (stdin by default)\n")); + fprintf (stream, _(" The options are:\n")); + + if (DEFAULT_STRINGS_ALL) + fprintf (stream, _("\ + -a - --all Scan the entire file, not just the data section [default]\n\ + -d --data Only scan the data sections in the file\n")); + else + fprintf (stream, _("\ + -a - --all Scan the entire file, not just the data section\n\ + -d --data Only scan the data sections in the file [default]\n")); + + fprintf (stream, _("\ + -f --print-file-name Print the name of the file before each string\n\ + -n --bytes=[number] Locate & print any NUL-terminated sequence of at\n\ + - least [number] characters (default 4).\n\ + -t --radix={o,d,x} Print the location of the string in base 8, 10 or 16\n\ + -w --include-all-whitespace Include all whitespace as valid string characters\n\ + -o An alias for --radix=o\n\ + -T --target= Specify the binary file format\n\ + -e --encoding={s,S,b,l,B,L} Select character size and endianness:\n\ + s = 7-bit, S = 8-bit, {b,l} = 16-bit, {B,L} = 32-bit\n\ + -s --output-separator= String used to separate strings in output.\n\ + @ Read options from \n\ + -h --help Display this information\n\ + -v -V --version Print the program's version number\n")); + list_supported_targets (program_name, stream); + if (REPORT_BUGS_TO[0] && status == 0) + fprintf (stream, _("Report bugs to %s\n"), REPORT_BUGS_TO); exit (status); }