/* strings -- print the strings of printable characters in files
- Copyright 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001,
- 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009
- Free Software Foundation, Inc.
+ Copyright (C) 1993-2023 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
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.
--bytes=min-len
-n min-len
-min-len Print graphic char sequences, MIN-LEN or more bytes long,
- that are followed by a NUL or a newline. Default is 4.
+ that are followed by a NUL or a non-displayable character.
+ Default is 4.
--radix={o,x,d}
-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.)
-T {bfdname}
Specify a non-default object file format.
+ --unicode={default|locale|invalid|hex|escape|highlight}
+ -U {d|l|i|x|e|h}
+ Determine how to handle UTF-8 unicode characters. The default
+ is no special treatment. All other versions of this option
+ only apply if the encoding is valid and enabling the option
+ implies --encoding=S.
+ The 'locale' option displays the characters according to the
+ current locale. The 'invalid' option treats them as
+ non-string characters. The 'hex' option displays them as hex
+ byte sequences. The 'escape' option displays them as escape
+ sequences and the 'highlight' option displays them as
+ coloured escape sequences.
+
+ --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.
#include "getopt.h"
#include "libiberty.h"
#include "safe-ctype.h"
-#include <sys/stat.h>
#include "bucomm.h"
+#ifndef streq
+#define streq(a,b) (strcmp ((a),(b)) == 0)
+#endif
+
+typedef enum unicode_display_type
+{
+ unicode_default = 0,
+ unicode_locale,
+ unicode_escape,
+ unicode_hex,
+ unicode_highlight,
+ unicode_invalid
+} unicode_display_type;
+
+static unicode_display_type unicode_display = unicode_default;
+
#define STRING_ISGRAPHIC(c) \
( (c) >= 0 \
&& (c) <= 255 \
- && ((c) == '\t' || ISPRINT (c) || (encoding == 'S' && (c) > 127)))
+ && ((c) == '\t' || ISPRINT (c) || (encoding == 'S' && (c) > 127) \
+ || (include_all_whitespace && ISSPACE (c))) \
+ )
#ifndef errno
extern int errno;
static int address_radix;
/* Minimum length of sequence of graphic chars to trigger output. */
-static int string_min;
+static unsigned int string_min;
+
+/* Whether or not we include all whitespace as a graphic char. */
+static bool include_all_whitespace;
/* TRUE means print address within file for each string. */
-static bfd_boolean print_addresses;
+static bool print_addresses;
/* TRUE means print filename for each string. */
-static bfd_boolean print_filenames;
+static bool print_filenames;
/* TRUE means for object files scan only the data section. */
-static bfd_boolean datasection_only;
-
-/* TRUE if we found an initialized data section in the current file. */
-static bfd_boolean got_a_section;
+static bool datasection_only;
/* The BFD object file format. */
static char *target;
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'},
- {"print-file-name", no_argument, NULL, 'f'},
{"bytes", required_argument, NULL, 'n'},
- {"radix", required_argument, NULL, 't'},
+ {"data", no_argument, NULL, 'd'},
{"encoding", required_argument, NULL, 'e'},
- {"target", required_argument, NULL, 'T'},
{"help", no_argument, NULL, 'h'},
+ {"include-all-whitespace", no_argument, NULL, 'w'},
+ {"output-separator", required_argument, NULL, 's'},
+ {"print-file-name", no_argument, NULL, 'f'},
+ {"radix", required_argument, NULL, 't'},
+ {"target", required_argument, NULL, 'T'},
+ {"unicode", required_argument, NULL, 'U'},
{"version", no_argument, NULL, 'v'},
{NULL, 0, NULL, 0}
};
-/* Records the size of a named file so that we
- do not repeatedly run bfd_stat() on it. */
-
-typedef struct
-{
- const char * filename;
- bfd_size_type filesize;
-} filename_and_size_t;
-
-static void strings_a_section (bfd *, asection *, void *);
-static bfd_boolean strings_object_file (const char *);
-static bfd_boolean strings_file (char *file);
-static void print_strings (const char *, FILE *, file_ptr, int, int, char *);
-static void usage (FILE *, int);
-static long get_char (FILE *, file_ptr *, int *, char **);
+static bool strings_file (char *);
+static void print_strings (const char *, FILE *, file_ptr, int, char *);
+static void usage (FILE *, int) ATTRIBUTE_NORETURN;
\f
int main (int, char **);
{
int optc;
int exit_status = 0;
- bfd_boolean files_given = FALSE;
+ bool 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];
xmalloc_set_program_name (program_name);
+ bfd_set_error_program_name (program_name);
expandargv (&argc, &argv);
string_min = 4;
- print_addresses = FALSE;
- print_filenames = FALSE;
- datasection_only = TRUE;
+ 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, "afhHn:ot:e:T:Vv0123456789",
+ while ((optc = getopt_long (argc, argv, "adfhHn:wot:e:T:s:U: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':
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])
encoding = optarg[0];
break;
+ case 's':
+ output_separator = optarg;
+ break;
+
+ case 'U':
+ if (streq (optarg, "default") || streq (optarg, "d"))
+ unicode_display = unicode_default;
+ else if (streq (optarg, "locale") || streq (optarg, "l"))
+ unicode_display = unicode_locale;
+ else if (streq (optarg, "escape") || streq (optarg, "e"))
+ unicode_display = unicode_escape;
+ else if (streq (optarg, "invalid") || streq (optarg, "i"))
+ unicode_display = unicode_invalid;
+ else if (streq (optarg, "hex") || streq (optarg, "x"))
+ unicode_display = unicode_hex;
+ else if (streq (optarg, "highlight") || streq (optarg, "h"))
+ unicode_display = unicode_highlight;
+ else
+ fatal (_("invalid argument to -U/--unicode: %s"), optarg);
+ break;
+
case 'V':
case 'v':
print_version ("strings");
}
}
+ if (unicode_display != unicode_default)
+ encoding = 'S';
+
if (numeric_opt != 0)
{
string_min = (int) strtoul (argv[numeric_opt - 1] + 1, &s, 0);
usage (stderr, 1);
}
- bfd_init ();
+ if (bfd_init () != BFD_INIT_MAGIC)
+ fatal (_("fatal error: libbfd ABI mismatch"));
set_default_bfd_target ();
if (optind >= argc)
{
- datasection_only = FALSE;
+ datasection_only = false;
SET_BINARY (fileno (stdin));
- print_strings ("{standard input}", stdin, 0, 0, 0, (char *) NULL);
- files_given = TRUE;
+ print_strings ("{standard input}", stdin, 0, 0, (char *) NULL);
+ files_given = true;
}
else
{
for (; optind < argc; ++optind)
{
- if (strcmp (argv[optind], "-") == 0)
- datasection_only = FALSE;
+ if (streq (argv[optind], "-"))
+ datasection_only = false;
else
{
- files_given = TRUE;
- exit_status |= strings_file (argv[optind]) == FALSE;
+ files_given = true;
+ exit_status |= !strings_file (argv[optind]);
}
}
}
return (exit_status);
}
\f
-/* Scan section SECT of the file ABFD, whose printable name is in
- ARG->filename and whose size might be in ARG->filesize. If it
- contains initialized data set `got_a_section' and print the
- strings in it.
-
- FIXME: We ought to be able to return error codes/messages for
- certain conditions. */
+/* 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 (bfd *abfd, asection *sect, void *arg)
+strings_a_section (bfd *abfd, asection *sect, const char *filename,
+ bool *got_a_section)
{
- filename_and_size_t * filename_and_sizep;
- bfd_size_type *filesizep;
bfd_size_type sectsize;
- void *mem;
-
+ bfd_byte *mem;
+
if ((sect->flags & DATA_FLAGS) != DATA_FLAGS)
return;
- sectsize = bfd_get_section_size (sect);
-
- if (sectsize <= 0)
+ sectsize = bfd_section_size (sect);
+ if (sectsize == 0)
return;
- /* Get the size of the file. This might have been cached for us. */
- filename_and_sizep = (filename_and_size_t *) arg;
- filesizep = & filename_and_sizep->filesize;
-
- if (*filesizep == 0)
+ if (!bfd_malloc_and_get_section (abfd, sect, &mem))
{
- struct stat st;
-
- if (bfd_stat (abfd, &st))
- return;
-
- /* Cache the result so that we do not repeatedly stat this file. */
- *filesizep = st.st_size;
- }
-
- /* Compare the size of the section against the size of the file.
- If the section is bigger then the file must be corrupt and
- we should not try dumping it. */
- if (sectsize >= *filesizep)
- return;
-
- mem = xmalloc (sectsize);
-
- if (bfd_get_section_contents (abfd, sect, mem, (file_ptr) 0, sectsize))
- {
- got_a_section = TRUE;
-
- print_strings (filename_and_sizep->filename, NULL, sect->filepos,
- 0, sectsize, (char *) 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, sectsize, (char *) mem);
free (mem);
}
Return TRUE if successful,
FALSE if not (such as if FILE is not an object file). */
-static bfd_boolean
+static bool
strings_object_file (const char *file)
{
- filename_and_size_t filename_and_size;
bfd *abfd;
+ asection *s;
+ bool got_a_section;
abfd = bfd_openr (file, target);
if (abfd == NULL)
/* Treat the file as a non-object file. */
- return FALSE;
+ 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
if (!bfd_check_format (abfd, bfd_object))
{
bfd_close (abfd);
- return FALSE;
+ return false;
}
- got_a_section = FALSE;
- filename_and_size.filename = file;
- filename_and_size.filesize = 0;
- bfd_map_over_sections (abfd, strings_a_section, & filename_and_size);
+ 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. */
-static bfd_boolean
+static bool
strings_file (char *file)
{
struct stat st;
else
non_fatal (_("Warning: could not locate '%s'. reason: %s"),
file, strerror (errno));
- return FALSE;
+ 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,
{
fprintf (stderr, "%s: ", program_name);
perror (file);
- return FALSE;
+ return false;
}
- print_strings (file, stream, (file_ptr) 0, 0, 0, (char *) 0);
+ print_strings (file, stream, (file_ptr) 0, 0, (char *) NULL);
if (fclose (stream) == EOF)
{
fprintf (stderr, "%s: ", program_name);
perror (file);
- return FALSE;
+ return false;
}
}
- return TRUE;
+ return true;
}
\f
/* Read the next character, return EOF if none available.
get_char (FILE *stream, file_ptr *address, int *magiccount, char **magic)
{
int c, i;
- long r = EOF;
- unsigned char buf[4];
+ long r = 0;
for (i = 0; i < encoding_bytes; i++)
{
}
(*address)++;
- buf[i] = c;
+ r = (r << 8) | (c & 0xff);
}
switch (encoding)
{
- case 'S':
- case 's':
- r = buf[0];
- break;
- case 'b':
- r = (buf[0] << 8) | buf[1];
+ default:
break;
case 'l':
- r = buf[0] | (buf[1] << 8);
- break;
- case 'B':
- r = ((long) buf[0] << 24) | ((long) buf[1] << 16) |
- ((long) buf[2] << 8) | buf[3];
+ r = ((r & 0xff) << 8) | ((r & 0xff00) >> 8);
break;
case 'L':
- r = buf[0] | ((long) buf[1] << 8) | ((long) buf[2] << 16) |
- ((long) buf[3] << 24);
+ r = (((r & 0xff) << 24) | ((r & 0xff00) << 8)
+ | ((r & 0xff0000) >> 8) | ((r & 0xff000000) >> 24));
+ break;
+ }
+
+ 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;
+ }
+ }
+}
+
+static void
+print_filename_and_address (const char * filename, file_ptr address)
+{
+ if (print_filenames)
+ printf ("%s: ", filename);
+
+ if (! print_addresses)
+ return;
+
+ switch (address_radix)
+ {
+ case 8:
+ if (sizeof (address) > sizeof (long))
+ {
+#ifndef __MSVCRT__
+ printf ("%7llo ", (unsigned long long) address);
+#else
+ printf ("%7I64o ", (unsigned long long) address);
+#endif
+ }
+ else
+ printf ("%7lo ", (unsigned long) address);
+ break;
+
+ case 10:
+ if (sizeof (address) > sizeof (long))
+ {
+#ifndef __MSVCRT__
+ printf ("%7llu ", (unsigned long long) address);
+#else
+ printf ("%7I64d ", (unsigned long long) address);
+#endif
+ }
+ else
+ printf ("%7ld ", (long) address);
+ break;
+
+ case 16:
+ if (sizeof (address) > sizeof (long))
+ {
+#ifndef __MSVCRT__
+ printf ("%7llx ", (unsigned long long) address);
+#else
+ printf ("%7I64x ", (unsigned long long) address);
+#endif
+ }
+ else
+ printf ("%7lx ", (unsigned long) address);
break;
}
+}
- if (r == EOF)
+/* Return non-zero if the bytes starting at BUFFER form a valid UTF-8 encoding.
+ If the encoding is valid then returns the number of bytes it uses. */
+
+static unsigned int
+is_valid_utf8 (const unsigned char * buffer, unsigned long buflen)
+{
+ if (buffer[0] < 0xc0)
return 0;
- return r;
+ if (buflen < 2)
+ return 0;
+
+ if ((buffer[1] & 0xc0) != 0x80)
+ return 0;
+
+ if ((buffer[0] & 0x20) == 0)
+ return 2;
+
+ if (buflen < 3)
+ return 0;
+
+ if ((buffer[2] & 0xc0) != 0x80)
+ return 0;
+
+ if ((buffer[0] & 0x10) == 0)
+ return 3;
+
+ if (buflen < 4)
+ return 0;
+
+ if ((buffer[3] & 0xc0) != 0x80)
+ return 0;
+
+ return 4;
+}
+
+/* Display a UTF-8 encoded character in BUFFER according to the setting
+ of unicode_display. The character is known to be valid.
+ Returns the number of bytes consumed. */
+
+static unsigned int
+display_utf8_char (const unsigned char * buffer)
+{
+ unsigned int j;
+ unsigned int utf8_len;
+
+ switch (buffer[0] & 0x30)
+ {
+ case 0x00:
+ case 0x10:
+ utf8_len = 2;
+ break;
+ case 0x20:
+ utf8_len = 3;
+ break;
+ default:
+ utf8_len = 4;
+ }
+
+ switch (unicode_display)
+ {
+ default:
+ fprintf (stderr, "ICE: unexpected unicode display type\n");
+ break;
+
+ case unicode_escape:
+ case unicode_highlight:
+ if (unicode_display == unicode_highlight && isatty (1))
+ printf ("\x1B[31;47m"); /* Red. */
+
+ switch (utf8_len)
+ {
+ case 2:
+ printf ("\\u%02x%02x",
+ ((buffer[0] & 0x1c) >> 2),
+ ((buffer[0] & 0x03) << 6) | (buffer[1] & 0x3f));
+ break;
+
+ case 3:
+ printf ("\\u%02x%02x",
+ ((buffer[0] & 0x0f) << 4) | ((buffer[1] & 0x3c) >> 2),
+ ((buffer[1] & 0x03) << 6) | ((buffer[2] & 0x3f)));
+ break;
+
+ case 4:
+ printf ("\\u%02x%02x%02x",
+ ((buffer[0] & 0x07) << 6) | ((buffer[1] & 0x3c) >> 2),
+ ((buffer[1] & 0x03) << 6) | ((buffer[2] & 0x3c) >> 2),
+ ((buffer[2] & 0x03) << 6) | ((buffer[3] & 0x3f)));
+ break;
+ default:
+ /* URG. */
+ break;
+ }
+
+ if (unicode_display == unicode_highlight && isatty (1))
+ printf ("\033[0m"); /* Default colour. */
+ break;
+
+ case unicode_hex:
+ putchar ('<');
+ printf ("0x");
+ for (j = 0; j < utf8_len; j++)
+ printf ("%02x", buffer [j]);
+ putchar ('>');
+ break;
+
+ case unicode_locale:
+ printf ("%.1s", buffer);
+ break;
+ }
+
+ return utf8_len;
+}
+
+/* Display strings in BUFFER. Treat any UTF-8 encoded characters encountered
+ according to the setting of the unicode_display variable. The buffer
+ contains BUFLEN bytes.
+
+ Display the characters as if they started at ADDRESS and are contained in
+ FILENAME. */
+
+static void
+print_unicode_buffer (const char * filename,
+ file_ptr address,
+ const unsigned char * buffer,
+ unsigned long buflen)
+{
+ /* Paranoia checks... */
+ if (filename == NULL
+ || buffer == NULL
+ || unicode_display == unicode_default
+ || encoding != 'S'
+ || encoding_bytes != 1)
+ {
+ fprintf (stderr, "ICE: bad arguments to print_unicode_buffer\n");
+ return;
+ }
+
+ if (buflen == 0)
+ return;
+
+ /* We must only display strings that are at least string_min *characters*
+ long. So we scan the buffer in two stages. First we locate the start
+ of a potential string. Then we walk along it until we have found
+ string_min characters. Then we go back to the start point and start
+ displaying characters according to the unicode_display setting. */
+
+ unsigned long start_point = 0;
+ unsigned long i = 0;
+ unsigned int char_len = 1;
+ unsigned int num_found = 0;
+
+ for (i = 0; i < buflen; i += char_len)
+ {
+ int c = buffer[i];
+
+ char_len = 1;
+
+ /* Find the first potential character of a string. */
+ if (! STRING_ISGRAPHIC (c))
+ {
+ num_found = 0;
+ continue;
+ }
+
+ if (c > 126)
+ {
+ if (c < 0xc0)
+ {
+ num_found = 0;
+ continue;
+ }
+
+ if ((char_len = is_valid_utf8 (buffer + i, buflen - i)) == 0)
+ {
+ char_len = 1;
+ num_found = 0;
+ continue;
+ }
+
+ if (unicode_display == unicode_invalid)
+ {
+ /* We have found a valid UTF-8 character, but we treat it as non-graphic. */
+ num_found = 0;
+ continue;
+ }
+ }
+
+ if (num_found == 0)
+ /* We have found a potential starting point for a string. */
+ start_point = i;
+
+ ++ num_found;
+
+ if (num_found >= string_min)
+ break;
+ }
+
+ if (num_found < string_min)
+ return;
+
+ print_filename_and_address (filename, address + start_point);
+
+ /* We have found string_min characters. Display them and any
+ more that follow. */
+ for (i = start_point; i < buflen; i += char_len)
+ {
+ int c = buffer[i];
+
+ char_len = 1;
+
+ if (! STRING_ISGRAPHIC (c))
+ break;
+ else if (c < 127)
+ putchar (c);
+ else if (! is_valid_utf8 (buffer + i, buflen - i))
+ break;
+ else if (unicode_display == unicode_invalid)
+ break;
+ else
+ char_len = display_utf8_char (buffer + i);
+ }
+
+ if (output_separator)
+ fputs (output_separator, stdout);
+ else
+ putchar ('\n');
+
+ /* FIXME: Using tail recursion here is lazy programming... */
+ print_unicode_buffer (filename, address + i, buffer + i, buflen - i);
+}
+
+static int
+get_unicode_byte (FILE * stream,
+ unsigned char * putback,
+ unsigned int * num_putback,
+ unsigned int * num_read)
+{
+ if (* num_putback > 0)
+ {
+ * num_putback = * num_putback - 1;
+ return putback [* num_putback];
+ }
+
+ * num_read = * num_read + 1;
+
+#if defined(HAVE_GETC_UNLOCKED) && HAVE_DECL_GETC_UNLOCKED
+ return getc_unlocked (stream);
+#else
+ return getc (stream);
+#endif
+}
+
+/* Helper function for print_unicode_stream. */
+
+static void
+print_unicode_stream_body (const char * filename,
+ file_ptr address,
+ FILE * stream,
+ unsigned char * putback_buf,
+ unsigned int num_putback,
+ unsigned char * print_buf)
+{
+ /* It would be nice if we could just read the stream into a buffer
+ and then process if with print_unicode_buffer. But the input
+ might be huge or it might time-locked (eg stdin). So instead
+ we go one byte at a time... */
+
+ file_ptr start_point = 0;
+ unsigned int num_read = 0;
+ unsigned int num_chars = 0;
+ unsigned int num_print = 0;
+ int c = 0;
+
+ /* Find a series of string_min characters. Put them into print_buf. */
+ do
+ {
+ if (num_chars >= string_min)
+ break;
+
+ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read);
+ if (c == EOF)
+ break;
+
+ if (! STRING_ISGRAPHIC (c))
+ {
+ num_chars = num_print = 0;
+ continue;
+ }
+
+ if (num_chars == 0)
+ start_point = num_read - 1;
+
+ if (c < 127)
+ {
+ print_buf[num_print] = c;
+ num_chars ++;
+ num_print ++;
+ continue;
+ }
+
+ if (c < 0xc0)
+ {
+ num_chars = num_print = 0;
+ continue;
+ }
+
+ /* We *might* have a UTF-8 sequence. Time to start peeking. */
+ char utf8[4];
+
+ utf8[0] = c;
+ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read);
+ if (c == EOF)
+ break;
+ utf8[1] = c;
+
+ if ((utf8[1] & 0xc0) != 0x80)
+ {
+ /* Invalid UTF-8. */
+ putback_buf[num_putback++] = utf8[1];
+ num_chars = num_print = 0;
+ continue;
+ }
+ else if ((utf8[0] & 0x20) == 0)
+ {
+ /* A valid 2-byte UTF-8 encoding. */
+ if (unicode_display == unicode_invalid)
+ {
+ putback_buf[num_putback++] = utf8[1];
+ num_chars = num_print = 0;
+ }
+ else
+ {
+ print_buf[num_print ++] = utf8[0];
+ print_buf[num_print ++] = utf8[1];
+ num_chars ++;
+ }
+ continue;
+ }
+
+ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read);
+ if (c == EOF)
+ break;
+ utf8[2] = c;
+
+ if ((utf8[2] & 0xc0) != 0x80)
+ {
+ /* Invalid UTF-8. */
+ putback_buf[num_putback++] = utf8[2];
+ putback_buf[num_putback++] = utf8[1];
+ num_chars = num_print = 0;
+ continue;
+ }
+ else if ((utf8[0] & 0x10) == 0)
+ {
+ /* A valid 3-byte UTF-8 encoding. */
+ if (unicode_display == unicode_invalid)
+ {
+ putback_buf[num_putback++] = utf8[2];
+ putback_buf[num_putback++] = utf8[1];
+ num_chars = num_print = 0;
+ }
+ else
+ {
+ print_buf[num_print ++] = utf8[0];
+ print_buf[num_print ++] = utf8[1];
+ print_buf[num_print ++] = utf8[2];
+ num_chars ++;
+ }
+ continue;
+ }
+
+ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read);
+ if (c == EOF)
+ break;
+ utf8[3] = c;
+
+ if ((utf8[3] & 0xc0) != 0x80)
+ {
+ /* Invalid UTF-8. */
+ putback_buf[num_putback++] = utf8[3];
+ putback_buf[num_putback++] = utf8[2];
+ putback_buf[num_putback++] = utf8[1];
+ num_chars = num_print = 0;
+ }
+ /* We have a valid 4-byte UTF-8 encoding. */
+ else if (unicode_display == unicode_invalid)
+ {
+ putback_buf[num_putback++] = utf8[3];
+ putback_buf[num_putback++] = utf8[1];
+ putback_buf[num_putback++] = utf8[2];
+ num_chars = num_print = 0;
+ }
+ else
+ {
+ print_buf[num_print ++] = utf8[0];
+ print_buf[num_print ++] = utf8[1];
+ print_buf[num_print ++] = utf8[2];
+ print_buf[num_print ++] = utf8[3];
+ num_chars ++;
+ }
+ }
+ while (1);
+
+ if (num_chars >= string_min)
+ {
+ /* We know that we have string_min valid characters in print_buf,
+ and there may be more to come in the stream. Start displaying
+ them. */
+
+ print_filename_and_address (filename, address + start_point);
+
+ unsigned int i;
+ for (i = 0; i < num_print;)
+ {
+ if (print_buf[i] < 127)
+ putchar (print_buf[i++]);
+ else
+ i += display_utf8_char (print_buf + i);
+ }
+
+ /* OK so now we have to start read unchecked bytes. */
+
+ /* Find a series of string_min characters. Put them into print_buf. */
+ do
+ {
+ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read);
+ if (c == EOF)
+ break;
+
+ if (! STRING_ISGRAPHIC (c))
+ break;
+
+ if (c < 127)
+ {
+ putchar (c);
+ continue;
+ }
+
+ if (c < 0xc0)
+ break;
+
+ /* We *might* have a UTF-8 sequence. Time to start peeking. */
+ unsigned char utf8[4];
+
+ utf8[0] = c;
+ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read);
+ if (c == EOF)
+ break;
+ utf8[1] = c;
+
+ if ((utf8[1] & 0xc0) != 0x80)
+ {
+ /* Invalid UTF-8. */
+ putback_buf[num_putback++] = utf8[1];
+ break;
+ }
+ else if ((utf8[0] & 0x20) == 0)
+ {
+ /* Valid 2-byte UTF-8. */
+ if (unicode_display == unicode_invalid)
+ {
+ putback_buf[num_putback++] = utf8[1];
+ break;
+ }
+ else
+ {
+ (void) display_utf8_char (utf8);
+ continue;
+ }
+ }
+
+ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read);
+ if (c == EOF)
+ break;
+ utf8[2] = c;
+
+ if ((utf8[2] & 0xc0) != 0x80)
+ {
+ /* Invalid UTF-8. */
+ putback_buf[num_putback++] = utf8[2];
+ putback_buf[num_putback++] = utf8[1];
+ break;
+ }
+ else if ((utf8[0] & 0x10) == 0)
+ {
+ /* Valid 3-byte UTF-8. */
+ if (unicode_display == unicode_invalid)
+ {
+ putback_buf[num_putback++] = utf8[2];
+ putback_buf[num_putback++] = utf8[1];
+ break;
+ }
+ else
+ {
+ (void) display_utf8_char (utf8);
+ continue;
+ }
+ }
+
+ c = get_unicode_byte (stream, putback_buf, & num_putback, & num_read);
+ if (c == EOF)
+ break;
+ utf8[3] = c;
+
+ if ((utf8[3] & 0xc0) != 0x80)
+ {
+ /* Invalid UTF-8. */
+ putback_buf[num_putback++] = utf8[3];
+ putback_buf[num_putback++] = utf8[2];
+ putback_buf[num_putback++] = utf8[1];
+ break;
+ }
+ else if (unicode_display == unicode_invalid)
+ {
+ putback_buf[num_putback++] = utf8[3];
+ putback_buf[num_putback++] = utf8[2];
+ putback_buf[num_putback++] = utf8[1];
+ break;
+ }
+ else
+ /* A valid 4-byte UTF-8 encoding. */
+ (void) display_utf8_char (utf8);
+ }
+ while (1);
+
+ if (output_separator)
+ fputs (output_separator, stdout);
+ else
+ putchar ('\n');
+ }
+
+ if (c != EOF)
+ /* FIXME: Using tail recursion here is lazy, but it works. */
+ print_unicode_stream_body (filename, address + num_read, stream, putback_buf, num_putback, print_buf);
+}
+
+/* Display strings read in from STREAM. Treat any UTF-8 encoded characters
+ encountered according to the setting of the unicode_display variable.
+ The stream is positioned at ADDRESS and is attached to FILENAME. */
+
+static void
+print_unicode_stream (const char * filename,
+ file_ptr address,
+ FILE * stream)
+{
+ /* Paranoia checks... */
+ if (filename == NULL
+ || stream == NULL
+ || unicode_display == unicode_default
+ || encoding != 'S'
+ || encoding_bytes != 1)
+ {
+ fprintf (stderr, "ICE: bad arguments to print_unicode_stream\n");
+ return;
+ }
+
+ /* Allocate space for string_min 4-byte utf-8 characters. */
+ unsigned char * print_buf = xmalloc ((4 * string_min) + 1);
+ /* We should never have to put back more than 4 bytes. */
+ unsigned char putback_buf[5];
+ unsigned int num_putback = 0;
+
+ print_unicode_stream_body (filename, address, stream, putback_buf, num_putback, print_buf);
+ free (print_buf);
}
\f
/* Find the strings in file FILENAME, read from STREAM.
Assume that STREAM is positioned so that the next byte read
is at address ADDRESS in the file.
- Stop reading at address STOP_POINT in the file, if nonzero.
If STREAM is NULL, do not read from it.
The caller can supply a buffer of characters
static void
print_strings (const char *filename, FILE *stream, file_ptr address,
- int stop_point, int magiccount, char *magic)
+ int magiccount, char *magic)
{
+ if (unicode_display != unicode_default)
+ {
+ if (magic != NULL)
+ print_unicode_buffer (filename, address,
+ (const unsigned char *) magic, magiccount);
+
+ if (stream != NULL)
+ print_unicode_stream (filename, address, stream);
+ return;
+ }
+
char *buf = (char *) xmalloc (sizeof (char) * (string_min + 1));
while (1)
{
file_ptr start;
- int i;
+ unsigned int i;
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++)
{
c = get_char (stream, &address, &magiccount, &magic);
if (c == EOF)
- return;
+ {
+ free (buf);
+ return;
+ }
+
if (! STRING_ISGRAPHIC (c))
- /* Found a non-graphic. Try again starting with next char. */
- goto tryline;
+ {
+ /* Found a non-graphic. Try again starting with next byte. */
+ unget_part_char (c, &address, &magiccount, &magic);
+ goto tryline;
+ }
buf[i] = c;
}
/* We found a run of `string_min' graphic characters. Print up
to the next non-graphic character. */
-
- if (print_filenames)
- printf ("%s: ", filename);
- if (print_addresses)
- switch (address_radix)
- {
- case 8:
-#if __STDC_VERSION__ >= 199901L || (defined(__GNUC__) && __GNUC__ >= 2)
- 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:
-#if __STDC_VERSION__ >= 199901L || (defined(__GNUC__) && __GNUC__ >= 2)
- if (sizeof (start) > sizeof (long))
- {
-#ifndef __MSVCRT__
- printf ("%7lld ", (unsigned long long) start);
-#else
- printf ("%7I64d ", (unsigned long long) start);
-#endif
- }
- else
-#elif !BFD_HOST_64BIT_LONG
- if (start != (unsigned long) start)
- printf ("++%7ld ", (unsigned long) start);
- else
-#endif
- printf ("%7ld ", (long) start);
- break;
-
- case 16:
-#if __STDC_VERSION__ >= 199901L || (defined(__GNUC__) && __GNUC__ >= 2)
- 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;
- }
+ print_filename_and_address (filename, start);
buf[i] = '\0';
fputs (buf, stdout);
if (c == EOF)
break;
if (! STRING_ISGRAPHIC (c))
- break;
+ {
+ unget_part_char (c, &address, &magiccount, &magic);
+ break;
+ }
putchar (c);
}
- putchar ('\n');
+ if (output_separator)
+ fputs (output_separator, stdout);
+ else
+ putchar ('\n');
}
+ free (buf);
}
\f
static void
{
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\
+ 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\
- -<number> least [number] characters (default 4).\n\
+ -n <number> Locate & print any sequence of at least <number>\n\
+ --bytes=<number> displayable characters. (The default is 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=<BFDNAME> 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\
+ --unicode={default|show|invalid|hex|escape|highlight}\n\
+ -U {d|s|i|x|e|h} Specify how to treat UTF-8 encoded unicode characters\n\
+ -s --output-separator=<string> String used to separate strings in output.\n\
@<file> Read options from <file>\n\
-h --help Display this information\n\
-v -V --version Print the program's version number\n"));