1 /* frcode -- front-compress a sorted list
2 Copyright (C) 1994,2005,2006,2007 Free Software Foundation, Inc.
4 This program is free software; you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation; either version 2, or (at your option)
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program; if not, write to the Free Software
16 Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,
20 /* Usage: frcode < sorted-list > compressed-list
22 Uses front compression (also known as incremental encoding);
23 see ";login:", March 1983, p. 8.
25 The input is a sorted list of NUL-terminated strings (or
26 newline-terminated if the -0 option is not given).
28 The output entries are in the same order as the input; each entry
29 consists of a signed offset-differential count byte (the additional
30 number of characters of prefix of the preceding entry to use beyond
31 the number that the preceding entry is using of its predecessor),
32 followed by a null-terminated ASCII remainder.
34 If the offset-differential count is larger than can be stored
35 in a byte (+/-127), the byte has the value LOCATEDB_ESCAPE
36 and the count follows in a 2-byte word, with the high byte first
41 Input, with NULs changed to newlines:
43 /usr/src/cmd/aardvark.c
44 /usr/src/cmd/armadillo.c
47 Length of the longest prefix of the preceding entry to share:
53 Output, with NULs changed to newlines and count bytes made printable:
60 (6 = 14 - 8, and -9 = 5 - 14)
62 Written by James A. Woods <jwoods@adobe.com>.
63 Modified by David MacKenzie <djm@gnu.org>.
64 Modified by James Youngman <jay@gnu.org>.
72 #include <sys/types.h>
74 #if defined(HAVE_STRING_H) || defined(STDC_HEADERS)
86 # define _(Text) gettext (Text)
89 #define textdomain(Domain)
90 #define bindtextdomain(Package, Directory)
93 # define N_(String) gettext_noop (String)
95 /* We used to use (String) instead of just String, but apparently ISO C
96 * doesn't allow this (at least, that's what HP said when someone reported
97 * this as a compiler bug). This is HP case number 1205608192. See
98 * also http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11250 (which references
99 * ANSI 3.5.7p14-15). The Intel icc compiler also rejects constructs
100 * like: static const char buf[] = ("string");
102 # define N_(String) String
106 #include "locatedb.h"
110 #include "closeout.h"
112 char *xmalloc
PARAMS((size_t));
114 /* The name this program was run with. */
117 /* Write out a 16-bit int, high byte first (network byte order).
118 * Return true iff all went well.
121 put_short (int c
, FILE *fp
)
123 /* XXX: The value of c may be negative. ANSI C 1989 (section 6.3.7)
124 * indicates that the result of shifting a negative value right is
125 * implementation defined.
127 assert(c
<= SHRT_MAX
);
128 assert(c
>= SHRT_MIN
);
129 return (putc (c
>> 8, fp
) != EOF
) && (putc (c
, fp
) != EOF
);
132 /* Return the length of the longest common prefix of strings S1 and S2. */
135 prefix_length (char *s1
, char *s2
)
137 register char *start
;
139 for (start
= s1
; *s1
== *s2
&& *s1
!= '\0'; s1
++, s2
++)
141 /* Don't emit a prefix length that will not fit into
150 static struct option
const longopts
[] =
152 {"help", no_argument
, NULL
, 'h'},
153 {"version", no_argument
, NULL
, 'v'},
154 {"null", no_argument
, NULL
, '0'},
155 {NULL
, no_argument
, NULL
, 0}
158 extern char *version_string
;
160 /* The name this program was run with. */
168 _("Usage: %s [-0 | --null] [--version] [--help]\n"),
170 fputs (_("\nReport bugs to <bug-findutils@gnu.org>.\n"), stream
);
174 get_seclevel(char *s
)
179 /* Reset errno in oreder to be able to distinguish LONG_MAX/LONG_MIN
180 * from values whichare actually out of range
184 result
= strtol(s
, &p
, 10);
185 if ((0==result
) && (p
== optarg
))
187 error(1, 0, _("You need to specify a security level as a decimal integer."));
191 else if ((LONG_MIN
==result
|| LONG_MAX
==result
) && errno
)
194 error(1, 0, _("Security level %s is outside the convertible range."), s
);
200 /* Some suffix exists */
201 error(1, 0, _("Security level %s has unexpected suffix %s."), s
, p
);
214 /* Issue the same error message as closeout() would. */
215 error(1, errno
, _("write error"));
219 main (int argc
, char **argv
)
221 char *path
; /* The current input entry. */
222 char *oldpath
; /* The previous input entry. */
223 size_t pathsize
, oldpathsize
; /* Amounts allocated for them. */
224 int count
, oldcount
, diffcount
; /* Their prefix lengths & the difference. */
225 int line_len
; /* Length of input line. */
226 int delimiter
= '\n';
228 int slocate_compat
= 0;
229 long slocate_seclevel
= 0L;
231 program_name
= argv
[0];
233 program_name
= "frcode";
234 atexit (close_stdout
);
236 pathsize
= oldpathsize
= 1026; /* Increased as necessary by getline. */
237 path
= xmalloc (pathsize
);
238 oldpath
= xmalloc (oldpathsize
);
244 while ((optc
= getopt_long (argc
, argv
, "hv0S:", longopts
, (int *) 0)) != -1)
253 slocate_seclevel
= get_seclevel(optarg
);
254 if (slocate_seclevel
< 0 || slocate_seclevel
> 1)
257 _("slocate security level %ld is unsupported."),
267 printf (_("GNU locate version %s\n"), version_string
);
275 /* We expect to have no arguments. */
285 fputc(slocate_seclevel
? '1' : '0', stdout
);
291 /* GNU LOCATE02 format */
292 if (fwrite (LOCATEDB_MAGIC
, 1, sizeof (LOCATEDB_MAGIC
), stdout
)
293 != sizeof(LOCATEDB_MAGIC
))
295 error(1, errno
, _("Failed to write to standard output"));
300 while ((line_len
= getdelim (&path
, &pathsize
, delimiter
, stdin
)) > 0)
302 path
[line_len
- 1] = '\0'; /* FIXME temporary: nuke the newline. */
304 count
= prefix_length (oldpath
, path
);
305 diffcount
= count
- oldcount
;
306 if ( (diffcount
> SHRT_MAX
) || (diffcount
< SHRT_MIN
) )
308 /* We do this to prevent overflow of the value we
309 * write with put_short()
312 diffcount
= (-oldcount
);
318 /* Emit no count for the first pathname. */
323 /* If the difference is small, it fits in one byte;
324 otherwise, two bytes plus a marker noting that fact. */
325 if (diffcount
< LOCATEDB_ONEBYTE_MIN
326 || diffcount
> LOCATEDB_ONEBYTE_MAX
)
328 if (EOF
== putc (LOCATEDB_ESCAPE
, stdout
))
330 if (!put_short (diffcount
, stdout
))
335 if (EOF
== putc (diffcount
, stdout
))
340 if ( (EOF
== fputs (path
+ count
, stdout
))
341 || (EOF
== putc ('\0', stdout
)))
348 /* Swap path and oldpath and their sizes. */
349 char *tmppath
= oldpath
;
350 size_t tmppathsize
= oldpathsize
;
352 oldpathsize
= pathsize
;
354 pathsize
= tmppathsize
;