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>.
74 #include <sys/types.h>
77 #if defined(HAVE_STRING_H) || defined(STDC_HEADERS)
89 # define _(Text) gettext (Text)
92 #define textdomain(Domain)
93 #define bindtextdomain(Package, Directory)
96 # define N_(String) gettext_noop (String)
98 /* We used to use (String) instead of just String, but apparently ISO C
99 * doesn't allow this (at least, that's what HP said when someone reported
100 * this as a compiler bug). This is HP case number 1205608192. See
101 * also http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11250 (which references
102 * ANSI 3.5.7p14-15). The Intel icc compiler also rejects constructs
103 * like: static const char buf[] = ("string");
105 # define N_(String) String
109 #include "locatedb.h"
113 #include "closeout.h"
115 char *xmalloc
PARAMS((size_t));
117 /* The name this program was run with. */
120 /* Write out a 16-bit int, high byte first (network byte order).
121 * Return true iff all went well.
124 put_short (int c
, FILE *fp
)
126 /* XXX: The value of c may be negative. ANSI C 1989 (section 6.3.7)
127 * indicates that the result of shifting a negative value right is
128 * implementation defined.
130 assert(c
<= SHRT_MAX
);
131 assert(c
>= SHRT_MIN
);
132 return (putc (c
>> 8, fp
) != EOF
) && (putc (c
, fp
) != EOF
);
135 /* Return the length of the longest common prefix of strings S1 and S2. */
138 prefix_length (char *s1
, char *s2
)
140 register char *start
;
142 for (start
= s1
; *s1
== *s2
&& *s1
!= '\0'; s1
++, s2
++)
144 /* Don't emit a prefix length that will not fit into
153 static struct option
const longopts
[] =
155 {"help", no_argument
, NULL
, 'h'},
156 {"version", no_argument
, NULL
, 'v'},
157 {"null", no_argument
, NULL
, '0'},
158 {NULL
, no_argument
, NULL
, 0}
161 extern char *version_string
;
163 /* The name this program was run with. */
171 _("Usage: %s [-0 | --null] [--version] [--help]\n"),
173 fputs (_("\nReport bugs to <bug-findutils@gnu.org>.\n"), stream
);
177 get_seclevel(char *s
)
182 /* Reset errno in oreder to be able to distinguish LONG_MAX/LONG_MIN
183 * from values whichare actually out of range
187 result
= strtol(s
, &p
, 10);
188 if ((0==result
) && (p
== optarg
))
190 error(1, 0, _("You need to specify a security level as a decimal integer."));
194 else if ((LONG_MIN
==result
|| LONG_MAX
==result
) && errno
)
197 error(1, 0, _("Security level %s is outside the convertible range."), s
);
203 /* Some suffix exists */
204 error(1, 0, _("Security level %s has unexpected suffix %s."), s
, p
);
217 /* Issue the same error message as closeout() would. */
218 error(1, errno
, _("write error"));
222 main (int argc
, char **argv
)
224 char *path
; /* The current input entry. */
225 char *oldpath
; /* The previous input entry. */
226 size_t pathsize
, oldpathsize
; /* Amounts allocated for them. */
227 int count
, oldcount
, diffcount
; /* Their prefix lengths & the difference. */
228 int line_len
; /* Length of input line. */
229 int delimiter
= '\n';
231 int slocate_compat
= 0;
232 long slocate_seclevel
= 0L;
234 program_name
= argv
[0];
236 program_name
= "frcode";
237 atexit (close_stdout
);
239 pathsize
= oldpathsize
= 1026; /* Increased as necessary by getline. */
240 path
= xmalloc (pathsize
);
241 oldpath
= xmalloc (oldpathsize
);
247 while ((optc
= getopt_long (argc
, argv
, "hv0S:", longopts
, (int *) 0)) != -1)
256 slocate_seclevel
= get_seclevel(optarg
);
257 if (slocate_seclevel
< 0 || slocate_seclevel
> 1)
260 _("slocate security level %ld is unsupported."),
270 printf (_("GNU locate version %s\n"), version_string
);
278 /* We expect to have no arguments. */
288 fputc(slocate_seclevel
? '1' : '0', stdout
);
294 /* GNU LOCATE02 format */
295 if (fwrite (LOCATEDB_MAGIC
, 1, sizeof (LOCATEDB_MAGIC
), stdout
)
296 != sizeof(LOCATEDB_MAGIC
))
298 error(1, errno
, _("Failed to write to standard output"));
303 while ((line_len
= getdelim (&path
, &pathsize
, delimiter
, stdin
)) > 0)
305 path
[line_len
- 1] = '\0'; /* FIXME temporary: nuke the newline. */
307 count
= prefix_length (oldpath
, path
);
308 diffcount
= count
- oldcount
;
309 if ( (diffcount
> SHRT_MAX
) || (diffcount
< SHRT_MIN
) )
311 /* We do this to prevent overflow of the value we
312 * write with put_short()
315 diffcount
= (-oldcount
);
321 /* Emit no count for the first pathname. */
326 /* If the difference is small, it fits in one byte;
327 otherwise, two bytes plus a marker noting that fact. */
328 if (diffcount
< LOCATEDB_ONEBYTE_MIN
329 || diffcount
> LOCATEDB_ONEBYTE_MAX
)
331 if (EOF
== putc (LOCATEDB_ESCAPE
, stdout
))
333 if (!put_short (diffcount
, stdout
))
338 if (EOF
== putc (diffcount
, stdout
))
343 if ( (EOF
== fputs (path
+ count
, stdout
))
344 || (EOF
== putc ('\0', stdout
)))
351 /* Swap path and oldpath and their sizes. */
352 char *tmppath
= oldpath
;
353 size_t tmppathsize
= oldpathsize
;
355 oldpathsize
= pathsize
;
357 pathsize
= tmppathsize
;