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 3 of the License, or
7 (at your option) any later version.
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, see <http://www.gnu.org/licenses/>.
18 /* Usage: frcode < sorted-list > compressed-list
20 Uses front compression (also known as incremental encoding);
21 see ";login:", March 1983, p. 8.
23 The input is a sorted list of NUL-terminated strings (or
24 newline-terminated if the -0 option is not given).
26 The output entries are in the same order as the input; each entry
27 consists of a signed offset-differential count byte (the additional
28 number of characters of prefix of the preceding entry to use beyond
29 the number that the preceding entry is using of its predecessor),
30 followed by a null-terminated ASCII remainder.
32 If the offset-differential count is larger than can be stored
33 in a byte (+/-127), the byte has the value LOCATEDB_ESCAPE
34 and the count follows in a 2-byte word, with the high byte first
39 Input, with NULs changed to newlines:
41 /usr/src/cmd/aardvark.c
42 /usr/src/cmd/armadillo.c
45 Length of the longest prefix of the preceding entry to share:
51 Output, with NULs changed to newlines and count bytes made printable:
58 (6 = 14 - 8, and -9 = 5 - 14)
60 Written by James A. Woods <jwoods@adobe.com>.
61 Modified by David MacKenzie <djm@gnu.org>.
62 Modified by James Youngman <jay@gnu.org>.
72 #include <sys/types.h>
75 #if defined HAVE_STRING_H || defined STDC_HEADERS
87 # define _(Text) gettext (Text)
90 #define textdomain(Domain)
91 #define bindtextdomain(Package, Directory)
94 # define N_(String) gettext_noop (String)
96 /* We used to use (String) instead of just String, but apparently ISO C
97 * doesn't allow this (at least, that's what HP said when someone reported
98 * this as a compiler bug). This is HP case number 1205608192. See
99 * also http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11250 (which references
100 * ANSI 3.5.7p14-15). The Intel icc compiler also rejects constructs
101 * like: static const char buf[] = ("string");
103 # define N_(String) String
107 #include "locatedb.h"
110 #include "closeout.h"
111 #include "findutils-version.h"
113 char *xmalloc
PARAMS((size_t));
115 /* The name this program was run with. */
118 /* Write out a 16-bit int, high byte first (network byte order).
119 * Return true iff all went well.
122 put_short (int c
, FILE *fp
)
124 /* XXX: The value of c may be negative. ANSI C 1989 (section 6.3.7)
125 * indicates that the result of shifting a negative value right is
126 * implementation defined.
128 assert (c
<= SHRT_MAX
);
129 assert (c
>= SHRT_MIN
);
130 return (putc (c
>> 8, fp
) != EOF
) && (putc (c
, fp
) != EOF
);
133 /* Return the length of the longest common prefix of strings S1 and S2. */
136 prefix_length (char *s1
, char *s2
)
138 register char *start
;
140 for (start
= s1
; *s1
== *s2
&& *s1
!= '\0'; s1
++, s2
++)
142 /* Don't emit a prefix length that will not fit into
151 static struct option
const longopts
[] =
153 {"help", no_argument
, NULL
, 'h'},
154 {"version", no_argument
, NULL
, 'v'},
155 {"null", no_argument
, NULL
, '0'},
156 {NULL
, no_argument
, NULL
, 0}
159 extern char *version_string
;
161 /* The name this program was run with. */
169 _("Usage: %s [-0 | --null] [--version] [--help]\n"),
171 fputs (_("\nReport bugs to <bug-findutils@gnu.org>.\n"), stream
);
175 get_seclevel(char *s
)
180 /* Reset errno in oreder to be able to distinguish LONG_MAX/LONG_MIN
181 * from values whichare actually out of range
185 result
= strtol(s
, &p
, 10);
186 if ((0==result
) && (p
== optarg
))
188 error(1, 0, _("You need to specify a security level as a decimal integer."));
192 else if ((LONG_MIN
==result
|| LONG_MAX
==result
) && errno
)
195 error(1, 0, _("Security level %s is outside the convertible range."), s
);
201 /* Some suffix exists */
202 error(1, 0, _("Security level %s has unexpected suffix %s."), s
, p
);
215 /* Issue the same error message as closeout() would. */
216 error(1, errno
, _("write error"));
220 main (int argc
, char **argv
)
222 char *path
; /* The current input entry. */
223 char *oldpath
; /* The previous input entry. */
224 size_t pathsize
, oldpathsize
; /* Amounts allocated for them. */
225 int count
, oldcount
, diffcount
; /* Their prefix lengths & the difference. */
226 int line_len
; /* Length of input line. */
227 int delimiter
= '\n';
229 int slocate_compat
= 0;
230 long slocate_seclevel
= 0L;
232 program_name
= argv
[0];
234 program_name
= "frcode";
235 atexit (close_stdout
);
237 pathsize
= oldpathsize
= 1026; /* Increased as necessary by getline. */
238 path
= xmalloc (pathsize
);
239 oldpath
= xmalloc (oldpathsize
);
245 while ((optc
= getopt_long (argc
, argv
, "hv0S:", longopts
, (int *) 0)) != -1)
254 slocate_seclevel
= get_seclevel(optarg
);
255 if (slocate_seclevel
< 0 || slocate_seclevel
> 1)
258 _("slocate security level %ld is unsupported."),
268 display_findutils_version("frcode");
276 /* We expect to have no arguments. */
286 fputc(slocate_seclevel
? '1' : '0', stdout
);
292 /* GNU LOCATE02 format */
293 if (fwrite (LOCATEDB_MAGIC
, 1, sizeof (LOCATEDB_MAGIC
), stdout
)
294 != sizeof(LOCATEDB_MAGIC
))
296 error(1, errno
, _("Failed to write to standard output"));
301 while ((line_len
= getdelim (&path
, &pathsize
, delimiter
, stdin
)) > 0)
303 path
[line_len
- 1] = '\0'; /* FIXME temporary: nuke the newline. */
305 count
= prefix_length (oldpath
, path
);
306 diffcount
= count
- oldcount
;
307 if ( (diffcount
> SHRT_MAX
) || (diffcount
< SHRT_MIN
) )
309 /* We do this to prevent overflow of the value we
310 * write with put_short()
313 diffcount
= (-oldcount
);
319 /* Emit no count for the first pathname. */
324 /* If the difference is small, it fits in one byte;
325 otherwise, two bytes plus a marker noting that fact. */
326 if (diffcount
< LOCATEDB_ONEBYTE_MIN
327 || diffcount
> LOCATEDB_ONEBYTE_MAX
)
329 if (EOF
== putc (LOCATEDB_ESCAPE
, stdout
))
331 if (!put_short (diffcount
, stdout
))
336 if (EOF
== putc (diffcount
, stdout
))
341 if ( (EOF
== fputs (path
+ count
, stdout
))
342 || (EOF
== putc ('\0', stdout
)))
349 /* Swap path and oldpath and their sizes. */
350 char *tmppath
= oldpath
;
351 size_t tmppathsize
= oldpathsize
;
353 oldpathsize
= pathsize
;
355 pathsize
= tmppathsize
;