file/ascmagic.c

   1 /*
   2  * Copyright (c) 1999 Apple Computer, Inc. All rights reserved.
   3  *
   4  * @APPLE_LICENSE_HEADER_START@
   5  *
   6  * Portions Copyright (c) 1999 Apple Computer, Inc.  All Rights
   7  * Reserved.  This file contains Original Code and/or Modifications of
   8  * Original Code as defined in and that are subject to the Apple Public
   9  * Source License Version 1.1 (the "License").  You may not use this file
  10  * except in compliance with the License.  Please obtain a copy of the
  11  * License at http://www.apple.com/publicsource and read it before using
  12  * this file.
  13  *
  14  * The Original Code and all software distributed under the License are
  15  * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  16  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  17  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  18  * FITNESS FOR A PARTICULAR PURPOSE OR NON- INFRINGEMENT.  Please see the
  19  * License for the specific language governing rights and limitations
  20  * under the License.
  21  *
  22  * @APPLE_LICENSE_HEADER_END@
  23  */
  24 /*      $OpenBSD: ascmagic.c,v 1.3 1997/02/09 23:58:18 millert Exp $    */
  25
  26 /*
  27  * ASCII magic -- file types that we know based on keywords
  28  * that can appear anywhere in the file.
  29  *
  30  * Copyright (c) Ian F. Darwin, 1987.
  31  * Written by Ian F. Darwin.
  32  *
  33  * This software is not subject to any license of the American Telephone
  34  * and Telegraph Company or of the Regents of the University of California.
  35  *
  36  * Permission is granted to anyone to use this software for any purpose on
  37  * any computer system, and to alter it and redistribute it freely, subject
  38  * to the following restrictions:
  39  *
  40  * 1. The author is not responsible for the consequences of use of this
  41  *    software, no matter how awful, even if they arise from flaws in it.
  42  *
  43  * 2. The origin of this software must not be misrepresented, either by
  44  *    explicit claim or by omission.  Since few users ever read sources,
  45  *    credits must appear in the documentation.
  46  *
  47  * 3. Altered versions must be plainly marked as such, and must not be
  48  *    misrepresented as being the original software.  Since few users
  49  *    ever read sources, credits must appear in the documentation.
  50  *
  51  * 4. This notice may not be removed or altered.
  52  */
  53
  54 #include <stdio.h>
  55 #include <string.h>
  56 #include <ctype.h>
  57 #include <stdlib.h>
  58 #include <unistd.h>
  59 #include "file.h"
  60 #include "names.h"
  61
  62 #ifndef lint
  63 #if 0
  64 static char *moduleid = "$OpenBSD: ascmagic.c,v 1.3 1997/02/09 23:58:18 millert Exp $";
  65 #endif
  66 #endif  /* lint */
  67
  68                         /* an optimisation over plain strcmp() */
  69 #define STREQ(a, b)     (*(a) == *(b) && strcmp((a), (b)) == 0)
  70
  71 int
  72 ascmagic(buf, nbytes)
  73 unsigned char *buf;
  74 int nbytes;     /* size actually read */
  75 {
  76         int i, has_escapes = 0;
  77         unsigned char *s;
  78         char nbuf[HOWMANY+1];   /* one extra for terminating '\0' */
  79         char *token;
  80         register struct names *p;
  81
  82         /*
  83          * Do the tar test first, because if the first file in the tar
  84          * archive starts with a dot, we can confuse it with an nroff file.
  85          */
  86         switch (is_tar(buf, nbytes)) {
  87         case 1:
  88                 ckfputs("tar archive", stdout);
  89                 return 1;
  90         case 2:
  91                 ckfputs("POSIX tar archive", stdout);
  92                 return 1;
  93         }
  94
  95         /*
  96          * for troff, look for . + letter + letter or .\";
  97          * this must be done to disambiguate tar archives' ./file
  98          * and other trash from real troff input.
  99          */
 100         if (*buf == '.') {
 101                 unsigned char *tp = buf + 1;
 102
 103                 while (isascii(*tp) && isspace(*tp))
 104                         ++tp;   /* skip leading whitespace */
 105                 if ((isascii(*tp) && (isalnum(*tp) || *tp=='\\') &&
 106                     isascii(tp[1]) && (isalnum(tp[1]) || tp[1] == '"'))) {
 107                         ckfputs("troff or preprocessor input text", stdout);
 108                         return 1;
 109                 }
 110         }
 111         if ((*buf == 'c' || *buf == 'C') &&
 112             isascii(buf[1]) && isspace(buf[1])) {
 113                 ckfputs("fortran program text", stdout);
 114                 return 1;
 115         }
 116
 117
 118         /* Make sure we are dealing with ascii text before looking for tokens */
 119         for (i = 0; i < nbytes; i++) {
 120                 if (!isascii(buf[i]))
 121                         return 0;       /* not all ASCII */
 122         }
 123
 124         /* look for tokens from names.h - this is expensive! */
 125         /* make a copy of the buffer here because strtok() will destroy it */
 126         s = (unsigned char*) memcpy(nbuf, buf, nbytes);
 127         s[nbytes] = '\0';
 128         has_escapes = (memchr(s, '\033', nbytes) != NULL);
 129         while ((token = strtok((char *) s, " \t\n\r\f")) != NULL) {
 130                 s = NULL;       /* make strtok() keep on tokin' */
 131                 for (p = names; p < names + NNAMES; p++) {
 132                         if (STREQ(p->name, token)) {
 133                                 ckfputs(types[p->type], stdout);
 134                                 if (has_escapes)
 135                                         ckfputs(" (with escape sequences)",
 136                                                 stdout);
 137                                 return 1;
 138                         }
 139                 }
 140         }
 141
 142         /* all else fails, but it is ASCII... */
 143         ckfputs("ASCII text", stdout);
 144         if (has_escapes) {
 145                 ckfputs(" (with escape sequences)", stdout);
 146         }
 147         return 1;
 148 }
 149
 150