xref: /PHP-5.4/ext/fileinfo/libmagic/names.h (revision 14ce4a1d)
1 /*
2  * Copyright (c) Ian F. Darwin 1986-1995.
3  * Software written by Ian F. Darwin and others;
4  * maintained 1995-present by Christos Zoulas and others.
5  *
6  * Redistribution and use in source and binary forms, with or without
7  * modification, are permitted provided that the following conditions
8  * are met:
9  * 1. Redistributions of source code must retain the above copyright
10  *    notice immediately at the beginning of the file, without modification,
11  *    this list of conditions, and the following disclaimer.
12  * 2. Redistributions in binary form must reproduce the above copyright
13  *    notice, this list of conditions and the following disclaimer in the
14  *    documentation and/or other materials provided with the distribution.
15  *
16  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
17  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19  * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
20  * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
26  * SUCH DAMAGE.
27  */
28 /*
29  * Names.h - names and types used by ascmagic in file(1).
30  * These tokens are here because they can appear anywhere in
31  * the first HOWMANY bytes, while tokens in MAGIC must
32  * appear at fixed offsets into the file. Don't make HOWMANY
33  * too high unless you have a very fast CPU.
34  *
35  * $File: names.h,v 1.32 2008/02/11 00:19:29 rrt Exp $
36  */
37 
38 /*
39 	modified by Chris Lowth - 9 April 2000
40 	to add mime type strings to the types table.
41 */
42 
43 /* these types are used to index the table 'types': keep em in sync! */
44 #define	L_C	0		/* first and foremost on UNIX */
45 #define	L_CC	1		/* Bjarne's postincrement */
46 #define	L_MAKE	2		/* Makefiles */
47 #define	L_PLI	3		/* PL/1 */
48 #define	L_MACH	4		/* some kinda assembler */
49 #define	L_ENG	5		/* English */
50 #define	L_PAS	6		/* Pascal */
51 #define	L_MAIL	7		/* Electronic mail */
52 #define	L_NEWS	8		/* Usenet Netnews */
53 #define	L_JAVA	9		/* Java code */
54 #define	L_HTML	10		/* HTML */
55 #define	L_BCPL	11		/* BCPL */
56 #define	L_M4	12		/* M4 */
57 #define	L_PO	13		/* PO */
58 
59 static const struct {
60 	char human[48];
61 	char mime[16];
62 } types[] = {
63 	{ "C program",					"text/x-c", },
64 	{ "C++ program",				"text/x-c++" },
65 	{ "make commands",				"text/x-makefile" },
66 	{ "PL/1 program",				"text/x-pl1" },
67 	{ "assembler program",				"text/x-asm" },
68 	{ "English",					"text/plain" },
69 	{ "Pascal program",				"text/x-pascal" },
70 	{ "mail",					"text/x-mail" },
71 	{ "news",					"text/x-news" },
72 	{ "Java program",				"text/x-java" },
73 	{ "HTML document",				"text/html", },
74 	{ "BCPL program",				"text/x-bcpl" },
75 	{ "M4 macro language pre-processor",		"text/x-m4" },
76 	{ "PO (gettext message catalogue)",             "text/x-po" },
77 	{ "cannot happen error on names.h/types",	"error/x-error" }
78 };
79 
80 /*
81  * XXX - how should we distinguish Java from C++?
82  * The trick used in a Debian snapshot, of having "extends" or "implements"
83  * as tags for Java, doesn't work very well, given that those keywords
84  * are often preceded by "class", which flags it as C++.
85  *
86  * Perhaps we need to be able to say
87  *
88  *	If "class" then
89  *
90  *		if "extends" or "implements" then
91  *			Java
92  *		else
93  *			C++
94  *	endif
95  *
96  * Or should we use other keywords, such as "package" or "import"?
97  * Unfortunately, Ada95 uses "package", and Modula-3 uses "import",
98  * although I infer from the language spec at
99  *
100  *	http://www.research.digital.com/SRC/m3defn/html/m3.html
101  *
102  * that Modula-3 uses "IMPORT" rather than "import", i.e. it must be
103  * in all caps.
104  *
105  * So, for now, we go with "import".  We must put it before the C++
106  * stuff, so that we don't misidentify Java as C++.  Not using "package"
107  * means we won't identify stuff that defines a package but imports
108  * nothing; hopefully, very little Java code imports nothing (one of the
109  * reasons for doing OO programming is to import as much as possible
110  * and write only what you need to, right?).
111  *
112  * Unfortunately, "import" may cause us to misidentify English text
113  * as Java, as it comes after "the" and "The".  Perhaps we need a fancier
114  * heuristic to identify Java?
115  */
116 static const struct names {
117 	char name[14];
118 	short type;
119 } names[] = {
120 	/* These must be sorted by eye for optimal hit rate */
121 	/* Add to this list only after substantial meditation */
122 	{"msgid",	L_PO},
123 	{"dnl",		L_M4},
124 	{"import",	L_JAVA},
125 	{"\"libhdr\"",	L_BCPL},
126 	{"\"LIBHDR\"",	L_BCPL},
127 	{"//",		L_CC},
128 	{"template",	L_CC},
129 	{"virtual",	L_CC},
130 	{"class",	L_CC},
131 	{"public:",	L_CC},
132 	{"private:",	L_CC},
133 	{"/*",		L_C},	/* must precede "The", "the", etc. */
134 	{"#include",	L_C},
135 	{"char",	L_C},
136 	{"The",		L_ENG},
137 	{"the",		L_ENG},
138 	{"double",	L_C},
139 	{"extern",	L_C},
140 	{"float",	L_C},
141 	{"struct",	L_C},
142 	{"union",	L_C},
143 	{"CFLAGS",	L_MAKE},
144 	{"LDFLAGS",	L_MAKE},
145 	{"all:",	L_MAKE},
146 	{".PRECIOUS",	L_MAKE},
147 	{".ascii",	L_MACH},
148 	{".asciiz",	L_MACH},
149 	{".byte",	L_MACH},
150 	{".even",	L_MACH},
151 	{".globl",	L_MACH},
152 	{".text",	L_MACH},
153 	{"clr",		L_MACH},
154 	{"(input,",	L_PAS},
155 	{"program",	L_PAS},
156 	{"record",	L_PAS},
157 	{"dcl",		L_PLI},
158 	{"Received:",	L_MAIL},
159 	{">From",	L_MAIL},
160 	{"Return-Path:",L_MAIL},
161 	{"Cc:",		L_MAIL},
162 	{"Newsgroups:",	L_NEWS},
163 	{"Path:",	L_NEWS},
164 	{"Organization:",L_NEWS},
165 	{"href=",	L_HTML},
166 	{"HREF=",	L_HTML},
167 	{"<body",	L_HTML},
168 	{"<BODY",	L_HTML},
169 	{"<html",	L_HTML},
170 	{"<HTML",	L_HTML},
171 	{"<!--",	L_HTML},
172 };
173 #define NNAMES (sizeof(names)/sizeof(struct names))
174