summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorHartmut Holzgraefe <hholzgra@php.net>2002-03-24 23:00:47 +0000
committerHartmut Holzgraefe <hholzgra@php.net>2002-03-24 23:00:47 +0000
commit2c413f116aaf83eed98ba0c303e138d86a17633c (patch)
tree4423f6671b0a78075899b8ba4b9e7710f9639cde
parent0f3b3045062bc392700f1d630228917679106c6a (diff)
downloadphp-git-2c413f116aaf83eed98ba0c303e138d86a17633c.tar.gz
1st implementation of magic-file based mime_content_type(string filename)
-rw-r--r--ext/mime_magic/CREDITS2
-rw-r--r--ext/mime_magic/EXPERIMENTAL0
-rw-r--r--ext/mime_magic/config.m411
-rw-r--r--ext/mime_magic/mime_magic.c2120
-rw-r--r--ext/mime_magic/php_mime_magic.h454
5 files changed, 2587 insertions, 0 deletions
diff --git a/ext/mime_magic/CREDITS b/ext/mime_magic/CREDITS
new file mode 100644
index 0000000000..ce54609ecc
--- /dev/null
+++ b/ext/mime_magic/CREDITS
@@ -0,0 +1,2 @@
+mime_magic
+Hartmut Holzgraefe
diff --git a/ext/mime_magic/EXPERIMENTAL b/ext/mime_magic/EXPERIMENTAL
new file mode 100644
index 0000000000..e69de29bb2
--- /dev/null
+++ b/ext/mime_magic/EXPERIMENTAL
diff --git a/ext/mime_magic/config.m4 b/ext/mime_magic/config.m4
new file mode 100644
index 0000000000..21e1c88a1e
--- /dev/null
+++ b/ext/mime_magic/config.m4
@@ -0,0 +1,11 @@
+dnl $Id$
+dnl config.m4 for extension mime_magic
+
+PHP_ARG_ENABLE(mime_magic, whether to enable mime_magic support,
+[ --enable-mime_magic Enable mime_magic support])
+
+if test "$PHP_MIME_MAGIC" != "no"; then
+ dnl PHP_SUBST(MIME_MAGIC_SHARED_LIBADD)
+
+ PHP_NEW_EXTENSION(mime_magic, mime_magic.c, $ext_shared)
+fi
diff --git a/ext/mime_magic/mime_magic.c b/ext/mime_magic/mime_magic.c
new file mode 100644
index 0000000000..07af88ec07
--- /dev/null
+++ b/ext/mime_magic/mime_magic.c
@@ -0,0 +1,2120 @@
+/*
+ +----------------------------------------------------------------------+
+ | PHP Version 4 |
+ +----------------------------------------------------------------------+
+ | Copyright (c) 1997-2002 The PHP Group |
+ +----------------------------------------------------------------------+
+ | This source file is subject to version 2.02 of the PHP license, |
+ | that is bundled with this package in the file LICENSE, and is |
+ | available at through the world-wide-web at |
+ | http://www.php.net/license/2_02.txt. |
+ | If you did not receive a copy of the PHP license and are unable to |
+ | obtain it through the world-wide-web, please send a note to |
+ | license@php.net so we can mail you a copy immediately. |
+ +----------------------------------------------------------------------+
+ | Author: Hartmut Holzgraefe |
+ +----------------------------------------------------------------------+
+
+ $Id$
+
+ This module contains a lot of stuff taken from Apache mod_mime_magic,
+ so the lincense section is a little bit longer than usual:
+
+ ====================================================================
+ * The Apache Software License, Version 1.1
+ *
+ * Copyright (c) 2000 The Apache Software Foundation. All rights
+ * reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ *
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ *
+ * 3. The end-user documentation included with the redistribution,
+ * if any, must include the following acknowledgment:
+ * "This product includes software developed by the
+ * Apache Software Foundation (http://www.apache.org/)."
+ * Alternately, this acknowledgment may appear in the software itself,
+ * if and wherever such third-party acknowledgments normally appear.
+ *
+ * 4. The names "Apache" and "Apache Software Foundation" must
+ * not be used to endorse or promote products derived from this
+ * software without prior written permission. For written
+ * permission, please contact apache@apache.org.
+ *
+ * 5. Products derived from this software may not be called "Apache",
+ * nor may "Apache" appear in their name, without prior written
+ * permission of the Apache Software Foundation.
+ *
+ * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
+ * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
+ * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
+ * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+ * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
+ * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ * ====================================================================
+ *
+ * This software consists of voluntary contributions made by many
+ * individuals on behalf of the Apache Software Foundation. For more
+ * information on the Apache Software Foundation, please see
+ * <http://www.apache.org/>.
+ *
+ * Portions of this software are based upon public domain software
+ * originally written at the National Center for Supercomputing Applications,
+ * University of Illinois, Urbana-Champaign.
+ */
+
+/*
+ * mod_mime_magic: MIME type lookup via file magic numbers
+ * Copyright (c) 1996-1997 Cisco Systems, Inc.
+ *
+ * This software was submitted by Cisco Systems to the Apache Group in July
+ * 1997. Future revisions and derivatives of this source code must
+ * acknowledge Cisco Systems as the original contributor of this module.
+ * All other licensing and usage conditions are those of the Apache Group.
+ *
+ * Some of this code is derived from the free version of the file command
+ * originally posted to comp.sources.unix. Copyright info for that program
+ * is included below as required.
+ * ---------------------------------------------------------------------------
+ * - Copyright (c) Ian F. Darwin, 1987. Written by Ian F. Darwin.
+ *
+ * This software is not subject to any license of the American Telephone and
+ * Telegraph Company or of the Regents of the University of California.
+ *
+ * Permission is granted to anyone to use this software for any purpose on any
+ * computer system, and to alter it and redistribute it freely, subject to
+ * the following restrictions:
+ *
+ * 1. The author is not responsible for the consequences of use of this
+ * software, no matter how awful, even if they arise from flaws in it.
+ *
+ * 2. The origin of this software must not be misrepresented, either by
+ * explicit claim or by omission. Since few users ever read sources, credits
+ * must appear in the documentation.
+ *
+ * 3. Altered versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software. Since few users ever read
+ * sources, credits must appear in the documentation.
+ *
+ * 4. This notice may not be removed or altered.
+ * -------------------------------------------------------------------------
+ *
+ * For compliance with Mr Darwin's terms: this has been very significantly
+ * modified from the free "file" command.
+ * - all-in-one file for compilation convenience when moving from one
+ * version of Apache to the next.
+ * - Memory allocation is done through the Apache API's pool structure.
+ * - All functions have had necessary Apache API request or server
+ * structures passed to them where necessary to call other Apache API
+ * routines. (i.e. usually for logging, files, or memory allocation in
+ * itself or a called function.)
+ * - struct magic has been converted from an array to a single-ended linked
+ * list because it only grows one record at a time, it's only accessed
+ * sequentially, and the Apache API has no equivalent of realloc().
+ * - Functions have been changed to get their parameters from the server
+ * configuration instead of globals. (It should be reentrant now but has
+ * not been tested in a threaded environment.)
+ * - Places where it used to print results to stdout now saves them in a
+ * list where they're used to set the MIME type in the Apache request
+ * record.
+ * - Command-line flags have been removed since they will never be used here.
+ *
+ * Ian Kluft <ikluft@cisco.com>
+ * Engineering Information Framework
+ * Central Engineering
+ * Cisco Systems, Inc.
+ * San Jose, CA, USA
+ *
+ * Initial installation July/August 1996
+ * Misc bug fixes May 1997
+ * Submission to Apache Group July 1997
+ *
+ */
+
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php.h"
+#include "php_ini.h"
+#include "ext/standard/info.h"
+#include "php_mime_magic.h"
+
+#include <fcntl.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+#define MODNAME "mime_magic"
+
+ZEND_DECLARE_MODULE_GLOBALS(mime_magic)
+
+
+/* True global resources - no need for thread safety here */
+/* static int le_mime_magic; */
+static magic_server_config_rec mime_global;
+
+/* {{{ mime_magic_functions[]
+ *
+ * Every user visible function must have an entry in mime_magic_functions[].
+ */
+function_entry mime_magic_functions[] = {
+ PHP_FE(mime_content_type, NULL) /* For testing, remove later. */
+ {NULL, NULL, NULL} /* Must be the last line in mime_magic_functions[] */
+};
+/* }}} */
+
+/* {{{ mime_magic_module_entry
+ */
+zend_module_entry mime_magic_module_entry = {
+#if ZEND_MODULE_API_NO >= 20010901
+ STANDARD_MODULE_HEADER,
+#endif
+ "mime_magic",
+ mime_magic_functions,
+ PHP_MINIT(mime_magic),
+ PHP_MSHUTDOWN(mime_magic),
+ PHP_RINIT(mime_magic), /* Replace with NULL if there's nothing to do at request start */
+ PHP_RSHUTDOWN(mime_magic), /* Replace with NULL if there's nothing to do at request end */
+ PHP_MINFO(mime_magic),
+#if ZEND_MODULE_API_NO >= 20010901
+ "0.1", /* Replace with version number for your extension */
+#endif
+ STANDARD_MODULE_PROPERTIES
+};
+/* }}} */
+
+#ifdef COMPILE_DL_MIME_MAGIC
+ZEND_GET_MODULE(mime_magic)
+#endif
+
+/* {{{ PHP_INI
+ */
+PHP_INI_BEGIN()
+ STD_PHP_INI_ENTRY("mime_magic.magicfile", "/usr/share/misc/magic.mime", PHP_INI_SYSTEM, OnUpdateString, magicfile, zend_mime_magic_globals, mime_magic_globals)
+PHP_INI_END()
+/* }}} */
+
+/* {{{ php_mime_magic_init_globals
+ */
+static void php_mime_magic_init_globals(zend_mime_magic_globals *mime_magic_globals)
+{
+ mime_magic_globals->magicfile = NULL;
+}
+/* }}} */
+
+/* {{{ PHP_MINIT_FUNCTION
+ */
+PHP_MINIT_FUNCTION(mime_magic)
+{
+ ZEND_INIT_MODULE_GLOBALS(mime_magic, php_mime_magic_init_globals, NULL);
+ REGISTER_INI_ENTRIES();
+
+ if(mime_magic_globals.magicfile) {
+ mime_global.magicfile = mime_magic_globals.magicfile;
+ apprentice();
+ }
+
+ return SUCCESS;
+}
+/* }}} */
+
+/* {{{ PHP_MSHUTDOWN_FUNCTION
+ */
+PHP_MSHUTDOWN_FUNCTION(mime_magic)
+{
+ UNREGISTER_INI_ENTRIES();
+ return SUCCESS;
+}
+/* }}} */
+
+/* Remove if there's nothing to do at request start */
+/* {{{ PHP_RINIT_FUNCTION
+ */
+PHP_RINIT_FUNCTION(mime_magic)
+{
+ return SUCCESS;
+}
+/* }}} */
+
+/* Remove if there's nothing to do at request end */
+/* {{{ PHP_RSHUTDOWN_FUNCTION
+ */
+PHP_RSHUTDOWN_FUNCTION(mime_magic)
+{
+ return SUCCESS;
+}
+/* }}} */
+
+/* {{{ PHP_MINFO_FUNCTION
+ */
+PHP_MINFO_FUNCTION(mime_magic)
+{
+ php_info_print_table_start();
+ php_info_print_table_header(2, "mime_magic support", "enabled");
+ php_info_print_table_end();
+
+ /* Remove comments if you have entries in php.ini
+ DISPLAY_INI_ENTRIES();
+ */
+}
+/* }}} */
+
+
+/* Remove the following function when you have succesfully modified config.m4
+ so that your module can be compiled into PHP, it exists only for testing
+ purposes. */
+
+/* {{{ proto string mime_content_type(string filename)
+ Return content-type for file */
+PHP_FUNCTION(mime_content_type)
+{
+ char *filename = NULL;
+ int filename_len;
+ magic_server_config_rec *conf = &mime_global;
+ char *content_type=NULL, *content_encoding=NULL;
+
+ if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s", &filename, &filename_len) == FAILURE) {
+ return;
+ }
+
+ if(!conf->magic) {
+ php_error(E_WARNING, MODNAME " not initialized");
+ RETURN_FALSE;
+ }
+
+ magic_set_config();
+
+ if(OK != magic_process(filename)) {
+ RETVAL_FALSE;
+ } else if(OK != magic_rsl_get(&content_type, &content_encoding)) {
+ RETVAL_FALSE;
+ } else {
+ RETVAL_STRING(content_type, 1);
+ }
+
+ if(content_type) efree(content_type);
+ if(content_encoding) efree(content_encoding);
+
+ magic_free_config(MIME_MAGIC_G(req_dat));
+}
+/* }}} */
+
+#define EATAB {while (isspace((unsigned char) *l)) ++l;}
+
+/*
+ * apprentice - load configuration from the magic file r
+ * API request record
+ */
+static int apprentice(void)
+{
+ FILE *f;
+ char line[BUFSIZ + 1];
+ int errs = 0;
+ int lineno;
+#if MIME_MAGIC_DEBUG
+ int rule = 0;
+ struct magic *m, *prevm;
+#endif
+ char *fname;
+ magic_server_config_rec *conf = &mime_global;
+
+ fname = conf->magicfile; // todo cwd?
+ f = fopen(fname, "r");
+ if (f == NULL) {
+ php_error(E_WARNING,
+ MODNAME ": can't read magic file %s", fname);
+ return -1;
+ }
+
+ /* set up the magic list (empty) */
+ conf->magic = conf->last = NULL;
+
+ /* parse it */
+ for (lineno = 1; fgets(line, BUFSIZ, f) != NULL; lineno++) {
+ int ws_offset;
+
+ /* delete newline */
+ if (line[0]) {
+ line[strlen(line) - 1] = '\0';
+ }
+
+ /* skip leading whitespace */
+ ws_offset = 0;
+ while (line[ws_offset] && isspace(line[ws_offset])) {
+ ws_offset++;
+ }
+
+ /* skip blank lines */
+ if (line[ws_offset] == 0) {
+ continue;
+ }
+
+ /* comment, do not parse */
+ if (line[ws_offset] == '#')
+ continue;
+
+#if MIME_MAGIC_DEBUG
+ /* if we get here, we're going to use it so count it */
+ rule++;
+#endif
+
+ /* parse it */
+ if (parse(line + ws_offset, lineno) != 0)
+ ++errs;
+ }
+
+ (void) fclose(f);
+
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ MODNAME ": apprentice conf=%x file=%s m=%s m->next=%s last=%s",
+ conf,
+ conf->magicfile ? conf->magicfile : "NULL",
+ conf->magic ? "set" : "NULL",
+ (conf->magic && conf->magic->next) ? "set" : "NULL",
+ conf->last ? "set" : "NULL");
+ php_error(E_NOTICE,
+ MODNAME ": apprentice read %d lines, %d rules, %d errors",
+ lineno, rule, errs);
+#endif
+
+#if MIME_MAGIC_DEBUG
+ prevm = 0;
+ php_error(E_NOTICE,
+ MODNAME ": apprentice test");
+ for (m = conf->magic; m; m = m->next) {
+ if (isprint((((unsigned long) m) >> 24) & 255) &&
+ isprint((((unsigned long) m) >> 16) & 255) &&
+ isprint((((unsigned long) m) >> 8) & 255) &&
+ isprint(((unsigned long) m) & 255)) {
+ php_error(E_NOTICE,
+ MODNAME ": apprentice: POINTER CLOBBERED! "
+ "m=\"%c%c%c%c\" line=%d",
+ (((unsigned long) m) >> 24) & 255,
+ (((unsigned long) m) >> 16) & 255,
+ (((unsigned long) m) >> 8) & 255,
+ ((unsigned long) m) & 255,
+ prevm ? prevm->lineno : -1);
+ break;
+ }
+ prevm = m;
+ }
+#endif
+
+ return (errs ? -1 : 0);
+}
+
+/*
+ * extend the sign bit if the comparison is to be signed
+ */
+static unsigned long signextend(struct magic *m, unsigned long v)
+{
+ if (!(m->flag & UNSIGNED))
+ switch (m->type) {
+ /*
+ * Do not remove the casts below. They are vital. When later
+ * compared with the data, the sign extension must have happened.
+ */
+ case BYTE:
+ v = (char) v;
+ break;
+ case SHORT:
+ case BESHORT:
+ case LESHORT:
+ v = (short) v;
+ break;
+ case DATE:
+ case BEDATE:
+ case LEDATE:
+ case LONG:
+ case BELONG:
+ case LELONG:
+ v = (long) v;
+ break;
+ case STRING:
+ break;
+ default:
+ php_error(E_WARNING,
+ MODNAME ": can't happen: m->type=%d", m->type);
+ return -1;
+ }
+ return v;
+}
+
+/*
+ * parse one line from magic file, put into magic[index++] if valid
+ */
+static int parse(char *l, int lineno)
+{
+ struct magic *m;
+ char *t, *s;
+ magic_server_config_rec *conf = &mime_global;
+
+ /* allocate magic structure entry */
+ m = (struct magic *) calloc(1, sizeof(struct magic));
+
+ /* append to linked list */
+ m->next = NULL;
+ if (!conf->magic || !conf->last) {
+ conf->magic = conf->last = m;
+ }
+ else {
+ conf->last->next = m;
+ conf->last = m;
+ }
+
+ /* set values in magic structure */
+ m->flag = 0;
+ m->cont_level = 0;
+ m->lineno = lineno;
+
+ while (*l == '>') {
+ ++l; /* step over */
+ m->cont_level++;
+ }
+
+ if (m->cont_level != 0 && *l == '(') {
+ ++l; /* step over */
+ m->flag |= INDIR;
+ }
+
+ /* get offset, then skip over it */
+ m->offset = (int) strtol(l, &t, 0);
+ if (l == t) {
+ php_error(E_WARNING,
+ MODNAME ": offset %s invalid", l);
+ }
+ l = t;
+
+ if (m->flag & INDIR) {
+ m->in.type = LONG;
+ m->in.offset = 0;
+ /*
+ * read [.lbs][+-]nnnnn)
+ */
+ if (*l == '.') {
+ switch (*++l) {
+ case 'l':
+ m->in.type = LONG;
+ break;
+ case 's':
+ m->in.type = SHORT;
+ break;
+ case 'b':
+ m->in.type = BYTE;
+ break;
+ default:
+ php_error(E_WARNING,
+ MODNAME ": indirect offset type %c invalid", *l);
+ break;
+ }
+ l++;
+ }
+ s = l;
+ if (*l == '+' || *l == '-')
+ l++;
+ if (isdigit((unsigned char) *l)) {
+ m->in.offset = strtol(l, &t, 0);
+ if (*s == '-')
+ m->in.offset = -m->in.offset;
+ }
+ else
+ t = l;
+ if (*t++ != ')') {
+ php_error(E_WARNING,
+ MODNAME ": missing ')' in indirect offset");
+ }
+ l = t;
+ }
+
+
+ while (isdigit((unsigned char) *l))
+ ++l;
+ EATAB;
+
+#define NBYTE 4
+#define NSHORT 5
+#define NLONG 4
+#define NSTRING 6
+#define NDATE 4
+#define NBESHORT 7
+#define NBELONG 6
+#define NBEDATE 6
+#define NLESHORT 7
+#define NLELONG 6
+#define NLEDATE 6
+
+ if (*l == 'u') {
+ ++l;
+ m->flag |= UNSIGNED;
+ }
+
+ /* get type, skip it */
+ if (strncmp(l, "byte", NBYTE) == 0) {
+ m->type = BYTE;
+ l += NBYTE;
+ }
+ else if (strncmp(l, "short", NSHORT) == 0) {
+ m->type = SHORT;
+ l += NSHORT;
+ }
+ else if (strncmp(l, "long", NLONG) == 0) {
+ m->type = LONG;
+ l += NLONG;
+ }
+ else if (strncmp(l, "string", NSTRING) == 0) {
+ m->type = STRING;
+ l += NSTRING;
+ }
+ else if (strncmp(l, "date", NDATE) == 0) {
+ m->type = DATE;
+ l += NDATE;
+ }
+ else if (strncmp(l, "beshort", NBESHORT) == 0) {
+ m->type = BESHORT;
+ l += NBESHORT;
+ }
+ else if (strncmp(l, "belong", NBELONG) == 0) {
+ m->type = BELONG;
+ l += NBELONG;
+ }
+ else if (strncmp(l, "bedate", NBEDATE) == 0) {
+ m->type = BEDATE;
+ l += NBEDATE;
+ }
+ else if (strncmp(l, "leshort", NLESHORT) == 0) {
+ m->type = LESHORT;
+ l += NLESHORT;
+ }
+ else if (strncmp(l, "lelong", NLELONG) == 0) {
+ m->type = LELONG;
+ l += NLELONG;
+ }
+ else if (strncmp(l, "ledate", NLEDATE) == 0) {
+ m->type = LEDATE;
+ l += NLEDATE;
+ }
+ else {
+ php_error(E_WARNING,
+ MODNAME ": type %s invalid", l);
+ return -1;
+ }
+ /* New-style anding: "0 byte&0x80 =0x80 dynamically linked" */
+ if (*l == '&') {
+ ++l;
+ m->mask = signextend(m, strtol(l, &l, 0));
+ }
+ else
+ m->mask = ~0L;
+ EATAB;
+
+ switch (*l) {
+ case '>':
+ case '<':
+ /* Old-style anding: "0 byte &0x80 dynamically linked" */
+ case '&':
+ case '^':
+ case '=':
+ m->reln = *l;
+ ++l;
+ break;
+ case '!':
+ if (m->type != STRING) {
+ m->reln = *l;
+ ++l;
+ break;
+ }
+ /* FALL THROUGH */
+ default:
+ if (*l == 'x' && isspace((unsigned char) l[1])) {
+ m->reln = *l;
+ ++l;
+ goto GetDesc; /* Bill The Cat */
+ }
+ m->reln = '=';
+ break;
+ }
+ EATAB;
+
+ if (getvalue(m, &l))
+ return -1;
+ /*
+ * now get last part - the description
+ */
+ GetDesc:
+ EATAB;
+ if (l[0] == '\b') {
+ ++l;
+ m->nospflag = 1;
+ }
+ else if ((l[0] == '\\') && (l[1] == 'b')) {
+ ++l;
+ ++l;
+ m->nospflag = 1;
+ }
+ else
+ m->nospflag = 0;
+ strncpy(m->desc, l, sizeof(m->desc) - 1);
+ m->desc[sizeof(m->desc) - 1] = '\0';
+
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ MODNAME ": parse line=%d m=%x next=%x cont=%d desc=%s",
+ lineno, m, m->next, m->cont_level, m->desc);
+#endif /* MIME_MAGIC_DEBUG */
+
+ return 0;
+}
+
+/*
+ * Read a numeric value from a pointer, into the value union of a magic
+ * pointer, according to the magic type. Update the string pointer to point
+ * just after the number read. Return 0 for success, non-zero for failure.
+ */
+static int getvalue(struct magic *m, char **p)
+{
+ int slen;
+
+ if (m->type == STRING) {
+ *p = getstr(*p, m->value.s, sizeof(m->value.s), &slen);
+ m->vallen = slen;
+ }
+ else if (m->reln != 'x')
+ m->value.l = signextend(m, strtol(*p, p, 0));
+ return 0;
+}
+
+/*
+ * Convert a string containing C character escapes. Stop at an unescaped
+ * space or tab. Copy the converted version to "p", returning its length in
+ * *slen. Return updated scan pointer as function result.
+ */
+static char *getstr(register char *s, register char *p,
+ int plen, int *slen)
+{
+ char *origs = s, *origp = p;
+ char *pmax = p + plen - 1;
+ register int c;
+ register int val;
+
+ while ((c = *s++) != '\0') {
+ if (isspace((unsigned char) c))
+ break;
+ if (p >= pmax) {
+ php_error(E_WARNING,
+ MODNAME ": string too long: %s", origs);
+ break;
+ }
+ if (c == '\\') {
+ switch (c = *s++) {
+
+ case '\0':
+ goto out;
+
+ default:
+ *p++ = (char) c;
+ break;
+
+ case 'n':
+ *p++ = '\n';
+ break;
+
+ case 'r':
+ *p++ = '\r';
+ break;
+
+ case 'b':
+ *p++ = '\b';
+ break;
+
+ case 't':
+ *p++ = '\t';
+ break;
+
+ case 'f':
+ *p++ = '\f';
+ break;
+
+ case 'v':
+ *p++ = '\v';
+ break;
+
+ /* \ and up to 3 octal digits */
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ val = c - '0';
+ c = *s++; /* try for 2 */
+ if (c >= '0' && c <= '7') {
+ val = (val << 3) | (c - '0');
+ c = *s++; /* try for 3 */
+ if (c >= '0' && c <= '7')
+ val = (val << 3) | (c - '0');
+ else
+ --s;
+ }
+ else
+ --s;
+ *p++ = (char) val;
+ break;
+
+ /* \x and up to 3 hex digits */
+ case 'x':
+ val = 'x'; /* Default if no digits */
+ c = hextoint(*s++); /* Get next char */
+ if (c >= 0) {
+ val = c;
+ c = hextoint(*s++);
+ if (c >= 0) {
+ val = (val << 4) + c;
+ c = hextoint(*s++);
+ if (c >= 0) {
+ val = (val << 4) + c;
+ }
+ else
+ --s;
+ }
+ else
+ --s;
+ }
+ else
+ --s;
+ *p++ = (char) val;
+ break;
+ }
+ }
+ else
+ *p++ = (char) c;
+ }
+ out:
+ *p = '\0';
+ *slen = p - origp;
+ return s;
+}
+
+/* Single hex char to int; -1 if not a hex char. */
+static int hextoint(int c)
+{
+ if (isdigit((unsigned char) c))
+ return c - '0';
+ if ((c >= 'a') && (c <= 'f'))
+ return c + 10 - 'a';
+ if ((c >= 'A') && (c <= 'F'))
+ return c + 10 - 'A';
+ return -1;
+}
+
+
+/*
+ * RSL (result string list) processing routines
+ *
+ * These collect strings that would have been printed in fragments by file(1)
+ * into a list of magic_rsl structures with the strings. When complete,
+ * they're concatenated together to become the MIME content and encoding
+ * types.
+ *
+ * return value conventions for these functions: functions which return int:
+ * failure = -1, other = result functions which return pointers: failure = 0,
+ * other = result
+ */
+
+/* allocate a per-request structure and put it in the request record */
+static magic_req_rec *magic_set_config(void)
+{
+ magic_req_rec *req_dat = (magic_req_rec *) emalloc(sizeof(magic_req_rec));
+
+ req_dat->head = req_dat->tail = (magic_rsl *) NULL;
+ MIME_MAGIC_G(req_dat) = req_dat;
+ return req_dat;
+}
+
+static void magic_free_config(magic_req_rec *req_dat) {
+ magic_rsl *curr, *next;
+
+ if(!req_dat) return;
+
+ curr = req_dat->head;
+ while(curr) {
+ next = curr->next;
+ efree(curr->str);
+ efree(curr);
+ curr = next;
+ }
+
+ efree(req_dat);
+}
+
+/* add a string to the result string list for this request */
+/* it is the responsibility of the caller to allocate "str" */
+static int magic_rsl_add(char *str)
+{
+ magic_req_rec *req_dat = MIME_MAGIC_G(req_dat);
+ magic_rsl *rsl;
+
+ /* make sure we have a list to put it in */
+ if (!req_dat) {
+ php_error(E_WARNING,
+ MODNAME ": request config should not be NULL");
+ if (!(req_dat = magic_set_config())) {
+ /* failure */
+ return -1;
+ }
+ }
+
+ /* allocate the list entry */
+ rsl = (magic_rsl *) emalloc(sizeof(magic_rsl));
+
+ /* fill it */
+ rsl->str = estrdup(str);
+ rsl->next = (magic_rsl *) NULL;
+
+ /* append to the list */
+ if (req_dat->head && req_dat->tail) {
+ req_dat->tail->next = rsl;
+ req_dat->tail = rsl;
+ }
+ else {
+ req_dat->head = req_dat->tail = rsl;
+ }
+
+ /* success */
+ return 0;
+}
+
+/* RSL hook for puts-type functions */
+static int magic_rsl_puts(char *str)
+{
+ return magic_rsl_add(str);
+}
+
+/* RSL hook for printf-type functions */
+static int magic_rsl_printf(char *str,...)
+{
+ va_list ap;
+
+ char buf[MAXMIMESTRING];
+
+ /* assemble the string into the buffer */
+ va_start(ap, str);
+ vsnprintf(buf, sizeof(buf), str, ap);
+ va_end(ap);
+
+ /* add the buffer to the list */
+ return magic_rsl_add(buf);
+}
+
+/* RSL hook for putchar-type functions */
+static int magic_rsl_putchar(char c)
+{
+ char str[2];
+
+ /* high overhead for 1 char - just hope they don't do this much */
+ str[0] = c;
+ str[1] = '\0';
+ return magic_rsl_add(str);
+}
+
+/* allocate and copy a contiguous string from a result string list */
+static char *rsl_strdup(int start_frag, int start_pos, int len)
+{
+ char *result; /* return value */
+ int cur_frag, /* current fragment number/counter */
+ cur_pos, /* current position within fragment */
+ res_pos; /* position in result string */
+ magic_rsl *frag; /* list-traversal pointer */
+ magic_req_rec *req_dat = MIME_MAGIC_G(req_dat);
+
+ /* allocate the result string */
+ result = (char *) emalloc(len + 1);
+
+ /* loop through and collect the string */
+ res_pos = 0;
+ for (frag = req_dat->head, cur_frag = 0;
+ frag->next;
+ frag = frag->next, cur_frag++) {
+ /* loop to the first fragment */
+ if (cur_frag < start_frag)
+ continue;
+
+ /* loop through and collect chars */
+ for (cur_pos = (cur_frag == start_frag) ? start_pos : 0;
+ frag->str[cur_pos];
+ cur_pos++) {
+ if (cur_frag >= start_frag
+ && cur_pos >= start_pos
+ && res_pos <= len) {
+ result[res_pos++] = frag->str[cur_pos];
+ if (res_pos > len) {
+ break;
+ }
+ }
+ }
+ }
+
+ /* clean up and return */
+ result[res_pos] = 0;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ MODNAME ": rsl_strdup() %d chars: %s", res_pos - 1, result);
+#endif
+ return result;
+}
+
+/*
+ * magic_process - process input file r Apache API request record
+ * (formerly called "process" in file command, prefix added for clarity) Opens
+ * the file and reads a fixed-size buffer to begin processing the contents.
+ */
+static int magic_process(char *filename)
+{
+ int fd = 0;
+ unsigned char buf[HOWMANY + 1]; /* one extra for terminating '\0' */
+ int nbytes = 0; /* number of bytes read from a datafile */
+ int result;
+
+ /*
+ * first try judging the file based on its filesystem status
+ */
+ switch ((result = fsmagic(filename))) {
+ case DONE:
+ magic_rsl_putchar('\n');
+ return OK;
+ case OK:
+ break;
+ default:
+ /* fatal error, bail out */
+ return result;
+ }
+
+ if ((fd = open(filename, O_RDONLY, 0)) < 0) {
+ /* We can't open it, but we were able to stat it. */
+ php_error(E_WARNING,
+ MODNAME ": can't read `%s'", filename);
+ /* let some other handler decide what the problem is */
+ return DECLINED;
+ }
+
+ /*
+ * try looking at the first HOWMANY bytes
+ */
+ if ((nbytes = read(fd, (char *) buf, sizeof(buf) - 1)) == -1) {
+ php_error(E_WARNING,
+ MODNAME ": read failed: %s", filename);
+ return ERROR;
+ }
+
+ if (nbytes == 0)
+ magic_rsl_puts(MIME_TEXT_UNKNOWN);
+ else {
+ buf[nbytes++] = '\0'; /* null-terminate it */
+ tryit(buf, nbytes, 1);
+ }
+
+ (void) close(fd);
+ (void) magic_rsl_putchar('\n');
+
+ return OK;
+}
+
+static void tryit(unsigned char *buf, int nb, int checkzmagic)
+{
+ /*
+ * Try compression stuff
+ */
+#if 0 /* TODO */
+ if (checkzmagic == 1) {
+ if (zmagic(buf, nb) == 1)
+ return;
+ }
+#endif
+
+ /*
+ * try tests in /etc/magic (or surrogate magic file)
+ */
+ if (softmagic(buf, nb) == 1)
+ return;
+
+ /*
+ * try known keywords, check for ascii-ness too.
+ */
+ if (ascmagic(buf, nb) == 1)
+ return;
+
+ /*
+ * abandon hope, all ye who remain here
+ */
+ magic_rsl_puts(MIME_BINARY_UNKNOWN);
+}
+
+
+/*
+ * return DONE to indicate it's been handled
+ * return OK to indicate it's a regular file still needing handling
+ * other returns indicate a failure of some sort
+ */
+static int fsmagic(const char *fn)
+{
+ struct stat finfo;
+
+ stat(fn, &finfo);
+
+ switch (finfo.st_mode & S_IFMT) {
+ case S_IFDIR:
+ magic_rsl_puts(DIR_MAGIC_TYPE);
+ return DONE;
+ case S_IFCHR:
+ /*
+ * (void) magic_rsl_printf(r,"character special (%d/%d)",
+ * major(sb->st_rdev), minor(sb->st_rdev));
+ */
+ (void) magic_rsl_puts(MIME_BINARY_UNKNOWN);
+ return DONE;
+#ifdef S_IFBLK
+ case S_IFBLK:
+ /*
+ * (void) magic_rsl_printf(r,"block special (%d/%d)",
+ * major(sb->st_rdev), minor(sb->st_rdev));
+ */
+ (void) magic_rsl_puts(MIME_BINARY_UNKNOWN);
+ return DONE;
+ /* TODO add code to handle V7 MUX and Blit MUX files */
+#endif
+#ifdef S_IFIFO
+ case S_IFIFO:
+ /*
+ * magic_rsl_puts(r,"fifo (named pipe)");
+ */
+ (void) magic_rsl_puts(MIME_BINARY_UNKNOWN);
+ return DONE;
+#endif
+#ifdef S_IFLNK
+ case S_IFLNK:
+ /* We used stat(), the only possible reason for this is that the
+ * symlink is broken.
+ */
+ php_error(E_WARNING,
+ MODNAME ": broken symlink (%s)", fn);
+ return ERROR;
+#endif
+#ifdef S_IFSOCK
+#ifndef __COHERENT__
+ case S_IFSOCK:
+ magic_rsl_puts(MIME_BINARY_UNKNOWN);
+ return DONE;
+#endif
+#endif
+ case S_IFREG:
+ break;
+ default:
+ php_error(E_WARNING,
+ MODNAME ": invalid mode 0%o.", (unsigned int)finfo.st_mode);
+ return ERROR;
+ }
+
+ /*
+ * regular file, check next possibility
+ */
+ if (finfo.st_size == 0) {
+ magic_rsl_puts(MIME_TEXT_UNKNOWN);
+ return DONE;
+ }
+ return OK;
+}
+
+/*
+ * softmagic - lookup one file in database (already read from /etc/magic by
+ * apprentice.c). Passed the name and FILE * of one file to be typed.
+ */
+/* ARGSUSED1 *//* nbytes passed for regularity, maybe need later */
+static int softmagic(unsigned char *buf, int nbytes)
+{
+ if (match(buf, nbytes))
+ return 1;
+
+ return 0;
+}
+
+/*
+ * Go through the whole list, stopping if you find a match. Process all the
+ * continuations of that match before returning.
+ *
+ * We support multi-level continuations:
+ *
+ * At any time when processing a successful top-level match, there is a current
+ * continuation level; it represents the level of the last successfully
+ * matched continuation.
+ *
+ * Continuations above that level are skipped as, if we see one, it means that
+ * the continuation that controls them - i.e, the lower-level continuation
+ * preceding them - failed to match.
+ *
+ * Continuations below that level are processed as, if we see one, it means
+ * we've finished processing or skipping higher-level continuations under the
+ * control of a successful or unsuccessful lower-level continuation, and are
+ * now seeing the next lower-level continuation and should process it. The
+ * current continuation level reverts to the level of the one we're seeing.
+ *
+ * Continuations at the current level are processed as, if we see one, there's
+ * no lower-level continuation that may have failed.
+ *
+ * If a continuation matches, we bump the current continuation level so that
+ * higher-level continuations are processed.
+ */
+static int match(unsigned char *s, int nbytes)
+{
+#if MIME_MAGIC_DEBUG
+ int rule_counter = 0;
+#endif
+ int cont_level = 0;
+ int need_separator = 0;
+ union VALUETYPE p;
+ magic_server_config_rec *conf = &mime_global;
+ struct magic *m;
+
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ MODNAME ": match conf=%x file=%s m=%s m->next=%s last=%s",
+ conf,
+ conf->magicfile ? conf->magicfile : "NULL",
+ conf->magic ? "set" : "NULL",
+ (conf->magic && conf->magic->next) ? "set" : "NULL",
+ conf->last ? "set" : "NULL");
+#endif
+
+#if MIME_MAGIC_DEBUG
+ for (m = conf->magic; m; m = m->next) {
+ if (isprint((((unsigned long) m) >> 24) & 255) &&
+ isprint((((unsigned long) m) >> 16) & 255) &&
+ isprint((((unsigned long) m) >> 8) & 255) &&
+ isprint(((unsigned long) m) & 255)) {
+ php_error(E_NOTICE,
+ MODNAME ": match: POINTER CLOBBERED! "
+ "m=\"%c%c%c%c\"",
+ (((unsigned long) m) >> 24) & 255,
+ (((unsigned long) m) >> 16) & 255,
+ (((unsigned long) m) >> 8) & 255,
+ ((unsigned long) m) & 255);
+ break;
+ }
+ }
+#endif
+
+ for (m = conf->magic; m; m = m->next) {
+#if MIME_MAGIC_DEBUG
+ rule_counter++;
+ php_error(E_NOTICE,
+ MODNAME ": line=%d desc=%s", m->lineno, m->desc);
+#endif
+
+ /* check if main entry matches */
+ if (!mget(&p, s, m, nbytes) ||
+ !mcheck(&p, m)) {
+ struct magic *m_cont;
+
+ /*
+ * main entry didn't match, flush its continuations
+ */
+ if (!m->next || (m->next->cont_level == 0)) {
+ continue;
+ }
+
+ m_cont = m->next;
+ while (m_cont && (m_cont->cont_level != 0)) {
+#if MIME_MAGIC_DEBUG
+ rule_counter++;
+ php_error(E_NOTICE,
+ MODNAME ": line=%d mc=%x mc->next=%x cont=%d desc=%s",
+ m_cont->lineno, m_cont,
+ m_cont->next, m_cont->cont_level,
+ m_cont->desc);
+#endif
+ /*
+ * this trick allows us to keep *m in sync when the continue
+ * advances the pointer
+ */
+ m = m_cont;
+ m_cont = m_cont->next;
+ }
+ continue;
+ }
+
+ /* if we get here, the main entry rule was a match */
+ /* this will be the last run through the loop */
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ MODNAME ": rule matched, line=%d type=%d %s",
+ m->lineno, m->type,
+ (m->type == STRING) ? m->value.s : "");
+#endif
+
+ /* print the match */
+ mprint(&p, m);
+
+ /*
+ * If we printed something, we'll need to print a blank before we
+ * print something else.
+ */
+ if (m->desc[0])
+ need_separator = 1;
+ /* and any continuations that match */
+ cont_level++;
+ /*
+ * while (m && m->next && m->next->cont_level != 0 && ( m = m->next
+ * ))
+ */
+ m = m->next;
+ while (m && (m->cont_level != 0)) {
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ MODNAME ": match line=%d cont=%d type=%d %s",
+ m->lineno, m->cont_level, m->type,
+ (m->type == STRING) ? m->value.s : "");
+#endif
+ if (cont_level >= m->cont_level) {
+ if (cont_level > m->cont_level) {
+ /*
+ * We're at the end of the level "cont_level"
+ * continuations.
+ */
+ cont_level = m->cont_level;
+ }
+ if (mget(&p, s, m, nbytes) &&
+ mcheck(&p, m)) {
+ /*
+ * This continuation matched. Print its message, with a
+ * blank before it if the previous item printed and this
+ * item isn't empty.
+ */
+ /* space if previous printed */
+ if (need_separator
+ && (m->nospflag == 0)
+ && (m->desc[0] != '\0')
+ ) {
+ (void) magic_rsl_putchar(' ');
+ need_separator = 0;
+ }
+ mprint(&p, m);
+ if (m->desc[0])
+ need_separator = 1;
+
+ /*
+ * If we see any continuations at a higher level, process
+ * them.
+ */
+ cont_level++;
+ }
+ }
+
+ /* move to next continuation record */
+ m = m->next;
+ }
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ MODNAME ": matched after %d rules", rule_counter);
+#endif
+ return 1; /* all through */
+ }
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ MODNAME ": failed after %d rules", rule_counter);
+#endif
+ return 0; /* no match at all */
+}
+
+/* an optimization over plain strcmp() */
+#define STREQ(a, b) (*(a) == *(b) && strcmp((a), (b)) == 0)
+
+static int ascmagic(unsigned char *buf, int nbytes)
+{
+ int has_escapes = 0;
+ unsigned char *s;
+ char nbuf[HOWMANY + 1]; /* one extra for terminating '\0' */
+ char *token;
+ register struct names *p;
+ int small_nbytes;
+
+ /* these are easy, do them first */
+
+ /*
+ * for troff, look for . + letter + letter or .\"; this must be done to
+ * disambiguate tar archives' ./file and other trash from real troff
+ * input.
+ */
+ if (*buf == '.') {
+ unsigned char *tp = buf + 1;
+
+ while (isspace(*tp))
+ ++tp; /* skip leading whitespace */
+ if ((isalnum(*tp) || *tp == '\\') &&
+ (isalnum(*(tp + 1)) || *tp == '"')) {
+ magic_rsl_puts("application/x-troff");
+ return 1;
+ }
+ }
+ if ((*buf == 'c' || *buf == 'C') && isspace(*(buf + 1))) {
+ /* Fortran */
+ magic_rsl_puts("text/plain");
+ return 1;
+ }
+
+ /* look for tokens from names.h - this is expensive!, so we'll limit
+ * ourselves to only SMALL_HOWMANY bytes */
+ small_nbytes = (nbytes > SMALL_HOWMANY) ? SMALL_HOWMANY : nbytes;
+ /* make a copy of the buffer here because strtok() will destroy it */
+ s = (unsigned char *) memcpy(nbuf, buf, small_nbytes);
+ s[small_nbytes] = '\0';
+ has_escapes = (memchr(s, '\033', small_nbytes) != NULL);
+ /* XXX: not multithread safe */
+ while ((token = strtok((char *) s, " \t\n\r\f")) != NULL) {
+ s = NULL; /* make strtok() keep on tokin' */
+ for (p = names; p < names + NNAMES; p++) {
+ if (STREQ(p->name, token)) {
+ magic_rsl_puts(types[p->type]);
+ if (has_escapes)
+ magic_rsl_puts(" (with escape sequences)");
+ return 1;
+ }
+ }
+ }
+
+ switch (is_tar(buf, nbytes)) {
+ case 1:
+ /* V7 tar archive */
+ magic_rsl_puts("application/x-tar");
+ return 1;
+ case 2:
+ /* POSIX tar archive */
+ magic_rsl_puts("application/x-tar");
+ return 1;
+ }
+
+ /* all else fails, but it is ascii... */
+ if (has_escapes) {
+ /* text with escape sequences */
+ /* we leave this open for further differentiation later */
+ magic_rsl_puts("text/plain");
+ }
+ else {
+ /* plain text */
+ magic_rsl_puts("text/plain");
+ }
+ return 1;
+}
+
+/*
+ * is_tar() -- figure out whether file is a tar archive.
+ *
+ * Stolen (by author of file utility) from the public domain tar program: Public
+ * Domain version written 26 Aug 1985 John Gilmore (ihnp4!hoptoad!gnu).
+ *
+ * @(#)list.c 1.18 9/23/86 Public Domain - gnu $Id: mod_mime_magic.c,v 1.7
+ * 1997/06/24 00:41:02 ikluft Exp ikluft $
+ *
+ * Comments changed and some code/comments reformatted for file command by Ian
+ * Darwin.
+ */
+
+#define isodigit(c) ( ((c) >= '0') && ((c) <= '7') )
+
+/*
+ * Return 0 if the checksum is bad (i.e., probably not a tar archive), 1 for
+ * old UNIX tar file, 2 for Unix Std (POSIX) tar file.
+ */
+
+static int is_tar(unsigned char *buf, int nbytes)
+{
+ register union record *header = (union record *) buf;
+ register int i;
+ register long sum, recsum;
+ register char *p;
+
+ if (nbytes < sizeof(union record))
+ return 0;
+
+ recsum = from_oct(8, header->header.chksum);
+
+ sum = 0;
+ p = header->charptr;
+ for (i = sizeof(union record); --i >= 0;) {
+ /*
+ * We can't use unsigned char here because of old compilers, e.g. V7.
+ */
+ sum += 0xFF & *p++;
+ }
+
+ /* Adjust checksum to count the "chksum" field as blanks. */
+ for (i = sizeof(header->header.chksum); --i >= 0;)
+ sum -= 0xFF & header->header.chksum[i];
+ sum += ' ' * sizeof header->header.chksum;
+
+ if (sum != recsum)
+ return 0; /* Not a tar archive */
+
+ if (0 == strcmp(header->header.magic, TMAGIC))
+ return 2; /* Unix Standard tar archive */
+
+ return 1; /* Old fashioned tar archive */
+}
+
+
+/*
+ * Quick and dirty octal conversion.
+ *
+ * Result is -1 if the field is invalid (all blank, or nonoctal).
+ */
+static long from_oct(int digs, char *where)
+{
+ register long value;
+
+ while (isspace(*where)) { /* Skip spaces */
+ where++;
+ if (--digs <= 0)
+ return -1; /* All blank field */
+ }
+ value = 0;
+ while (digs > 0 && isodigit(*where)) { /* Scan til nonoctal */
+ value = (value << 3) | (*where++ - '0');
+ --digs;
+ }
+
+ if (digs > 0 && *where && !isspace(*where))
+ return -1; /* Ended on non-space/nul */
+
+ return value;
+}
+
+static int mget(union VALUETYPE *p, unsigned char *s,
+ struct magic *m, int nbytes)
+{
+ long offset = m->offset;
+
+ if (offset + sizeof(union VALUETYPE) > nbytes)
+ return 0;
+
+ memcpy(p, s + offset, sizeof(union VALUETYPE));
+
+ if (!mconvert(p, m))
+ return 0;
+
+ if (m->flag & INDIR) {
+
+ switch (m->in.type) {
+ case BYTE:
+ offset = p->b + m->in.offset;
+ break;
+ case SHORT:
+ offset = p->h + m->in.offset;
+ break;
+ case LONG:
+ offset = p->l + m->in.offset;
+ break;
+ }
+
+ if (offset + sizeof(union VALUETYPE) > nbytes)
+ return 0;
+
+ memcpy(p, s + offset, sizeof(union VALUETYPE));
+
+ if (!mconvert(p, m))
+ return 0;
+ }
+ return 1;
+}
+
+static int mcheck(union VALUETYPE *p, struct magic *m)
+{
+ register unsigned long l = m->value.l;
+ register unsigned long v;
+ int matched;
+
+ if ((m->value.s[0] == 'x') && (m->value.s[1] == '\0')) {
+ php_error(E_WARNING,
+ MODNAME ": BOINK");
+ return 1;
+ }
+
+ switch (m->type) {
+ case BYTE:
+ v = p->b;
+ break;
+
+ case SHORT:
+ case BESHORT:
+ case LESHORT:
+ v = p->h;
+ break;
+
+ case LONG:
+ case BELONG:
+ case LELONG:
+ case DATE:
+ case BEDATE:
+ case LEDATE:
+ v = p->l;
+ break;
+
+ case STRING:
+ l = 0;
+ /*
+ * What we want here is: v = strncmp(m->value.s, p->s, m->vallen);
+ * but ignoring any nulls. bcmp doesn't give -/+/0 and isn't
+ * universally available anyway.
+ */
+ v = 0;
+ {
+ register unsigned char *a = (unsigned char *) m->value.s;
+ register unsigned char *b = (unsigned char *) p->s;
+ register int len = m->vallen;
+
+ while (--len >= 0)
+ if ((v = *b++ - *a++) != 0)
+ break;
+ }
+ break;
+ default:
+ /* bogosity, pretend that it just wasn't a match */
+ php_error(E_WARNING,
+ MODNAME ": invalid type %d in mcheck().", m->type);
+ return 0;
+ }
+
+ v = signextend(m, v) & m->mask;
+
+ switch (m->reln) {
+ case 'x':
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "%lu == *any* = 1", v);
+#endif
+ matched = 1;
+ break;
+
+ case '!':
+ matched = v != l;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "%lu != %lu = %d", v, l, matched);
+#endif
+ break;
+
+ case '=':
+ matched = v == l;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "%lu == %lu = %d", v, l, matched);
+#endif
+ break;
+
+ case '>':
+ if (m->flag & UNSIGNED) {
+ matched = v > l;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "%lu > %lu = %d", v, l, matched);
+#endif
+ }
+ else {
+ matched = (long) v > (long) l;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "%ld > %ld = %d", v, l, matched);
+#endif
+ }
+ break;
+
+ case '<':
+ if (m->flag & UNSIGNED) {
+ matched = v < l;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "%lu < %lu = %d", v, l, matched);
+#endif
+ }
+ else {
+ matched = (long) v < (long) l;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "%ld < %ld = %d", v, l, matched);
+#endif
+ }
+ break;
+
+ case '&':
+ matched = (v & l) == l;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "((%lx & %lx) == %lx) = %d", v, l, l, matched);
+#endif
+ break;
+
+ case '^':
+ matched = (v & l) != l;
+#if MIME_MAGIC_DEBUG
+ php_error(E_NOTICE,
+ "((%lx & %lx) != %lx) = %d", v, l, l, matched);
+#endif
+ break;
+
+ default:
+ /* bogosity, pretend it didn't match */
+ matched = 0;
+ php_error(E_WARNING,
+ MODNAME ": mcheck: can't happen: invalid relation %d.",
+ m->reln);
+ break;
+ }
+
+ return matched;
+}
+
+#if 0 /* TODO */
+/*
+ * compress routines: zmagic() - returns 0 if not recognized, uncompresses
+ * and prints information if recognized uncompress(s, method, old, n, newch)
+ * - uncompress old into new, using method, return sizeof new
+ */
+
+static struct {
+ char *magic;
+ int maglen;
+ char *argv[3];
+ int silent;
+ char *encoding; /* MUST be lowercase */
+} compr[] = {
+
+ /* we use gzip here rather than uncompress because we have to pass
+ * it a full filename -- and uncompress only considers filenames
+ * ending with .Z
+ */
+ {
+ "\037\235", 2, {
+ "gzip", "-dcq", NULL
+ }, 0, "x-compress"
+ },
+ {
+ "\037\213", 2, {
+ "gzip", "-dcq", NULL
+ }, 1, "x-gzip"
+ },
+ /*
+ * XXX pcat does not work, cause I don't know how to make it read stdin,
+ * so we use gzip
+ */
+ {
+ "\037\036", 2, {
+ "gzip", "-dcq", NULL
+ }, 0, "x-gzip"
+ },
+};
+
+static int ncompr = sizeof(compr) / sizeof(compr[0]);
+
+static int zmagic(unsigned char *buf, int nbytes)
+{
+ unsigned char *newbuf;
+ int newsize;
+ int i;
+
+ for (i = 0; i < ncompr; i++) {
+ if (nbytes < compr[i].maglen)
+ continue;
+ if (memcmp(buf, compr[i].magic, compr[i].maglen) == 0)
+ break;
+ }
+
+ if (i == ncompr)
+ return 0;
+
+ if ((newsize = uncompress(i, &newbuf, nbytes)) > 0) {
+ tryit(newbuf, newsize, 0);
+
+ /* set encoding type in the request record */
+ /* TODO r->content_encoding = compr[i].encoding; */
+ }
+ return 1;
+}
+
+
+struct uncompress_parms {
+ request_rec *r;
+ int method;
+};
+
+static int uncompress_child(void *data, child_info *pinfo)
+{
+ struct uncompress_parms *parm = data;
+#ifndef WIN32
+ char *new_argv[4];
+
+ new_argv[0] = compr[parm->method].argv[0];
+ new_argv[1] = compr[parm->method].argv[1];
+ new_argv[2] = parm->r->filename;
+ new_argv[3] = NULL;
+
+ if (compr[parm->method].silent) {
+ close(STDERR_FILENO);
+ }
+
+ execvp(compr[parm->method].argv[0], new_argv);
+ ap_log_rerror(APLOG_MARK, APLOG_ERR, parm->r,
+ MODNAME ": could not execute `%s'.",
+ compr[parm->method].argv[0]);
+ return -1;
+#else
+ char *pCommand;
+ STARTUPINFO si;
+ PROCESS_INFORMATION pi;
+ pid_t pid;
+
+ memset(&si, 0, sizeof(si));
+ memset(&pi, 0, sizeof(pi));
+
+ pid = -1;
+
+ /*
+ * Look at the arguments...
+ */
+ pCommand = ap_pstrcat(parm->r->pool, compr[parm->method].argv[0], " ",
+ compr[parm->method].argv[1], " \"",
+ parm->r->filename, "\"", NULL);
+
+ /*
+ * Make child process use hPipeOutputWrite as standard out,
+ * and make sure it does not show on screen.
+ */
+ si.cb = sizeof(si);
+ si.dwFlags = STARTF_USESTDHANDLES | STARTF_USESHOWWINDOW;
+ si.wShowWindow = SW_HIDE;
+ si.hStdInput = pinfo->hPipeInputRead;
+ si.hStdOutput = pinfo->hPipeOutputWrite;
+ si.hStdError = pinfo->hPipeErrorWrite;
+
+ if (CreateProcess(NULL, pCommand, NULL, NULL, TRUE, 0, NULL,
+ ap_make_dirstr_parent(parm->r->pool, parm->r->filename),
+ &si, &pi)) {
+ pid = pi.dwProcessId;
+ /*
+ * We must close the handles to the new process and its main thread
+ * to prevent handle and memory leaks.
+ */
+ CloseHandle(pi.hProcess);
+ CloseHandle(pi.hThread);
+ }
+ return (pid);
+#endif
+}
+
+static int uncompress(int method, unsigned char **newch, int n)
+{
+ struct uncompress_parms parm;
+ BUFF *bout;
+ pool *sub_pool;
+
+ parm.r = r;
+ parm.method = method;
+
+ /* We make a sub_pool so that we can collect our child early, otherwise
+ * there are cases (i.e. generating directory indicies with mod_autoindex)
+ * where we would end up with LOTS of zombies.
+ */
+ sub_pool = ap_make_sub_pool(r->pool);
+
+ if (!ap_bspawn_child(sub_pool, uncompress_child, &parm, kill_always,
+ NULL, &bout, NULL)) {
+ ap_log_rerror(APLOG_MARK, APLOG_ERR, r,
+ MODNAME ": couldn't spawn uncompress process: %s", r->uri);
+ return -1;
+ }
+
+ *newch = (unsigned char *) emalloc(n);
+ if ((n = ap_bread(bout, *newch, n)) <= 0) {
+ ap_destroy_pool(sub_pool);
+ ap_log_rerror(APLOG_MARK, APLOG_ERR, r,
+ MODNAME ": read failed %s", r->filename);
+ return -1;
+ }
+ ap_destroy_pool(sub_pool);
+ return n;
+}
+#endif
+
+static void mprint(union VALUETYPE *p, struct magic *m)
+{
+ char *pp, *rt;
+ unsigned long v;
+
+ switch (m->type) {
+ case BYTE:
+ v = p->b;
+ break;
+
+ case SHORT:
+ case BESHORT:
+ case LESHORT:
+ v = p->h;
+ break;
+
+ case LONG:
+ case BELONG:
+ case LELONG:
+ v = p->l;
+ break;
+
+ case STRING:
+ if (m->reln == '=') {
+ (void) magic_rsl_printf(m->desc, m->value.s);
+ }
+ else {
+ (void) magic_rsl_printf(m->desc, p->s);
+ }
+ return;
+
+ case DATE:
+ case BEDATE:
+ case LEDATE:
+ /* XXX: not multithread safe */
+ pp = ctime((time_t *) & p->l);
+ if ((rt = strchr(pp, '\n')) != NULL)
+ *rt = '\0';
+ (void) magic_rsl_printf(m->desc, pp);
+ return;
+ default:
+ php_error(E_WARNING,
+ MODNAME ": invalid m->type (%d) in mprint().",
+ m->type);
+ return;
+ }
+
+ v = signextend(m, v) & m->mask;
+ (void) magic_rsl_printf(m->desc, (unsigned long) v);
+}
+
+/*
+ * Convert the byte order of the data we are looking at
+ */
+static int mconvert(union VALUETYPE *p, struct magic *m)
+{
+ char *rt;
+
+ switch (m->type) {
+ case BYTE:
+ case SHORT:
+ case LONG:
+ case DATE:
+ return 1;
+ case STRING:
+ /* Null terminate and eat the return */
+ p->s[sizeof(p->s) - 1] = '\0';
+ if ((rt = strchr(p->s, '\n')) != NULL)
+ *rt = '\0';
+ return 1;
+ case BESHORT:
+ p->h = (short) ((p->hs[0] << 8) | (p->hs[1]));
+ return 1;
+ case BELONG:
+ case BEDATE:
+ p->l = (long)
+ ((p->hl[0] << 24) | (p->hl[1] << 16) | (p->hl[2] << 8) | (p->hl[3]));
+ return 1;
+ case LESHORT:
+ p->h = (short) ((p->hs[1] << 8) | (p->hs[0]));
+ return 1;
+ case LELONG:
+ case LEDATE:
+ p->l = (long)
+ ((p->hl[3] << 24) | (p->hl[2] << 16) | (p->hl[1] << 8) | (p->hl[0]));
+ return 1;
+ default:
+ php_error(E_WARNING,
+ MODNAME ": invalid type %d in mconvert().", m->type);
+ return 0;
+ }
+}
+
+
+/* states for the state-machine algorithm in magic_rsl_to_request() */
+typedef enum {
+ rsl_leading_space, rsl_type, rsl_subtype, rsl_separator, rsl_encoding
+} rsl_states;
+
+/* process the RSL and set the MIME info in the request record */
+static int magic_rsl_get(char **content_type, char **content_encoding)
+{
+ int cur_frag, /* current fragment number/counter */
+ cur_pos, /* current position within fragment */
+ type_frag, /* content type starting point: fragment */
+ type_pos, /* content type starting point: position */
+ type_len, /* content type length */
+ encoding_frag, /* content encoding starting point: fragment */
+ encoding_pos, /* content encoding starting point: position */
+ encoding_len; /* content encoding length */
+
+ magic_rsl *frag; /* list-traversal pointer */
+ rsl_states state;
+
+ magic_req_rec *req_dat = MIME_MAGIC_G(req_dat);
+
+ /* check if we have a result */
+ if (!req_dat || !req_dat->head) {
+ /* empty - no match, we defer to other Apache modules */
+ return DECLINED;
+ }
+
+ /* start searching for the type and encoding */
+ state = rsl_leading_space;
+ type_frag = type_pos = type_len = 0;
+ encoding_frag = encoding_pos = encoding_len = 0;
+ for (frag = req_dat->head, cur_frag = 0;
+ frag && frag->next;
+ frag = frag->next, cur_frag++) {
+ /* loop through the characters in the fragment */
+ for (cur_pos = 0; frag->str[cur_pos]; cur_pos++) {
+ if (isspace(frag->str[cur_pos])) {
+ /* process whitespace actions for each state */
+ if (state == rsl_leading_space) {
+ /* eat whitespace in this state */
+ continue;
+ }
+ else if (state == rsl_type) {
+ /* whitespace: type has no slash! */
+ return DECLINED;
+ }
+ else if (state == rsl_subtype) {
+ /* whitespace: end of MIME type */
+ state++;
+ continue;
+ }
+ else if (state == rsl_separator) {
+ /* eat whitespace in this state */
+ continue;
+ }
+ else if (state == rsl_encoding) {
+ /* whitespace: end of MIME encoding */
+ /* we're done */
+ frag = req_dat->tail;
+ break;
+ }
+ else {
+ /* should not be possible */
+ /* abandon malfunctioning module */
+ php_error(E_WARNING,
+ MODNAME ": bad state %d (ws)", state);
+ return DECLINED;
+ }
+ /* NOTREACHED */
+ }
+ else if (state == rsl_type &&
+ frag->str[cur_pos] == '/') {
+ /* copy the char and go to rsl_subtype state */
+ type_len++;
+ state++;
+ }
+ else {
+ /* process non-space actions for each state */
+ if (state == rsl_leading_space) {
+ /* non-space: begin MIME type */
+ state++;
+ type_frag = cur_frag;
+ type_pos = cur_pos;
+ type_len = 1;
+ continue;
+ }
+ else if (state == rsl_type ||
+ state == rsl_subtype) {
+ /* non-space: adds to type */
+ type_len++;
+ continue;
+ }
+ else if (state == rsl_separator) {
+ /* non-space: begin MIME encoding */
+ state++;
+ encoding_frag = cur_frag;
+ encoding_pos = cur_pos;
+ encoding_len = 1;
+ continue;
+ }
+ else if (state == rsl_encoding) {
+ /* non-space: adds to encoding */
+ encoding_len++;
+ continue;
+ }
+ else {
+ /* should not be possible */
+ /* abandon malfunctioning module */
+ php_error(E_WARNING,
+ MODNAME ": bad state %d (ns)", state);
+ return DECLINED;
+ }
+ /* NOTREACHED */
+ }
+ /* NOTREACHED */
+ }
+ }
+
+ /* if we ended prior to state rsl_subtype, we had incomplete info */
+ if (state != rsl_subtype && state != rsl_separator &&
+ state != rsl_encoding) {
+ /* defer to other modules */
+ return DECLINED;
+ }
+
+ /* save the info in the request record */
+ if (state == rsl_subtype || state == rsl_encoding ||
+ state == rsl_encoding) {
+ char *tmp;
+ tmp = rsl_strdup(type_frag, type_pos, type_len);
+ /* XXX: this could be done at config time I'm sure... but I'm
+ * confused by all this magic_rsl stuff. -djg */
+ /* TODO ap_content_type_tolower(tmp); */
+ *content_type = tmp;
+ }
+ if (state == rsl_encoding) {
+ char *tmp;
+ tmp = rsl_strdup(encoding_frag,
+ encoding_pos, encoding_len);
+ /* XXX: this could be done at config time I'm sure... but I'm
+ * confused by all this magic_rsl stuff. -djg */
+ /* TODO ap_str_tolower(tmp); */
+ *content_encoding = tmp;
+ }
+
+ /* detect memory allocation errors */
+ if (!content_type ||
+ (state == rsl_encoding && !*content_encoding)) {
+ return ERROR;
+ }
+
+ /* success! */
+ return OK;
+}
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * End:
+ * vim600: noet sw=4 ts=4 fdm=marker
+ * vim<600: noet sw=4 ts=4
+ */
diff --git a/ext/mime_magic/php_mime_magic.h b/ext/mime_magic/php_mime_magic.h
new file mode 100644
index 0000000000..60a9e4e64d
--- /dev/null
+++ b/ext/mime_magic/php_mime_magic.h
@@ -0,0 +1,454 @@
+/*
+ +----------------------------------------------------------------------+
+ | PHP Version 4 |
+ +----------------------------------------------------------------------+
+ | Copyright (c) 1997-2002 The PHP Group |
+ +----------------------------------------------------------------------+
+ | This source file is subject to version 2.02 of the PHP license, |
+ | that is bundled with this package in the file LICENSE, and is |
+ | available at through the world-wide-web at |
+ | http://www.php.net/license/2_02.txt. |
+ | If you did not receive a copy of the PHP license and are unable to |
+ | obtain it through the world-wide-web, please send a note to |
+ | license@php.net so we can mail you a copy immediately. |
+ +----------------------------------------------------------------------+
+ | Author: |
+ +----------------------------------------------------------------------+
+
+ $Id$
+*/
+
+#ifndef PHP_MIME_MAGIC_H
+#define PHP_MIME_MAGIC_H
+
+#define MIME_MAGIC_DEBUG 0
+
+#define MIME_BINARY_UNKNOWN "application/octet-stream"
+#define MIME_TEXT_UNKNOWN "text/plain"
+
+#define MAXMIMESTRING 256
+
+#define OK 0
+#define DECLINED -1
+#define DONE -2
+#define ERROR -3
+
+#define DIR_MAGIC_TYPE "httpd/unix-directory"
+
+/* HOWMANY must be at least 4096 to make gzip -dcq work */
+#define HOWMANY 4096
+/* SMALL_HOWMANY limits how much work we do to figure out text files */
+#define SMALL_HOWMANY 1024
+#define MAXDESC 50 /* max leng of text description */
+#define MAXstring 64 /* max leng of "string" types */
+
+struct magic {
+ struct magic *next; /* link to next entry */
+ int lineno; /* line number from magic file */
+
+ short flag;
+#define INDIR 1 /* if '>(...)' appears, */
+#define UNSIGNED 2 /* comparison is unsigned */
+ short cont_level; /* level of ">" */
+ struct {
+ char type; /* byte short long */
+ long offset; /* offset from indirection */
+ } in;
+ long offset; /* offset to magic number */
+ unsigned char reln; /* relation (0=eq, '>'=gt, etc) */
+ char type; /* int, short, long or string. */
+ char vallen; /* length of string value, if any */
+#define BYTE 1
+#define SHORT 2
+#define LONG 4
+#define STRING 5
+#define DATE 6
+#define BESHORT 7
+#define BELONG 8
+#define BEDATE 9
+#define LESHORT 10
+#define LELONG 11
+#define LEDATE 12
+ union VALUETYPE {
+ unsigned char b;
+ unsigned short h;
+ unsigned long l;
+ char s[MAXstring];
+ unsigned char hs[2]; /* 2 bytes of a fixed-endian "short" */
+ unsigned char hl[4]; /* 2 bytes of a fixed-endian "long" */
+ } value; /* either number or string */
+ unsigned long mask; /* mask before comparison with value */
+ char nospflag; /* supress space character */
+
+ /* NOTE: this string is suspected of overrunning - find it! */
+ char desc[MAXDESC]; /* description */
+};
+
+/*
+ * data structures for tar file recognition
+ * --------------------------------------------------------------------------
+ * Header file for public domain tar (tape archive) program.
+ *
+ * @(#)tar.h 1.20 86/10/29 Public Domain. Created 25 August 1985 by John
+ * Gilmore, ihnp4!hoptoad!gnu.
+ *
+ * Header block on tape.
+ *
+ * I'm going to use traditional DP naming conventions here. A "block" is a big
+ * chunk of stuff that we do I/O on. A "record" is a piece of info that we
+ * care about. Typically many "record"s fit into a "block".
+ */
+#define RECORDSIZE 512
+#define NAMSIZ 100
+#define TUNMLEN 32
+#define TGNMLEN 32
+
+union record {
+ char charptr[RECORDSIZE];
+ struct header {
+ char name[NAMSIZ];
+ char mode[8];
+ char uid[8];
+ char gid[8];
+ char size[12];
+ char mtime[12];
+ char chksum[8];
+ char linkflag;
+ char linkname[NAMSIZ];
+ char magic[8];
+ char uname[TUNMLEN];
+ char gname[TGNMLEN];
+ char devmajor[8];
+ char devminor[8];
+ } header;
+};
+
+/* The magic field is filled with this if uname and gname are valid. */
+#define TMAGIC "ustar " /* 7 chars and a null */
+
+/*
+ * file-function prototypes
+ */
+/*TODO*/
+#define request_rec void
+#define server_rec void
+#define pool void
+#define cmd_parms void
+
+static int apprentice(void);
+static int ascmagic(unsigned char *, int);
+static int is_tar(unsigned char *, int);
+static int softmagic(unsigned char *, int);
+static void tryit(unsigned char *, int, int);
+static int zmagic(unsigned char *, int);
+
+static int getvalue(struct magic *, char **);
+static int hextoint(int);
+static char *getstr(char *, char *, int, int *);
+static int parse(char *, int);
+
+static int match(unsigned char *, int);
+static int mget(union VALUETYPE *, unsigned char *,
+ struct magic *, int);
+static int mcheck(union VALUETYPE *, struct magic *);
+static void mprint(union VALUETYPE *, struct magic *);
+static int mconvert(union VALUETYPE *, struct magic *);
+static int magic_rsl_get(char **, char **);
+static int magic_process(char *);
+
+static int uncompress(int,
+ unsigned char **, int);
+static long from_oct(int, char *);
+static int fsmagic(const char *fn);
+
+/*
+ * includes for ASCII substring recognition formerly "names.h" in file
+ * command
+ *
+ * Original notes: names and types used by ascmagic in file(1). These tokens are
+ * here because they can appear anywhere in the first HOWMANY bytes, while
+ * tokens in /etc/magic must appear at fixed offsets into the file. Don't
+ * make HOWMANY too high unless you have a very fast CPU.
+ */
+
+/* these types are used to index the table 'types': keep em in sync! */
+/* HTML inserted in first because this is a web server module now */
+#define L_HTML 0 /* HTML */
+#define L_C 1 /* first and foremost on UNIX */
+#define L_FORT 2 /* the oldest one */
+#define L_MAKE 3 /* Makefiles */
+#define L_PLI 4 /* PL/1 */
+#define L_MACH 5 /* some kinda assembler */
+#define L_ENG 6 /* English */
+#define L_PAS 7 /* Pascal */
+#define L_MAIL 8 /* Electronic mail */
+#define L_NEWS 9 /* Usenet Netnews */
+
+static char *types[] =
+ {
+ "text/html", /* HTML */
+ "text/plain", /* "c program text", */
+ "text/plain", /* "fortran program text", */
+ "text/plain", /* "make commands text", */
+ "text/plain", /* "pl/1 program text", */
+ "text/plain", /* "assembler program text", */
+ "text/plain", /* "English text", */
+ "text/plain", /* "pascal program text", */
+ "message/rfc822", /* "mail text", */
+ "message/news", /* "news text", */
+ "application/binary", /* "can't happen error on names.h/types", */
+ 0
+ };
+
+static struct names {
+ char *name;
+ short type;
+} names[] = {
+
+ /* These must be sorted by eye for optimal hit rate */
+ /* Add to this list only after substantial meditation */
+ {
+ "<html>", L_HTML
+ },
+ {
+ "<HTML>", L_HTML
+ },
+ {
+ "<head>", L_HTML
+ },
+ {
+ "<HEAD>", L_HTML
+ },
+ {
+ "<title>", L_HTML
+ },
+ {
+ "<TITLE>", L_HTML
+ },
+ {
+ "<h1>", L_HTML
+ },
+ {
+ "<H1>", L_HTML
+ },
+ {
+ "<!--", L_HTML
+ },
+ {
+ "<!DOCTYPE HTML", L_HTML
+ },
+ {
+ "/*", L_C
+ }, /* must precede "The", "the", etc. */
+ {
+ "#include", L_C
+ },
+ {
+ "char", L_C
+ },
+ {
+ "The", L_ENG
+ },
+ {
+ "the", L_ENG
+ },
+ {
+ "double", L_C
+ },
+ {
+ "extern", L_C
+ },
+ {
+ "float", L_C
+ },
+ {
+ "real", L_C
+ },
+ {
+ "struct", L_C
+ },
+ {
+ "union", L_C
+ },
+ {
+ "CFLAGS", L_MAKE
+ },
+ {
+ "LDFLAGS", L_MAKE
+ },
+ {
+ "all:", L_MAKE
+ },
+ {
+ ".PRECIOUS", L_MAKE
+ },
+ /*
+ * Too many files of text have these words in them. Find another way to
+ * recognize Fortrash.
+ */
+#ifdef NOTDEF
+ {
+ "subroutine", L_FORT
+ },
+ {
+ "function", L_FORT
+ },
+ {
+ "block", L_FORT
+ },
+ {
+ "common", L_FORT
+ },
+ {
+ "dimension", L_FORT
+ },
+ {
+ "integer", L_FORT
+ },
+ {
+ "data", L_FORT
+ },
+#endif /* NOTDEF */
+ {
+ ".ascii", L_MACH
+ },
+ {
+ ".asciiz", L_MACH
+ },
+ {
+ ".byte", L_MACH
+ },
+ {
+ ".even", L_MACH
+ },
+ {
+ ".globl", L_MACH
+ },
+ {
+ "clr", L_MACH
+ },
+ {
+ "(input,", L_PAS
+ },
+ {
+ "dcl", L_PLI
+ },
+ {
+ "Received:", L_MAIL
+ },
+ {
+ ">From", L_MAIL
+ },
+ {
+ "Return-Path:", L_MAIL
+ },
+ {
+ "Cc:", L_MAIL
+ },
+ {
+ "Newsgroups:", L_NEWS
+ },
+ {
+ "Path:", L_NEWS
+ },
+ {
+ "Organization:", L_NEWS
+ },
+ {
+ NULL, 0
+ }
+};
+
+#define NNAMES ((sizeof(names)/sizeof(struct names)) - 1)
+
+/*
+ * Result String List (RSL)
+ *
+ * The file(1) command prints its output. Instead, we store the various
+ * "printed" strings in a list (allocating memory as we go) and concatenate
+ * them at the end when we finally know how much space they'll need.
+ */
+
+typedef struct magic_rsl_s {
+ char *str; /* string, possibly a fragment */
+ struct magic_rsl_s *next; /* pointer to next fragment */
+} magic_rsl;
+
+/*
+ * Apache module configuration structures
+ */
+
+/* per-server info */
+typedef struct {
+ char *magicfile; /* where magic be found */
+ struct magic *magic; /* head of magic config list */
+ struct magic *last;
+} magic_server_config_rec;
+
+/* per-request info */
+typedef struct {
+ magic_rsl *head; /* result string list */
+ magic_rsl *tail;
+ unsigned suf_recursion; /* recursion depth in suffix check */
+} magic_req_rec;
+
+static magic_req_rec *magic_set_config(void);
+static void magic_free_config(magic_req_rec *);
+
+
+
+extern zend_module_entry mime_magic_module_entry;
+#define phpext_mime_magic_ptr &mime_magic_module_entry
+
+#ifdef PHP_WIN32
+#define PHP_MIME_MAGIC_API __declspec(dllexport)
+#else
+#define PHP_MIME_MAGIC_API
+#endif
+
+#ifdef ZTS
+#include "TSRM.h"
+#endif
+
+PHP_MINIT_FUNCTION(mime_magic);
+PHP_MSHUTDOWN_FUNCTION(mime_magic);
+PHP_RINIT_FUNCTION(mime_magic);
+PHP_RSHUTDOWN_FUNCTION(mime_magic);
+PHP_MINFO_FUNCTION(mime_magic);
+
+PHP_FUNCTION(mime_content_type);
+
+ZEND_BEGIN_MODULE_GLOBALS(mime_magic)
+ char *magicfile;
+ magic_req_rec *req_dat;
+ZEND_END_MODULE_GLOBALS(mime_magic)
+
+/* In every utility function you add that needs to use variables
+ in php_mime_magic_globals, call TSRM_FETCH(); after declaring other
+ variables used by that function, or better yet, pass in TSRMG_CC
+ after the last function argument and declare your utility function
+ with TSRMG_DC after the last declared argument. Always refer to
+ the globals in your function as MIME_MAGIC_G(variable). You are
+ encouraged to rename these macros something shorter, see
+ examples in any other php module directory.
+*/
+
+#ifdef ZTS
+#define MIME_MAGIC_G(v) TSRMG(mime_magic_globals_id, zend_mime_magic_globals *, v)
+#else
+#define MIME_MAGIC_G(v) (mime_magic_globals.v)
+#endif
+
+
+
+
+#endif /* PHP_MIME_MAGIC_H */
+
+
+/*
+ * Local variables:
+ * tab-width: 4
+ * c-basic-offset: 4
+ * indent-tabs-mode: t
+ * End:
+ */