gimp/app/file/file-utils.c

592 lines
16 KiB
C

/* The GIMP -- an image manipulation program
* Copyright (C) 1995, 1996, 1997 Spencer Kimball and Peter Mattis
* Copyright (C) 1997 Josh MacDonald
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*/
#include "config.h"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#ifdef HAVE_SYS_PARAM_H
#include <sys/param.h>
#endif
#include <sys/types.h>
#include <sys/stat.h>
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif
#include <glib-object.h>
#include "libgimpmath/gimpmath.h"
#include "core/core-types.h"
#include "base/temp-buf.h"
#include "core/gimp.h"
#include "core/gimpimage.h"
#include "pdb/procedural_db.h"
#include "plug-in/plug-in.h"
#include "plug-in/plug-in-proc.h"
#include "file-utils.h"
#include "gimp-intl.h"
/* local function prototypes */
static PlugInProcDef * file_proc_find_by_prefix (GSList *procs,
const gchar *uri,
gboolean skip_magic);
static PlugInProcDef * file_proc_find_by_extension (GSList *procs,
const gchar *uri,
gboolean skip_magic);
static PlugInProcDef * file_proc_find_by_name (GSList *procs,
const gchar *uri,
gboolean skip_magic);
static void file_convert_string (gchar *instr,
gchar *outmem,
gint maxmem,
gint *nmem);
static gint file_check_single_magic (gchar *offset,
gchar *type,
gchar *value,
gint headsize,
guchar *file_head,
FILE *ifp);
static gint file_check_magic_list (GSList *magics_list,
gint headsize,
guchar *head,
FILE *ifp);
/* public functions */
gchar *
file_utils_filename_to_uri (GSList *procs,
const gchar *filename,
GError **error)
{
gchar *absolute;
gchar *uri;
g_return_val_if_fail (procs != NULL, NULL);
g_return_val_if_fail (filename != NULL, NULL);
/* check for prefixes like http or ftp */
if (file_proc_find_by_prefix (procs, filename, FALSE))
{
if (g_utf8_validate (filename, -1, NULL))
{
return g_strdup (filename);
}
else
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
_("Invalid character sequence in URI"));
return NULL;
}
}
if (! g_path_is_absolute (filename))
{
gchar *current;
current = g_get_current_dir ();
absolute = g_build_filename (current, filename, NULL);
g_free (current);
}
else
{
absolute = g_strdup (filename);
}
uri = g_filename_to_uri (absolute, NULL, error);
g_free (absolute);
return uri;
}
PlugInProcDef *
file_utils_find_proc (GSList *procs,
const gchar *uri)
{
PlugInProcDef *file_proc;
GSList *all_procs = procs;
gchar *filename;
g_return_val_if_fail (procs != NULL, NULL);
g_return_val_if_fail (uri != NULL, NULL);
/* First, check magicless prefixes/suffixes */
file_proc = file_proc_find_by_name (all_procs, uri, TRUE);
if (file_proc)
return file_proc;
filename = g_filename_from_uri (uri, NULL, NULL);
/* Then look for magics */
if (filename)
{
PlugInProcDef *size_matched_proc = NULL;
FILE *ifp = NULL;
gint head_size = -2;
gint size_match_count = 0;
gint match_val;
guchar head[256];
while (procs)
{
file_proc = procs->data;
procs = procs->next;
if (file_proc->magics_list)
{
if (head_size == -2)
{
head_size = 0;
if ((ifp = fopen (filename, "rb")) != NULL)
head_size = fread ((gchar *) head, 1, sizeof (head), ifp);
}
if (head_size >= 4)
{
match_val = file_check_magic_list (file_proc->magics_list,
head_size, head, ifp);
if (match_val == 2) /* size match ? */
{ /* Use it only if no other magic matches */
size_match_count++;
size_matched_proc = file_proc;
}
else if (match_val)
{
fclose (ifp);
g_free (filename);
return file_proc;
}
}
}
}
if (ifp)
fclose (ifp);
g_free (filename);
if (size_match_count == 1)
return size_matched_proc;
}
/* As a last ditch, try matching by name */
return file_proc_find_by_name (all_procs, uri, FALSE);
}
gchar *
file_utils_uri_to_utf8_basename (const gchar *uri)
{
gchar *filename;
gchar *basename;
g_assert (uri != NULL);
g_return_val_if_fail (uri != NULL, NULL);
filename = file_utils_uri_to_utf8_filename (uri);
if (strstr (filename, G_DIR_SEPARATOR_S))
{
basename = g_path_get_basename (filename);
g_free (filename);
return basename;
}
else if (strstr (filename, "://"))
{
basename = strrchr (uri, '/');
basename = g_strdup (basename + 1);
g_free (filename);
return basename;
}
return filename;
}
gchar *
file_utils_uri_to_utf8_filename (const gchar *uri)
{
g_return_val_if_fail (uri != NULL, NULL);
if (! strncmp (uri, "file:", strlen ("file:")))
{
gchar *filename;
filename = g_filename_from_uri (uri, NULL, NULL);
if (filename)
{
gchar *utf8 = g_filename_to_utf8 (filename, -1, NULL, NULL, NULL);
g_free (filename);
return utf8;
}
}
return g_strdup (uri);
}
/* private functions */
static PlugInProcDef *
file_proc_find_by_prefix (GSList *procs,
const gchar *uri,
gboolean skip_magic)
{
GSList *p;
for (p = procs; p; p = g_slist_next (p))
{
PlugInProcDef *proc = p->data;
GSList *prefixes;
if (skip_magic && proc->magics_list)
continue;
for (prefixes = proc->prefixes_list;
prefixes;
prefixes = g_slist_next (prefixes))
{
if (strncmp (uri, prefixes->data, strlen (prefixes->data)) == 0)
return proc;
}
}
return NULL;
}
static PlugInProcDef *
file_proc_find_by_extension (GSList *procs,
const gchar *uri,
gboolean skip_magic)
{
GSList *p;
gchar *ext;
ext = strrchr (uri, '.');
if (ext)
ext++;
for (p = procs; p; p = g_slist_next (p))
{
PlugInProcDef *proc = p->data;
GSList *extensions;
for (extensions = proc->extensions_list;
ext && extensions;
extensions = g_slist_next (extensions))
{
gchar *p1 = ext;
gchar *p2 = (gchar *) extensions->data;
if (skip_magic && proc->magics_list)
continue;
while (*p1 && *p2)
{
if (g_ascii_tolower (*p1) != g_ascii_tolower (*p2))
break;
p1++;
p2++;
}
if (!(*p1) && !(*p2))
return proc;
}
}
return NULL;
}
static PlugInProcDef *
file_proc_find_by_name (GSList *procs,
const gchar *uri,
gboolean skip_magic)
{
PlugInProcDef *proc;
proc = file_proc_find_by_prefix (procs, uri, skip_magic);
if (! proc)
proc = file_proc_find_by_extension (procs, uri, skip_magic);
return proc;
}
static void
file_convert_string (gchar *instr,
gchar *outmem,
gint maxmem,
gint *nmem)
{
/* Convert a string in C-notation to array of char */
guchar *uin = (guchar *) instr;
guchar *uout = (guchar *) outmem;
guchar tmp[5], *tmpptr;
guint k;
while ((*uin != '\0') && ((((gchar *) uout) - outmem) < maxmem))
{
if (*uin != '\\') /* Not an escaped character ? */
{
*(uout++) = *(uin++);
continue;
}
if (*(++uin) == '\0')
{
*(uout++) = '\\';
break;
}
switch (*uin)
{
case '0': case '1': case '2': case '3': /* octal */
for (tmpptr = tmp; (tmpptr-tmp) <= 3;)
{
*(tmpptr++) = *(uin++);
if ( (*uin == '\0') || (!g_ascii_isdigit (*uin))
|| (*uin == '8') || (*uin == '9'))
break;
}
*tmpptr = '\0';
sscanf ((gchar *) tmp, "%o", &k);
*(uout++) = k;
break;
case 'a': *(uout++) = '\a'; uin++; break;
case 'b': *(uout++) = '\b'; uin++; break;
case 't': *(uout++) = '\t'; uin++; break;
case 'n': *(uout++) = '\n'; uin++; break;
case 'v': *(uout++) = '\v'; uin++; break;
case 'f': *(uout++) = '\f'; uin++; break;
case 'r': *(uout++) = '\r'; uin++; break;
default : *(uout++) = *(uin++); break;
}
}
*nmem = ((gchar *) uout) - outmem;
}
static gint
file_check_single_magic (gchar *offset,
gchar *type,
gchar *value,
gint headsize,
guchar *file_head,
FILE *ifp)
{
/* Return values are 0: no match, 1: magic match, 2: size match */
glong offs;
gulong num_testval, num_operatorval;
gulong fileval;
gint numbytes, k, c = 0, found = 0;
gchar *num_operator_ptr, num_operator, num_test;
guchar mem_testval[256];
/* Check offset */
if (sscanf (offset, "%ld", &offs) != 1) return (0);
if (offs < 0) return (0);
/* Check type of test */
num_operator_ptr = NULL;
num_operator = '\0';
num_test = '=';
if (strncmp (type, "byte", 4) == 0)
{
numbytes = 1;
num_operator_ptr = type+4;
}
else if (strncmp (type, "short", 5) == 0)
{
numbytes = 2;
num_operator_ptr = type+5;
}
else if (strncmp (type, "long", 4) == 0)
{
numbytes = 4;
num_operator_ptr = type+4;
}
else if (strncmp (type, "size", 4) == 0)
{
numbytes = 5;
}
else if (strcmp (type, "string") == 0)
{
numbytes = 0;
}
else return (0);
/* Check numerical operator value if present */
if (num_operator_ptr && (*num_operator_ptr == '&'))
{
if (g_ascii_isdigit (num_operator_ptr[1]))
{
if (num_operator_ptr[1] != '0') /* decimal */
sscanf (num_operator_ptr+1, "%lu", &num_operatorval);
else if (num_operator_ptr[2] == 'x') /* hexadecimal */
sscanf (num_operator_ptr+3, "%lx", &num_operatorval);
else /* octal */
sscanf (num_operator_ptr+2, "%lo", &num_operatorval);
num_operator = *num_operator_ptr;
}
}
if (numbytes > 0) /* Numerical test ? */
{
/* Check test value */
if ((value[0] == '=') || (value[0] == '>') || (value[0] == '<'))
{
num_test = value[0];
value++;
}
if (!g_ascii_isdigit (value[0])) return (0);
/*
* to anybody reading this: is strtol's parsing behaviour
* (e.g. "0x" prefix) broken on some systems or why do we
* do the base detection ourselves?
* */
if (value[0] != '0') /* decimal */
num_testval = strtol(value, NULL, 10);
else if (value[1] == 'x') /* hexadecimal */
num_testval = (unsigned long)strtoul(value+2, NULL, 16);
else /* octal */
num_testval = strtol(value+1, NULL, 8);
fileval = 0;
if (numbytes == 5) /* Check for file size ? */
{
struct stat buf;
if (fstat (fileno (ifp), &buf) < 0) return (0);
fileval = buf.st_size;
}
else if (offs + numbytes <= headsize) /* We have it in memory ? */
{
for (k = 0; k < numbytes; k++)
fileval = (fileval << 8) | (long)file_head[offs+k];
}
else /* Read it from file */
{
if (fseek (ifp, offs, SEEK_SET) < 0) return (0);
for (k = 0; k < numbytes; k++)
fileval = (fileval << 8) | (c = getc (ifp));
if (c == EOF) return (0);
}
if (num_operator == '&')
fileval &= num_operatorval;
if (num_test == '<')
found = (fileval < num_testval);
else if (num_test == '>')
found = (fileval > num_testval);
else
found = (fileval == num_testval);
if (found && (numbytes == 5)) found = 2;
}
else if (numbytes == 0) /* String test */
{
file_convert_string ((char *)value, (char *)mem_testval,
sizeof (mem_testval), &numbytes);
if (numbytes <= 0) return (0);
if (offs + numbytes <= headsize) /* We have it in memory ? */
{
found = (memcmp (mem_testval, file_head+offs, numbytes) == 0);
}
else /* Read it from file */
{
if (fseek (ifp, offs, SEEK_SET) < 0) return (0);
found = 1;
for (k = 0; found && (k < numbytes); k++)
{
c = getc (ifp);
found = (c != EOF) && (c == (int)mem_testval[k]);
}
}
}
return found;
}
/*
* Return values are 0: no match, 1: magic match, 2: size match
*/
static gint
file_check_magic_list (GSList *magics_list,
gint headsize,
guchar *head,
FILE *ifp)
{
gchar *offset;
gchar *type;
gchar *value;
gint and = 0;
gint found = 0;
gint match_val;
while (magics_list)
{
if ((offset = (gchar *)magics_list->data) == NULL) break;
if ((magics_list = magics_list->next) == NULL) break;
if ((type = (gchar *)magics_list->data) == NULL) break;
if ((magics_list = magics_list->next) == NULL) break;
if ((value = (gchar *)magics_list->data) == NULL) break;
magics_list = magics_list->next;
match_val = file_check_single_magic (offset, type, value,
headsize, head, ifp);
if (and)
found = found && match_val;
else
found = match_val;
and = (strchr (offset, '&') != NULL);
if ((!and) && found)
return match_val;
}
return 0;
}