#include "config.h"
#include <iconv.h>
#include <errno.h>
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include "glib.h"
#include "gprintfint.h"
#include "gthreadinit.h"
#ifdef G_PLATFORM_WIN32
#define STRICT
#include <windows.h>
#undef STRICT
#endif
#include "glibintl.h"
#if defined(USE_LIBICONV_GNU) && !defined (_LIBICONV_H)
#error GNU libiconv in use but included iconv.h not from libiconv
#endif
#if !defined(USE_LIBICONV_GNU) && defined (_LIBICONV_H)
#error GNU libiconv not in use but included iconv.h is from libiconv
#endif
GQuark
g_convert_error_quark (void)
{
static GQuark quark;
if (!quark)
quark = g_quark_from_static_string ("g_convert_error");
return quark;
}
static gboolean
try_conversion (const char *to_codeset,
const char *from_codeset,
iconv_t *cd)
{
*cd = iconv_open (to_codeset, from_codeset);
if (*cd == (iconv_t)-1 && errno == EINVAL)
return FALSE;
else
return TRUE;
}
static gboolean
try_to_aliases (const char **to_aliases,
const char *from_codeset,
iconv_t *cd)
{
if (to_aliases)
{
const char **p = to_aliases;
while (*p)
{
if (try_conversion (*p, from_codeset, cd))
return TRUE;
p++;
}
}
return FALSE;
}
extern const char **_g_charset_get_aliases (const char *canonical_name);
GIConv
g_iconv_open (const gchar *to_codeset,
const gchar *from_codeset)
{
iconv_t cd;
if (!try_conversion (to_codeset, from_codeset, &cd))
{
const char **to_aliases = _g_charset_get_aliases (to_codeset);
const char **from_aliases = _g_charset_get_aliases (from_codeset);
if (from_aliases)
{
const char **p = from_aliases;
while (*p)
{
if (try_conversion (to_codeset, *p, &cd))
goto out;
if (try_to_aliases (to_aliases, *p, &cd))
goto out;
p++;
}
}
if (try_to_aliases (to_aliases, from_codeset, &cd))
goto out;
}
out:
return (cd == (iconv_t)-1) ? (GIConv)-1 : (GIConv)cd;
}
size_t
g_iconv (GIConv converter,
gchar **inbuf,
gsize *inbytes_left,
gchar **outbuf,
gsize *outbytes_left)
{
iconv_t cd = (iconv_t)converter;
return iconv (cd, inbuf, inbytes_left, outbuf, outbytes_left);
}
gint
g_iconv_close (GIConv converter)
{
iconv_t cd = (iconv_t)converter;
return iconv_close (cd);
}
#define ICONV_CACHE_SIZE (16)
struct _iconv_cache_bucket {
gchar *key;
guint32 refcount;
gboolean used;
GIConv cd;
};
static GList *iconv_cache_list;
static GHashTable *iconv_cache;
static GHashTable *iconv_open_hash;
static guint iconv_cache_size = 0;
G_LOCK_DEFINE_STATIC (iconv_cache_lock);
static void
iconv_cache_init (void)
{
static gboolean initialized = FALSE;
if (initialized)
return;
iconv_cache_list = NULL;
iconv_cache = g_hash_table_new (g_str_hash, g_str_equal);
iconv_open_hash = g_hash_table_new (g_direct_hash, g_direct_equal);
initialized = TRUE;
}
static struct _iconv_cache_bucket *
iconv_cache_bucket_new (const gchar *key, GIConv cd)
{
struct _iconv_cache_bucket *bucket;
bucket = g_new (struct _iconv_cache_bucket, 1);
bucket->key = g_strdup (key);
bucket->refcount = 1;
bucket->used = TRUE;
bucket->cd = cd;
g_hash_table_insert (iconv_cache, bucket->key, bucket);
iconv_cache_list = g_list_prepend (iconv_cache_list, bucket);
iconv_cache_size++;
return bucket;
}
static void
iconv_cache_bucket_expire (GList *node, struct _iconv_cache_bucket *bucket)
{
g_hash_table_remove (iconv_cache, bucket->key);
if (node == NULL)
node = g_list_find (iconv_cache_list, bucket);
g_assert (node != NULL);
if (node->prev)
{
node->prev->next = node->next;
if (node->next)
node->next->prev = node->prev;
}
else
{
iconv_cache_list = node->next;
if (node->next)
node->next->prev = NULL;
}
g_list_free_1 (node);
g_free (bucket->key);
g_iconv_close (bucket->cd);
g_free (bucket);
iconv_cache_size--;
}
static void
iconv_cache_expire_unused (void)
{
struct _iconv_cache_bucket *bucket;
GList *node, *next;
node = iconv_cache_list;
while (node && iconv_cache_size >= ICONV_CACHE_SIZE)
{
next = node->next;
bucket = node->data;
if (bucket->refcount == 0)
iconv_cache_bucket_expire (node, bucket);
node = next;
}
}
static GIConv
open_converter (const gchar *to_codeset,
const gchar *from_codeset,
GError **error)
{
struct _iconv_cache_bucket *bucket;
gchar *key;
GIConv cd;
key = g_alloca (strlen (from_codeset) + strlen (to_codeset) + 2);
_g_sprintf (key, "%s:%s", from_codeset, to_codeset);
G_LOCK (iconv_cache_lock);
iconv_cache_init ();
bucket = g_hash_table_lookup (iconv_cache, key);
if (bucket)
{
if (bucket->used)
{
cd = g_iconv_open (to_codeset, from_codeset);
if (cd == (GIConv) -1)
goto error;
}
else
{
gsize inbytes_left = 0;
gchar *outbuf = NULL;
gsize outbytes_left = 0;
cd = bucket->cd;
bucket->used = TRUE;
g_iconv (cd, NULL, &inbytes_left, &outbuf, &outbytes_left);
}
bucket->refcount++;
}
else
{
cd = g_iconv_open (to_codeset, from_codeset);
if (cd == (GIConv) -1)
goto error;
iconv_cache_expire_unused ();
bucket = iconv_cache_bucket_new (key, cd);
}
g_hash_table_insert (iconv_open_hash, cd, bucket->key);
G_UNLOCK (iconv_cache_lock);
return cd;
error:
G_UNLOCK (iconv_cache_lock);
if (errno == EINVAL)
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_NO_CONVERSION,
_("Conversion from character set '%s' to '%s' is not supported"),
from_codeset, to_codeset);
else
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_FAILED,
_("Could not open converter from '%s' to '%s'"),
from_codeset, to_codeset);
return cd;
}
static int
close_converter (GIConv converter)
{
struct _iconv_cache_bucket *bucket;
const gchar *key;
GIConv cd;
cd = converter;
if (cd == (GIConv) -1)
return 0;
G_LOCK (iconv_cache_lock);
key = g_hash_table_lookup (iconv_open_hash, cd);
if (key)
{
g_hash_table_remove (iconv_open_hash, cd);
bucket = g_hash_table_lookup (iconv_cache, key);
g_assert (bucket);
bucket->refcount--;
if (cd == bucket->cd)
bucket->used = FALSE;
else
g_iconv_close (cd);
if (!bucket->refcount && iconv_cache_size > ICONV_CACHE_SIZE)
{
iconv_cache_bucket_expire (NULL, bucket);
}
}
else
{
G_UNLOCK (iconv_cache_lock);
g_warning ("This iconv context wasn't opened using open_converter");
return g_iconv_close (converter);
}
G_UNLOCK (iconv_cache_lock);
return 0;
}
gchar*
g_convert (const gchar *str,
gssize len,
const gchar *to_codeset,
const gchar *from_codeset,
gsize *bytes_read,
gsize *bytes_written,
GError **error)
{
gchar *res;
GIConv cd;
g_return_val_if_fail (str != NULL, NULL);
g_return_val_if_fail (to_codeset != NULL, NULL);
g_return_val_if_fail (from_codeset != NULL, NULL);
cd = open_converter (to_codeset, from_codeset, error);
if (cd == (GIConv) -1)
{
if (bytes_read)
*bytes_read = 0;
if (bytes_written)
*bytes_written = 0;
return NULL;
}
res = g_convert_with_iconv (str, len, cd,
bytes_read, bytes_written,
error);
close_converter (cd);
return res;
}
gchar*
g_convert_with_iconv (const gchar *str,
gssize len,
GIConv converter,
gsize *bytes_read,
gsize *bytes_written,
GError **error)
{
gchar *dest;
gchar *outp;
const gchar *p;
gsize inbytes_remaining;
gsize outbytes_remaining;
gsize err;
gsize outbuf_size;
gboolean have_error = FALSE;
g_return_val_if_fail (str != NULL, NULL);
g_return_val_if_fail (converter != (GIConv) -1, NULL);
if (len < 0)
len = strlen (str);
p = str;
inbytes_remaining = len;
outbuf_size = len + 1;
outbytes_remaining = outbuf_size - 1;
outp = dest = g_malloc (outbuf_size);
again:
err = g_iconv (converter, (char **)&p, &inbytes_remaining, &outp, &outbytes_remaining);
if (err == (size_t) -1)
{
switch (errno)
{
case EINVAL:
break;
case E2BIG:
{
size_t used = outp - dest;
outbuf_size *= 2;
dest = g_realloc (dest, outbuf_size);
outp = dest + used;
outbytes_remaining = outbuf_size - used - 1;
goto again;
}
case EILSEQ:
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
_("Invalid byte sequence in conversion input"));
have_error = TRUE;
break;
default:
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_FAILED,
_("Error during conversion: %s"),
g_strerror (errno));
have_error = TRUE;
break;
}
}
*outp = '\0';
if (bytes_read)
*bytes_read = p - str;
else
{
if ((p - str) != len)
{
if (!have_error)
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_PARTIAL_INPUT,
_("Partial character sequence at end of input"));
have_error = TRUE;
}
}
}
if (bytes_written)
*bytes_written = outp - dest;
if (have_error)
{
g_free (dest);
return NULL;
}
else
return dest;
}
gchar*
g_convert_with_fallback (const gchar *str,
gssize len,
const gchar *to_codeset,
const gchar *from_codeset,
gchar *fallback,
gsize *bytes_read,
gsize *bytes_written,
GError **error)
{
gchar *utf8;
gchar *dest;
gchar *outp;
const gchar *insert_str = NULL;
const gchar *p;
gsize inbytes_remaining;
const gchar *save_p = NULL;
gsize save_inbytes = 0;
gsize outbytes_remaining;
gsize err;
GIConv cd;
gsize outbuf_size;
gboolean have_error = FALSE;
gboolean done = FALSE;
GError *local_error = NULL;
g_return_val_if_fail (str != NULL, NULL);
g_return_val_if_fail (to_codeset != NULL, NULL);
g_return_val_if_fail (from_codeset != NULL, NULL);
if (len < 0)
len = strlen (str);
dest = g_convert (str, len, to_codeset, from_codeset,
bytes_read, bytes_written, &local_error);
if (!local_error)
return dest;
if (!g_error_matches (local_error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE))
{
g_propagate_error (error, local_error);
return NULL;
}
else
g_error_free (local_error);
local_error = NULL;
cd = open_converter (to_codeset, "UTF-8", error);
if (cd == (GIConv) -1)
{
if (bytes_read)
*bytes_read = 0;
if (bytes_written)
*bytes_written = 0;
return NULL;
}
utf8 = g_convert (str, len, "UTF-8", from_codeset,
bytes_read, &inbytes_remaining, error);
if (!utf8)
{
close_converter (cd);
if (bytes_written)
*bytes_written = 0;
return NULL;
}
p = utf8;
outbuf_size = len + 1;
outbytes_remaining = outbuf_size - 1;
outp = dest = g_malloc (outbuf_size);
while (!done && !have_error)
{
size_t inbytes_tmp = inbytes_remaining;
err = g_iconv (cd, (char **)&p, &inbytes_tmp, &outp, &outbytes_remaining);
inbytes_remaining = inbytes_tmp;
if (err == (size_t) -1)
{
switch (errno)
{
case EINVAL:
g_assert_not_reached();
break;
case E2BIG:
{
size_t used = outp - dest;
outbuf_size *= 2;
dest = g_realloc (dest, outbuf_size);
outp = dest + used;
outbytes_remaining = outbuf_size - used - 1;
break;
}
case EILSEQ:
if (save_p)
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
_("Cannot convert fallback '%s' to codeset '%s'"),
insert_str, to_codeset);
have_error = TRUE;
break;
}
else
{
if (!fallback)
{
gunichar ch = g_utf8_get_char (p);
insert_str = g_strdup_printf (ch < 0x10000 ? "\\u%04x" : "\\U%08x",
ch);
}
else
insert_str = fallback;
save_p = g_utf8_next_char (p);
save_inbytes = inbytes_remaining - (save_p - p);
p = insert_str;
inbytes_remaining = strlen (p);
}
break;
default:
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_FAILED,
_("Error during conversion: %s"),
g_strerror (errno));
have_error = TRUE;
break;
}
}
else
{
if (save_p)
{
if (!fallback)
g_free ((gchar *)insert_str);
p = save_p;
inbytes_remaining = save_inbytes;
save_p = NULL;
}
else
done = TRUE;
}
}
*outp = '\0';
close_converter (cd);
if (bytes_written)
*bytes_written = outp - dest;
g_free (utf8);
if (have_error)
{
if (save_p && !fallback)
g_free ((gchar *)insert_str);
g_free (dest);
return NULL;
}
else
return dest;
}
static gchar *
strdup_len (const gchar *string,
gssize len,
gsize *bytes_written,
gsize *bytes_read,
GError **error)
{
gsize real_len;
if (!g_utf8_validate (string, len, NULL))
{
if (bytes_read)
*bytes_read = 0;
if (bytes_written)
*bytes_written = 0;
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
_("Invalid byte sequence in conversion input"));
return NULL;
}
if (len < 0)
real_len = strlen (string);
else
{
real_len = 0;
while (real_len < len && string[real_len])
real_len++;
}
if (bytes_read)
*bytes_read = real_len;
if (bytes_written)
*bytes_written = real_len;
return g_strndup (string, real_len);
}
gchar *
g_locale_to_utf8 (const gchar *opsysstring,
gssize len,
gsize *bytes_read,
gsize *bytes_written,
GError **error)
{
const char *charset;
if (g_get_charset (&charset))
return strdup_len (opsysstring, len, bytes_read, bytes_written, error);
else
return g_convert (opsysstring, len,
"UTF-8", charset, bytes_read, bytes_written, error);
}
gchar *
g_locale_from_utf8 (const gchar *utf8string,
gssize len,
gsize *bytes_read,
gsize *bytes_written,
GError **error)
{
const gchar *charset;
if (g_get_charset (&charset))
return strdup_len (utf8string, len, bytes_read, bytes_written, error);
else
return g_convert (utf8string, len,
charset, "UTF-8", bytes_read, bytes_written, error);
}
#ifndef G_PLATFORM_WIN32
typedef struct _GFilenameCharsetCache GFilenameCharsetCache;
struct _GFilenameCharsetCache {
gboolean is_utf8;
gchar *charset;
gchar *filename_charset;
};
static void
filename_charset_cache_free (gpointer data)
{
GFilenameCharsetCache *cache = data;
g_free (cache->charset);
g_free (cache->filename_charset);
g_free (cache);
}
static gboolean
get_filename_charset (const gchar **filename_charset)
{
static GStaticPrivate cache_private = G_STATIC_PRIVATE_INIT;
GFilenameCharsetCache *cache = g_static_private_get (&cache_private);
const gchar *charset;
if (!cache)
{
cache = g_new0 (GFilenameCharsetCache, 1);
g_static_private_set (&cache_private, cache, filename_charset_cache_free);
}
g_get_charset (&charset);
if (!(cache->charset && strcmp (cache->charset, charset) == 0))
{
const gchar *new_charset;
gchar *p, *q;
g_free (cache->charset);
g_free (cache->filename_charset);
cache->charset = g_strdup (charset);
p = getenv ("G_FILENAME_ENCODING");
if (p != NULL)
{
q = strchr (p, ',');
if (!q)
q = p + strlen (p);
if (strncmp ("@locale", p, q - p) == 0)
{
cache->is_utf8 = g_get_charset (&new_charset);
cache->filename_charset = g_strdup (new_charset);
}
else
{
cache->filename_charset = g_strndup (p, q - p);
cache->is_utf8 = (strcmp (cache->filename_charset, "UTF-8") == 0);
}
}
else if (getenv ("G_BROKEN_FILENAMES") != NULL)
{
cache->is_utf8 = g_get_charset (&new_charset);
cache->filename_charset = g_strdup (new_charset);
}
else
{
cache->filename_charset = g_strdup ("UTF-8");
cache->is_utf8 = TRUE;
}
}
if (filename_charset)
*filename_charset = cache->filename_charset;
return cache->is_utf8;
}
#else
static gboolean
get_filename_charset (const gchar **filename_charset)
{
g_get_charset (filename_charset);
return FALSE;
}
#endif
void
_g_convert_thread_init (void)
{
const gchar *dummy;
(void) get_filename_charset (&dummy);
}
gchar*
g_filename_to_utf8 (const gchar *opsysstring,
gssize len,
gsize *bytes_read,
gsize *bytes_written,
GError **error)
{
const gchar *charset;
if (get_filename_charset (&charset))
return strdup_len (opsysstring, len, bytes_read, bytes_written, error);
else
return g_convert (opsysstring, len,
"UTF-8", charset, bytes_read, bytes_written, error);
}
gchar*
g_filename_from_utf8 (const gchar *utf8string,
gssize len,
gsize *bytes_read,
gsize *bytes_written,
GError **error)
{
const gchar *charset;
if (get_filename_charset (&charset))
return strdup_len (utf8string, len, bytes_read, bytes_written, error);
else
return g_convert (utf8string, len,
charset, "UTF-8", bytes_read, bytes_written, error);
}
static gboolean
has_case_prefix (const gchar *haystack, const gchar *needle)
{
const gchar *h, *n;
h = haystack;
n = needle;
while (*n && *h &&
g_ascii_tolower (*n) == g_ascii_tolower (*h))
{
n++;
h++;
}
return *n == '\0';
}
typedef enum {
UNSAFE_ALL = 0x1,
UNSAFE_ALLOW_PLUS = 0x2,
UNSAFE_PATH = 0x8,
UNSAFE_HOST = 0x10,
UNSAFE_SLASHES = 0x20
} UnsafeCharacterSet;
static const guchar acceptable[96] = {
0x00,0x3F,0x20,0x20,0x28,0x00,0x2C,0x3F,0x3F,0x3F,0x3F,0x2A,0x28,0x3F,0x3F,0x1C,
0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x38,0x20,0x20,0x2C,0x20,0x20,
0x38,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,
0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x20,0x20,0x20,0x20,0x3F,
0x20,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,
0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x3F,0x20,0x20,0x20,0x3F,0x20
};
static const gchar hex[16] = "0123456789ABCDEF";
static gchar *
g_escape_uri_string (const gchar *string,
UnsafeCharacterSet mask)
{
#define ACCEPTABLE(a) ((a)>=32 && (a)<128 && (acceptable[(a)-32] & use_mask))
const gchar *p;
gchar *q;
gchar *result;
int c;
gint unacceptable;
UnsafeCharacterSet use_mask;
g_return_val_if_fail (mask == UNSAFE_ALL
|| mask == UNSAFE_ALLOW_PLUS
|| mask == UNSAFE_PATH
|| mask == UNSAFE_HOST
|| mask == UNSAFE_SLASHES, NULL);
unacceptable = 0;
use_mask = mask;
for (p = string; *p != '\0'; p++)
{
c = (guchar) *p;
if (!ACCEPTABLE (c))
unacceptable++;
}
result = g_malloc (p - string + unacceptable * 2 + 1);
use_mask = mask;
for (q = result, p = string; *p != '\0'; p++)
{
c = (guchar) *p;
if (!ACCEPTABLE (c))
{
*q++ = '%';
*q++ = hex[c >> 4];
*q++ = hex[c & 15];
}
else
*q++ = *p;
}
*q = '\0';
return result;
}
static gchar *
g_escape_file_uri (const gchar *hostname,
const gchar *pathname)
{
char *escaped_hostname = NULL;
char *escaped_path;
char *res;
#ifdef G_OS_WIN32
char *p, *backslash;
pathname = g_strdup (pathname);
p = (char *) pathname;
while ((backslash = strchr (p, '\\')) != NULL)
{
*backslash = '/';
p = backslash + 1;
}
#endif
if (hostname && *hostname != '\0')
{
escaped_hostname = g_escape_uri_string (hostname, UNSAFE_HOST);
}
escaped_path = g_escape_uri_string (pathname, UNSAFE_PATH);
res = g_strconcat ("file://",
(escaped_hostname) ? escaped_hostname : "",
(*escaped_path != '/') ? "/" : "",
escaped_path,
NULL);
#ifdef G_OS_WIN32
g_free ((char *) pathname);
#endif
g_free (escaped_hostname);
g_free (escaped_path);
return res;
}
static int
unescape_character (const char *scanner)
{
int first_digit;
int second_digit;
first_digit = g_ascii_xdigit_value (scanner[0]);
if (first_digit < 0)
return -1;
second_digit = g_ascii_xdigit_value (scanner[1]);
if (second_digit < 0)
return -1;
return (first_digit << 4) | second_digit;
}
static gchar *
g_unescape_uri_string (const char *escaped,
int len,
const char *illegal_escaped_characters,
gboolean ascii_must_not_be_escaped)
{
const gchar *in, *in_end;
gchar *out, *result;
int c;
if (escaped == NULL)
return NULL;
if (len < 0)
len = strlen (escaped);
result = g_malloc (len + 1);
out = result;
for (in = escaped, in_end = escaped + len; in < in_end; in++)
{
c = *in;
if (c == '%')
{
if (in + 3 > in_end)
break;
c = unescape_character (in + 1);
if (c <= 0)
break;
if (ascii_must_not_be_escaped && c <= 0x7F)
break;
if (strchr (illegal_escaped_characters, c) != NULL)
break;
in += 2;
}
*out++ = c;
}
g_assert (out - result <= len);
*out = '\0';
if (in != in_end)
{
g_free (result);
return NULL;
}
return result;
}
static gboolean
is_asciialphanum (gunichar c)
{
return c <= 0x7F && g_ascii_isalnum (c);
}
static gboolean
is_asciialpha (gunichar c)
{
return c <= 0x7F && g_ascii_isalpha (c);
}
static gboolean
hostname_validate (const char *hostname)
{
const char *p;
gunichar c, first_char, last_char;
p = hostname;
if (*p == '\0')
return TRUE;
do
{
c = g_utf8_get_char (p);
p = g_utf8_next_char (p);
if (!is_asciialphanum (c))
return FALSE;
first_char = c;
do
{
last_char = c;
c = g_utf8_get_char (p);
p = g_utf8_next_char (p);
}
while (is_asciialphanum (c) || c == '-');
if (last_char == '-')
return FALSE;
if (c == '\0' || (c == '.' && *p == '\0'))
return is_asciialpha (first_char);
}
while (c == '.');
return FALSE;
}
gchar *
g_filename_from_uri (const gchar *uri,
gchar **hostname,
GError **error)
{
const char *path_part;
const char *host_part;
char *unescaped_hostname;
char *result;
char *filename;
int offs;
#ifdef G_OS_WIN32
char *p, *slash;
#endif
if (hostname)
*hostname = NULL;
if (!has_case_prefix (uri, "file:/"))
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_BAD_URI,
_("The URI '%s' is not an absolute URI using the file scheme"),
uri);
return NULL;
}
path_part = uri + strlen ("file:");
if (strchr (path_part, '#') != NULL)
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_BAD_URI,
_("The local file URI '%s' may not include a '#'"),
uri);
return NULL;
}
if (has_case_prefix (path_part, "///"))
path_part += 2;
else if (has_case_prefix (path_part, "//"))
{
path_part += 2;
host_part = path_part;
path_part = strchr (path_part, '/');
if (path_part == NULL)
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_BAD_URI,
_("The URI '%s' is invalid"),
uri);
return NULL;
}
unescaped_hostname = g_unescape_uri_string (host_part, path_part - host_part, "", TRUE);
if (unescaped_hostname == NULL ||
!hostname_validate (unescaped_hostname))
{
g_free (unescaped_hostname);
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_BAD_URI,
_("The hostname of the URI '%s' is invalid"),
uri);
return NULL;
}
if (hostname)
*hostname = unescaped_hostname;
else
g_free (unescaped_hostname);
}
filename = g_unescape_uri_string (path_part, -1, "/", FALSE);
if (filename == NULL)
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_BAD_URI,
_("The URI '%s' contains invalidly escaped characters"),
uri);
return NULL;
}
offs = 0;
#ifdef G_OS_WIN32
if (hostname && *hostname != NULL &&
g_ascii_strcasecmp (*hostname, "localhost") == 0)
{
g_free (*hostname);
*hostname = NULL;
}
p = filename;
while ((slash = strchr (p, '/')) != NULL)
{
*slash = '\\';
p = slash + 1;
}
if (g_ascii_isalpha (filename[1]))
{
if (filename[2] == ':')
offs = 1;
else if (filename[2] == '|')
{
filename[2] = ':';
offs = 1;
}
}
#endif
result = g_strdup (filename + offs);
g_free (filename);
return result;
}
gchar *
g_filename_to_uri (const gchar *filename,
const gchar *hostname,
GError **error)
{
char *escaped_uri;
g_return_val_if_fail (filename != NULL, NULL);
if (!g_path_is_absolute (filename))
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_NOT_ABSOLUTE_PATH,
_("The pathname '%s' is not an absolute path"),
filename);
return NULL;
}
if (hostname &&
!(g_utf8_validate (hostname, -1, NULL)
&& hostname_validate (hostname)))
{
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
_("Invalid hostname"));
return NULL;
}
#ifdef G_OS_WIN32
if (hostname && g_ascii_strcasecmp (hostname, "localhost") == 0)
hostname = NULL;
#endif
escaped_uri = g_escape_file_uri (hostname, filename);
return escaped_uri;
}