diff -rdup old/direntry.h new/direntry.h
--- old/direntry.h 2010-05-08 22:10:26.000000000 +0800
+++ new/direntry.h 2010-05-10 10:21:35.000000000 +0800
@@ -133,6 +133,7 @@ int unix2winfn(const u_char *un, int unl
int winChkName(const u_char *un, int unlen, struct winentry *wep,
int chksum);
int win2unixfn(struct winentry *wep, struct dirent *dp, int chksum);
+int dosdir_wchar_to_utf8(struct dirent *dp, int chksum);
u_int8_t winChksum(u_int8_t *name);
int winSlotCnt(const u_char *un, int unlen);
#endif /* _KERNEL */
diff -rdup old/msdosfs_conv.c new/msdosfs_conv.c
--- old/msdosfs_conv.c 2010-05-08 22:10:26.000000000 +0800
+++ new/msdosfs_conv.c 2010-05-10 10:21:35.000000000 +0800
@@ -91,6 +91,71 @@ u_long lastday;
u_short lastddate;
u_short lastdtime;
+static int
+wput_utf8(char *s, size_t n, u_int16_t wc)
+{
+ if (wc & 0xF800) {
+ if (n < 3)
+ return 0;
+ s[0] = 0xE0 | (wc >> 12);
+ s[1] = 0x80 | ((wc >> 6) & 0x3F);
+ s[2] = 0x80 | (wc & 0x3F);
+ return 3;
+ } else if (wc & 0x0780) {
+ if (n < 2)
+ return 0;
+ s[0] = 0xC0 | (wc >> 6);
+ s[1] = 0x80 | ((wc) & 0x3F); + return 2;
+ } else {
+ if (n < 1)
+ return 0;
+ s[0] = wc;
+ return 1;
+ }
+ return 0;
+}
+
+static u_int16_t
+wget_utf8(const char **str, size_t *sz)
+{
+ int c;
+ u_int16_t rune = 0;
+ const char *s = *str;
+ static const int _utf_count[16] = {
+ 1, 1, 1, 1, 1, 1, 1, 1,
+ 0, 0, 0, 0, 2, 2, 3, 0,
+ };
+
+ c = _utf_count[(s[0] & 0xF0) >> 4];
+ if (c == 0 || c > *sz) {
+decoding_error:
+ c = 1;
+ }
+
+ switch (c) {
+ case 1:
+ rune = s[0] & 0xFF;
+ break;
+ case 2:
+ if ((s[1] & 0xC0) != 0x80)
+ goto decoding_error;
+ rune = ((s[0] & 0x1F) << 6) | (s[1] & 0x3f);
+ break;
+ case 3:
+ if ((s[1] & 0xC0) != 0x80 || (s[2] & 0xC0) != 0x80)
+ goto decoding_error;
+ rune = ((s[0] & 0x0F) << 12) | ((s[1] & 0x3F) << 6) |
+ (s[2] & 0x3F);
+ break;
+ }
+
+ *str += c;
+ *sz -= c;
+ return rune;
+}
+
+
/*
* Convert the unix version of time to dos's idea of time to be used in
* file timestamps. The passed in unix time is assumed to be in GMT.
@@ -588,13 +653,30 @@ unix2winfn(un, unlen, wep, cnt, chksum)
u_int8_t *wcp;
int i;
+ size_t count1;
+ const char * name;
+ u_char buf[WIN_MAXLEN * 2];
+
/*
* Drop trailing blanks and dots
*/
for (cp = un + unlen; *--cp == ' ' || *cp == '.'; unlen--);
- un += (cnt - 1) * WIN_CHARS;
- unlen -= (cnt - 1) * WIN_CHARS;
+ name = un;
+ count1 = 0;
+ while (unlen > 0) {
+ u_int16_t wc;
+ size_t len1 = unlen;
+ wc = wget_utf8(&name, &len1);
+ buf[count1++] = (wc & 0xFF);
+ buf[count1++] = (wc >> 8);
+ unlen = len1;
+ }
+ unlen = count1;
+ un = buf;
+
+ un += (cnt - 1) * WIN_CHARS * 2;
+ unlen -= (cnt - 1) * WIN_CHARS * 2;
/*
* Initialize winentry to some useful default
@@ -610,22 +692,25 @@ unix2winfn(un, unlen, wep, cnt, chksum)
* Now convert the filename parts
*/
for (wcp = wep->wePart1, i = sizeof(wep->wePart1)/2; --i >= 0;) {
- if (--unlen < 0)
+ unlen -= 2;
+ if (unlen < 0)
goto done;
*wcp++ = *un++;
- *wcp++ = 0;
+ *wcp++ = *un++;
}
for (wcp = wep->wePart2, i = sizeof(wep->wePart2)/2; --i >= 0;) {
- if (--unlen < 0)
+ unlen -= 2;
+ if (unlen < 0)
goto done;
*wcp++ = *un++;
- *wcp++ = 0;
+ *wcp++ = *un++;
}
for (wcp = wep->wePart3, i = sizeof(wep->wePart3)/2; --i >= 0;) {
- if (--unlen < 0)
+ unlen -= 2;
+ if (unlen < 0)
goto done;
*wcp++ = *un++;
- *wcp++ = 0;
+ *wcp++ = *un++;
}
if (!unlen)
wep->weCnt |= WIN_LAST;
@@ -649,9 +734,14 @@ winChkName(un, unlen, wep, chksum)
struct winentry *wep;
int chksum;
{
- u_int8_t *cp;
+ u_char we_name[2 + sizeof(*wep)];
+ u_int8_t *cp, *we_p;
int i;
+ size_t count1;
+ const char * name;
+ u_char buf[WIN_MAXLEN * 2];
+
/*
* First compare checksums
*/
@@ -662,10 +751,26 @@ winChkName(un, unlen, wep, chksum)
if (chksum == -1)
return -1;
+ count1 = 0;
+ name = (const char *)un;
+ while (unlen > 0) {
+ u_int16_t wc;
+ size_t len1 = unlen;
+ /* const char * t_name = name; */
+ wc = wget_utf8(&name, &len1);
+ buf[count1++] = (wc & 0xFF);
+ buf[count1++] = (wc >> 8);
+ /* assert(t_name != name); */
+ unlen = len1;
+ }
+ unlen = count1;
+ un = buf;
+
+
/*
* Offset of this entry
*/
- i = ((wep->weCnt&WIN_CNT) - 1) * WIN_CHARS;
+ i = ((wep->weCnt&WIN_CNT) - 1) * WIN_CHARS * 2;
un += i;
if ((unlen -= i) < 0)
return -1;
@@ -682,12 +788,13 @@ winChkName(un, unlen, wep, chksum)
return -1;
}
- if ((wep->weCnt&WIN_LAST) && unlen > WIN_CHARS)
+ if ((wep->weCnt&WIN_LAST) && unlen > WIN_CHARS * 2)
return -1;
/*
* Compare the name parts
*/
+#if 0
for (cp = wep->wePart1, i = sizeof(wep->wePart1)/2; --i >= 0;) {
if (--unlen < 0) {
if (!*cp++ && !*cp)
@@ -715,6 +821,25 @@ winChkName(un, unlen, wep, chksum)
if (u2l[*cp++] != u2l[*un++] || *cp++)
return -1;
}
+#endif
+ cp = we_p = we_name;
+ memcpy(we_p, wep->wePart1, sizeof(wep->wePart1));
+ we_p += sizeof(wep->wePart1);
+ memcpy(we_p, wep->wePart2, sizeof(wep->wePart2));
+ we_p += sizeof(wep->wePart2);
+ memcpy(we_p, wep->wePart3, sizeof(wep->wePart3));
+ we_p += sizeof(wep->wePart3);
+ we_p[0] = we_p[1] = 0;
+
+ while (cp < we_p) {
+ if (--unlen < 0) {
+ if (cp[0] == 0 && cp[1] == 0)
+ return chksum;
+ return -1;
+ }
+ if (*cp++ != *un++)
+ return -1;
+ }
return chksum;
}
@@ -732,7 +858,7 @@ win2unixfn(wep, dp, chksum)
u_int8_t *np, *ep = dp->d_name + WIN_MAXLEN;
int i;
- if ((wep->weCnt&WIN_CNT) > howmany(WIN_MAXLEN, WIN_CHARS)
+ if ((wep->weCnt&WIN_CNT) > howmany(WIN_MAXLEN, WIN_CHARS * 2)
|| !(wep->weCnt&WIN_CNT))
return -1;
@@ -744,7 +870,7 @@ win2unixfn(wep, dp, chksum)
/*
* This works even though d_namlen is one byte!
*/
- dp->d_namlen = (wep->weCnt&WIN_CNT) * WIN_CHARS;
+ dp->d_namlen = (wep->weCnt&WIN_CNT) * WIN_CHARS * 2;
} else if (chksum != wep->weChksum)
chksum = -1;
if (chksum == -1)
@@ -753,9 +879,20 @@ win2unixfn(wep, dp, chksum)
/*
* Offset of this entry
*/
- i = ((wep->weCnt&WIN_CNT) - 1) * WIN_CHARS;
+ i = ((wep->weCnt&WIN_CNT) - 1) * WIN_CHARS * 2;
np = (u_int8_t *)dp->d_name + i;
+ memcpy(np, wep->wePart1, sizeof(wep->wePart1));
+ np += sizeof(wep->wePart1);
+ memcpy(np, wep->wePart2, sizeof(wep->wePart2));
+ np += sizeof(wep->wePart2);
+ memcpy(np, wep->wePart3, sizeof(wep->wePart3));
+ np += sizeof(wep->wePart3);
+
+ /* avoid warnning */
+ ep = cp = np;
+
+#if 0
/*
* Convert the name parts
*/
@@ -822,6 +959,7 @@ win2unixfn(wep, dp, chksum)
if (*cp++)
return -1;
}
+#endif
return chksum;
}
@@ -848,10 +986,65 @@ winSlotCnt(un, unlen)
const u_char *un;
int unlen;
{
- for (un += unlen; unlen > 0; unlen--)
- if (*--un != ' ' && *un != '.')
- break;
- if (unlen > WIN_MAXLEN)
+ u_int16_t wc;
+ size_t count = 0;
+
+ size_t count1 = 0;
+ const char * name = un;
+ while (unlen > 0) {
+ int len = unlen;
+ wc = wget_utf8(&name, &len);
+ count += 2;
+ if (wc != ' ' && wc != '.')
+ count1 = count;
+ unlen = len;
+ }
+ if (count1 > WIN_MAXLEN * 2)
return 0;
- return howmany(unlen, WIN_CHARS);
+ return howmany(count1, WIN_CHARS * 2);
}
+
+int
+dosdir_wchar_to_utf8(dp, chksum)
+struct dirent *dp;
+int chksum;
+{
+ u_int16_t wc;
+ int n, count;
+
+ char *cp, *ep;
+ u_int8_t *np;
+ u_int8_t name[sizeof(dp->d_name)];
+
+ if ((dp->d_namlen & 0x01) ||
+ dp->d_namlen > sizeof(name))
+ return -1;
+
+ memcpy(name, dp->d_name, dp->d_namlen);
+ count = (dp->d_namlen / 2);
+
+ np = name;
+ cp = dp->d_name;
+ ep = dp->d_name + sizeof(dp->d_name);
+ while (count-- > 0 && cp < ep) {
+ wc = *np++;
+ wc |= (*np++ << 8);
+ switch(wc) {
+ case '/':
+ dp->d_namlen = (cp - dp->d_name);
+ *cp = 0;
+ return -1;
+ case 0:
+ dp->d_namlen = (cp - dp->d_name);
+ *cp = 0;
+ return chksum;
+ }
+ n = wput_utf8(cp, ep - cp, wc);
+ if (n == 0)
+ break;
+ cp += n;
+ }
+ dp->d_namlen = (cp < ep)? (cp - dp->d_name): dp->d_namlen;
+ return (cp < ep)? chksum: -1;
+}
+
diff -rdup old/msdosfs_vnops.c new/msdosfs_vnops.c
--- old/msdosfs_vnops.c 2010-05-08 22:10:26.000000000 +0800
+++ new/msdosfs_vnops.c 2010-05-10 10:21:44.000000000 +0800
@@ -1637,6 +1637,7 @@ msdosfs_readdir(v)
offset / sizeof(struct direntry);
dirbuf->d_type = DT_REG;
}
+ chksum = dosdir_wchar_to_utf8(dirbuf, chksum);
if (chksum != winChksum(dentp->deName))
dirbuf->d_namlen = dos2unixfn(dentp->deName,
(u_char *)dirbuf->d_name, |
