X-Git-Url: https://vcs.fsf.org/?a=blobdiff_plain;f=src%2Fsrc%2Fsieve.c;h=e243f653c3433792358e00f1b88a3301d67840e3;hb=7b3a77e5b3464fa2533bd82f644c9145c1f11f71;hp=56b00027222ec9ec5e8f3c7238edf54cff3ecd01;hpb=a43a27c5e355a06131c36e68a63a0609b1da8307;p=exim.git

diff --git a/src/src/sieve.c b/src/src/sieve.c
index 56b000272..e243f653c 100644
--- a/src/src/sieve.c
+++ b/src/src/sieve.c
@@ -1,4 +1,4 @@
-/* $Cambridge: exim/src/src/sieve.c,v 1.27 2007/04/12 09:00:52 ph10 Exp $ */
+/* $Cambridge: exim/src/src/sieve.c,v 1.28 2007/04/19 13:19:06 ph10 Exp $ */
 
 /*************************************************
 *     Exim - an Internet mail transport agent    *
@@ -1197,12 +1197,13 @@ return 1;
 
 #ifdef ENCODED_CHARACTER
 /*************************************************
-*       Decode encoded-character string          *
+*      Decode hex-encoded-character string       *
 *************************************************/
 
 /*
 Encoding definition:
-   hex-pair-seq         = hex-pair *(WSP hex-pair)
+   blank                = SP / TAB / CRLF
+   hex-pair-seq         = *blank hex-pair *(1*blank hex-pair) *blank
    hex-pair             = 1*2HEXDIG
 
 Arguments:
@@ -1219,37 +1220,42 @@ static int hex_decode(uschar *src, uschar *end, uschar *dst)
 {
 int decoded=0;
 
-while (src<end)
+while (*src==' ' || *src=='\t' || *src=='\n') ++src;
+do
   {
-  int l,h;
+  int x,d,n;
 
-  if (*src==' ' || *src=='\t') ++src;
-  else if ((src+1)<end && isxdigit(h=tolower(*src)) && isxdigit(l=tolower(*(src+1))))
-    {
-    h=(h>='0' && h<='9') ? h-'0' : 10+(h-'a');
-    l=(l>='0' && l<='9') ? l-'0' : 10+(l-'a');
-    if (dst) *dst++=(h<<4)|l;
-    ++decoded;
-    src+=2;
-    }
-  else return -1;
+  for (x=0,d=0; d<2 && src<end && isxdigit(n=tolower(*src)); x=(x<<4)|(n>='0' && n<='9' ? n-'0' : 10+(n-'a')),++d,++src);
+  if (d==0) return -1;
+  if (dst) *dst++=x;
+  ++decoded;
+  if (src==end) return decoded;
+  if (*src==' ' || *src=='\t' || *src=='\n')
+    while (*src==' ' || *src=='\t' || *src=='\n') ++src;
+  else
+    return -1;
   }
-  return decoded;
+while (src<end);
+return decoded;
 }
 
 
 /*************************************************
-*       Decode encoded-character string          *
+*    Decode unicode-encoded-character string     *
 *************************************************/
 
 /*
 Encoding definition:
-   unicode-hex-seq      = unicode-hex *(WSP unicode-hex)
-   unicode-hex          = 1*6HEXDIG
+   blank                = SP / TAB / CRLF
+   unicode-hex-seq      = *blank unicode-hex *(blank unicode-hex) *blank
+   unicode-hex          = 1*HEXDIG
 
    It is an error for a script to use a hexadecimal value that isn't in
    either the range 0 to D7FF or the range E000 to 10FFFF.
 
+   At this time, strings are already scanned, thus the CRLF is converted
+   to the internally used \n (should RFC_EOL have been used).
+
 Arguments:
   src         points to a unicode-hex-seq
   end         points to its end
@@ -1265,69 +1271,61 @@ static int unicode_decode(uschar *src, uschar *end, uschar *dst)
 {
 int decoded=0;
 
-while (src<end)
+while (*src==' ' || *src=='\t' || *src=='\n') ++src;
+do
   {
-  int c,n5,n4,n3,n2,n1,n0;
+  uschar *hex_seq;
+  int c,d,n;
 
-  if (*src==' ' || *src=='\t') ++src;
-  else if (
-           (src+5)<end
-           && isxdigit(n5=tolower(*src))
-           && isxdigit(n4=tolower(*(src+1)))
-           && isxdigit(n3=tolower(*(src+2)))
-           && isxdigit(n2=tolower(*(src+3)))
-           && isxdigit(n1=tolower(*(src+4)))
-           && isxdigit(n0=tolower(*(src+5)))
-          )
+  unicode_hex:
+  for (hex_seq=src; src<end && *src=='0'; ++src);
+  for (c=0,d=0; d<7 && src<end && isxdigit(n=tolower(*src)); c=(c<<4)|(n>='0' && n<='9' ? n-'0' : 10+(n-'a')),++d,++src);
+  if (src==hex_seq) return -1;
+  if (d==7 || (!((c>=0 && c<=0xd7ff) || (c>=0xe000 && c<=0x10ffff)))) return -2;
+  if (c<128)
     {
-    n5=(n5>='0' && n5<='9') ? n5-'0' : 10+(n5-'a');
-    n4=(n4>='0' && n4<='9') ? n4-'0' : 10+(n4-'a');
-    n3=(n3>='0' && n3<='9') ? n3-'0' : 10+(n3-'a');
-    n2=(n2>='0' && n2<='9') ? n2-'0' : 10+(n2-'a');
-    n1=(n1>='0' && n1<='9') ? n1-'0' : 10+(n1-'a');
-    n0=(n0>='0' && n0<='9') ? n0-'0' : 10+(n0-'a');
-    c=(n5<<24)|(n4<<16)|(n3<<12)|(n2<<8)|(n1<<4)|n0;
-    if (!((c>=0 && c<=0xd7ff) || (c>=0xe000 && c<=0x10ffff))) return -2;
-    if (c<128)
-      {
-      if (dst) *dst++=c;
-      ++decoded;
-      }
-    else if (c>=0x80 && c<=0x7ff)
-      {
-        if (dst)
-          {
-          *dst++=192+(c>>6);
-          *dst++=128+(c&0x3f);
-          }
-        decoded+=2;
-      }
-    else if (c>=0x800 && c<=0xffff)
-      {
-        if (dst)
-          {
-          *dst++=224+(c>>12);
-          *dst++=128+((c>>6)&0x3f);
-          *dst++=128+(c&0x3f);
-          }
-        decoded+=3;
-      }
-    else if (c>=0x10000 && c<=0x1fffff)
-      {
-        if (dst)
-          {
-          *dst++=240+(c>>18);
-          *dst++=128+((c>>10)&0x3f);
-          *dst++=128+((c>>6)&0x3f);
-          *dst++=128+(c&0x3f);
-          }
-        decoded+=4;
-      }
-    src+=6;
+    if (dst) *dst++=c;
+    ++decoded;
+    }
+  else if (c>=0x80 && c<=0x7ff)
+    {
+      if (dst)
+        {
+        *dst++=192+(c>>6);
+        *dst++=128+(c&0x3f);
+        }
+      decoded+=2;
+    }
+  else if (c>=0x800 && c<=0xffff)
+    {
+      if (dst)
+        {
+        *dst++=224+(c>>12);
+        *dst++=128+((c>>6)&0x3f);
+        *dst++=128+(c&0x3f);
+        }
+      decoded+=3;
+    }
+  else if (c>=0x10000 && c<=0x1fffff)
+    {
+      if (dst)
+        {
+        *dst++=240+(c>>18);
+        *dst++=128+((c>>10)&0x3f);
+        *dst++=128+((c>>6)&0x3f);
+        *dst++=128+(c&0x3f);
+        }
+      decoded+=4;
+    }
+  if (*src==' ' || *src=='\t' || *src=='\n')
+    {
+    while (*src==' ' || *src=='\t' || *src=='\n') ++src;
+    if (src==end) return decoded;
+    goto unicode_hex;
     }
-  else return -1;
   }
-  return decoded;
+while (src<end);
+return decoded;
 }
 
 
@@ -1360,7 +1358,7 @@ while (src<end)
   uschar *brace;
 
   if (
-      Ustrncmp(src,CUS "${hex:",6)==0
+      strncmpic(src,US "${hex:",6)==0
       && (brace=Ustrchr(src+6,'}'))!=(uschar*)0
       && (hex_decode(src+6,brace,(uschar*)0))>=0
      )
@@ -1369,7 +1367,7 @@ while (src<end)
     src=brace+1;
     }
   else if (
-           Ustrncmp(src,CUS "${unicode:",10)==0
+           strncmpic(src,US "${unicode:",10)==0
            && (brace=Ustrchr(src+10,'}'))!=(uschar*)0
           )
     {
@@ -1584,7 +1582,7 @@ static int parse_identifier(struct Sieve *filter, const uschar *id)
 {
   size_t idlen=Ustrlen(id);
 
-  if (Ustrncmp(filter->pc,id,idlen)==0)
+  if (strncmpic(US filter->pc,US id,idlen)==0)
   {
     uschar next=filter->pc[idlen];