-/* $Cambridge: exim/src/src/sieve.c,v 1.26 2007/03/21 15:15:12 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 *
}
+#ifdef ENCODED_CHARACTER
+/*************************************************
+* Decode hex-encoded-character string *
+*************************************************/
+
+/*
+Encoding definition:
+ blank = SP / TAB / CRLF
+ hex-pair-seq = *blank hex-pair *(1*blank hex-pair) *blank
+ hex-pair = 1*2HEXDIG
+
+Arguments:
+ src points to a hex-pair-seq
+ end points to its end
+ dst points to the destination of the decoded octets,
+ optionally to (uschar*)0 for checking only
+
+Returns: >=0 number of decoded octets
+ -1 syntax error
+*/
+
+static int hex_decode(uschar *src, uschar *end, uschar *dst)
+{
+int decoded=0;
+
+while (*src==' ' || *src=='\t' || *src=='\n') ++src;
+do
+ {
+ int x,d,n;
+
+ 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;
+ }
+while (src<end);
+return decoded;
+}
+
+
+/*************************************************
+* Decode unicode-encoded-character string *
+*************************************************/
+
+/*
+Encoding definition:
+ 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
+ dst points to the destination of the decoded octets,
+ optionally to (uschar*)0 for checking only
+
+Returns: >=0 number of decoded octets
+ -1 syntax error
+ -2 semantic error (character range violation)
+*/
+
+static int unicode_decode(uschar *src, uschar *end, uschar *dst)
+{
+int decoded=0;
+
+while (*src==' ' || *src=='\t' || *src=='\n') ++src;
+do
+ {
+ uschar *hex_seq;
+ int c,d,n;
+
+ 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)
+ {
+ 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;
+ }
+ }
+while (src<end);
+return decoded;
+}
+
+
+/*************************************************
+* Decode encoded-character string *
+*************************************************/
+
+/*
+Encoding definition:
+ encoded-arb-octets = "${hex:" hex-pair-seq "}"
+ encoded-unicode-char = "${unicode:" unicode-hex-seq "}"
+
+Arguments:
+ encoded points to an encoded string, returns decoded string
+ filter points to the Sieve filter including its state
+
+Returns: 1 success
+ -1 syntax error
+*/
+
+static int string_decode(struct Sieve *filter, struct String *data)
+{
+uschar *src,*dst,*end;
+
+src=data->character;
+dst=src;
+end=data->character+data->length;
+while (src<end)
+ {
+ uschar *brace;
+
+ if (
+ strncmpic(src,US "${hex:",6)==0
+ && (brace=Ustrchr(src+6,'}'))!=(uschar*)0
+ && (hex_decode(src+6,brace,(uschar*)0))>=0
+ )
+ {
+ dst+=hex_decode(src+6,brace,dst);
+ src=brace+1;
+ }
+ else if (
+ strncmpic(src,US "${unicode:",10)==0
+ && (brace=Ustrchr(src+10,'}'))!=(uschar*)0
+ )
+ {
+ switch (unicode_decode(src+10,brace,(uschar*)0))
+ {
+ case -2:
+ {
+ filter->errmsg=CUS "unicode character out of range";
+ return -1;
+ }
+ case -1:
+ {
+ *dst++=*src++;
+ break;
+ }
+ default:
+ {
+ dst+=unicode_decode(src+10,brace,dst);
+ src=brace+1;
+ }
+ }
+ }
+ else *dst++=*src++;
+ }
+ data->length=dst-data->character;
+ *dst='\0';
+return 1;
+}
+#endif
+
+
/*************************************************
-* Parse a optional string *
+* Parse an optional string *
*************************************************/
/*
++filter->pc;
/* that way, there will be at least one character allocated */
data->character=string_cat(data->character,&dataCapacity,&foo,CUS "",1);
+#ifdef ENCODED_CHARACTER
+ if (filter->require_encoded_character
+ && string_decode(filter,data)==-1)
+ return -1;
+#endif
return 1;
}
else if (*filter->pc=='\\' && *(filter->pc+1)) /* quoted character */
filter->pc+=2;
#endif
++filter->line;
+#ifdef ENCODED_CHARACTER
+ if (filter->require_encoded_character
+ && string_decode(filter,data)==-1)
+ return -1;
+#endif
return 1;
}
else if (*filter->pc=='.' && *(filter->pc+1)=='.') /* remove dot stuffing */
{
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];
projects / exim.git / blobdiff