Fix filter noerror command. Bug 2318

[exim.git] / src / src / sieve.c

diff --git a/src/src/sieve.c b/src/src/sieve.c
index 3d7e99b27b332c27317917f4bd8db12a865fcbef..f5329da0c69523d983beb0f14fe9d078cffd3054 100644 (file)
--- a/src/src/sieve.c
+++ b/src/src/sieve.c
@@ -2,8 +2,10 @@
 *     Exim - an Internet mail transport agent    *
 *************************************************/
 
 *     Exim - an Internet mail transport agent    *
 *************************************************/
 

-/* Copyright (c) Michael Haardt 2003 - 2015 */
-/* See the file NOTICE for conditions of use and distribution. */
+/* Copyright (c) Michael Haardt 2003 - 2015
+ * Copyright (c) The Exim Maintainers 2016 - 2018
+ * See the file NOTICE for conditions of use and distribution.
+ */

 
 /* This code was contributed by Michael Haardt. */
 
 
 /* This code was contributed by Michael Haardt. */
 


@@ -19,7 +21,7 @@
 #include "exim.h"
 
 #if HAVE_ICONV
 #include "exim.h"
 
 #if HAVE_ICONV

-#include <iconv.h>
+# include <iconv.h>

 #endif
 
 /* Define this for RFC compliant \r\n end-of-line terminators.      */
 #endif
 
 /* Define this for RFC compliant \r\n end-of-line terminators.      */


@@ -151,8 +153,10 @@ static uschar str_cc_c[]="Cc";
 static const struct String str_cc={ str_cc_c, 2 };
 static uschar str_bcc_c[]="Bcc";
 static const struct String str_bcc={ str_bcc_c, 3 };
 static const struct String str_cc={ str_cc_c, 2 };
 static uschar str_bcc_c[]="Bcc";
 static const struct String str_bcc={ str_bcc_c, 3 };

+#ifdef ENVELOPE_AUTH

 static uschar str_auth_c[]="auth";
 static const struct String str_auth={ str_auth_c, 4 };
 static uschar str_auth_c[]="auth";
 static const struct String str_auth={ str_auth_c, 4 };

+#endif

 static uschar str_sender_c[]="Sender";
 static const struct String str_sender={ str_sender_c, 6 };
 static uschar str_resent_from_c[]="Resent-From";
 static uschar str_sender_c[]="Sender";
 static const struct String str_sender={ str_sender_c, 6 };
 static uschar str_resent_from_c[]="Resent-From";


@@ -232,6 +236,9 @@ uschar *new = NULL;
 uschar ch;
 size_t line;
 
 uschar ch;
 size_t line;
 

+/* Two passes: one to count output allocation size, second
+to do the encoding */
+

 for (pass=0; pass<=1; ++pass)
   {
   line=0;
 for (pass=0; pass<=1; ++pass)
   {
   line=0;


@@ -245,55 +252,47 @@ for (pass=0; pass<=1; ++pass)
   for (start=src->character,end=start+src->length; start<end; ++start)
     {
     ch=*start;
   for (start=src->character,end=start+src->length; start<end; ++start)
     {
     ch=*start;

-    if (line>=73)
+    if (line>=73)      /* line length limit */

       {
       if (pass==0)
         dst->length+=2;
       else
         {
       {
       if (pass==0)
         dst->length+=2;
       else
         {

-        *new++='=';
+        *new++='=';    /* line split */

         *new++='\n';
         }
       line=0;
       }
         *new++='\n';
         }
       line=0;
       }

-    if
-      (
-      (ch>=33 && ch<=60)
-      || (ch>=62 && ch<=126)
-      ||
-        (
-        (ch==9 || ch==32)
-        && start+2<end
-        && (*(start+1)!='\r' || *(start+2)!='\n')
-        )
-      )
+    if (  (ch>='!' && ch<='<')
+       || (ch>='>' && ch<='~')
+       || (  (ch=='\t' || ch==' ')
+          && start+2<end
+          && (*(start+1)!='\r' || *(start+2)!='\n')    /* CRLF */
+          )
+       )

       {
       if (pass==0)
         ++dst->length;
       else
       {
       if (pass==0)
         ++dst->length;
       else

-        *new++=*start;
+        *new++=*start; /* copy char */

       ++line;
       }
       ++line;
       }

-    else if (ch=='\r' && start+1<end && *(start+1)=='\n')
+    else if (ch=='\r' && start+1<end && *(start+1)=='\n') /* CRLF */

       {
       if (pass==0)
       {
       if (pass==0)

-        {

         ++dst->length;
         ++dst->length;

-        line=0;
-        }

       else
       else

-        *new++='\n';
-        line=0;
-      ++start;
+        *new++='\n';                                   /* NL */
+      line=0;
+      ++start; /* consume extra input char */

       }
     else
       {
       if (pass==0)
         dst->length+=3;
       else
       }
     else
       {
       if (pass==0)
         dst->length+=3;
       else

-        {
-        sprintf(CS new,"=%02X",ch);
-        new+=3;
+        {              /* encoded char */
+        new += sprintf(CS new,"=%02X",ch);

         }
       line+=3;
       }
         }
       line+=3;
       }


@@ -412,11 +411,14 @@ Returns
  -1           syntax error
 */
 
  -1           syntax error
 */
 

-static int parse_mailto_uri(struct Sieve *filter, const uschar *uri, string_item **recipient, struct String *header, struct String *subject, struct String *body)
+static int
+parse_mailto_uri(struct Sieve *filter, const uschar *uri,
+  string_item **recipient, struct String *header, struct String *subject,
+  struct String *body)

 {
 const uschar *start;
 {
 const uschar *start;

-struct String to,hname,hvalue;
-int capacity;
+struct String to, hname;
+struct String hvalue = {.character = NULL, .length = 0};

 string_item *new;
 
 if (Ustrncmp(uri,"mailto:",7))
 string_item *new;
 
 if (Ustrncmp(uri,"mailto:",7))


@@ -424,6 +426,7 @@ if (Ustrncmp(uri,"mailto:",7))
   filter->errmsg=US "Unknown URI scheme";
   return 0;
   }
   filter->errmsg=US "Unknown URI scheme";
   return 0;
   }

+

 uri+=7;
 if (*uri && *uri!='?')
   for (;;)
 uri+=7;
 if (*uri && *uri!='?')
   for (;;)


@@ -432,22 +435,21 @@ if (*uri && *uri!='?')
     for (start=uri; *uri && *uri!='?' && (*uri!='%' || *(uri+1)!='2' || tolower(*(uri+2))!='c'); ++uri);
     if (uri>start)
       {
     for (start=uri; *uri && *uri!='?' && (*uri!='%' || *(uri+1)!='2' || tolower(*(uri+2))!='c'); ++uri);
     if (uri>start)
       {

-      capacity=0;
-      to.character=(uschar*)0;
-      to.length=0;
-      to.character=string_cat(to.character,&capacity,&to.length,start,uri-start);
-      to.character[to.length]='\0';
+      gstring * g = string_catn(NULL, start, uri-start);
+
+      to.character = string_from_gstring(g);
+      to.length = g->ptr;

       if (uri_decode(&to)==-1)
         {
         filter->errmsg=US"Invalid URI encoding";
         return -1;
         }
       if (uri_decode(&to)==-1)
         {
         filter->errmsg=US"Invalid URI encoding";
         return -1;
         }

-        new=store_get(sizeof(string_item));
-        new->text=store_get(to.length+1);
-        if (to.length) memcpy(new->text,to.character,to.length);
-        new->text[to.length]='\0';
-        new->next=*recipient;
-        *recipient=new;
+      new=store_get(sizeof(string_item));
+      new->text=store_get(to.length+1);
+      if (to.length) memcpy(new->text,to.character,to.length);
+      new->text[to.length]='\0';
+      new->next=*recipient;
+      *recipient=new;

       }
     else
       {
       }
     else
       {


@@ -466,11 +468,10 @@ if (*uri=='?')
     for (start=uri; *uri && (isalnum(*uri) || strchr("$-_.+!*'(),%",*uri)); ++uri);
     if (uri>start)
       {
     for (start=uri; *uri && (isalnum(*uri) || strchr("$-_.+!*'(),%",*uri)); ++uri);
     if (uri>start)
       {

-      capacity=0;
-      hname.character=(uschar*)0;
-      hname.length=0;
-      hname.character=string_cat(hname.character,&capacity,&hname.length,start,uri-start);
-      hname.character[hname.length]='\0';
+      gstring * g = string_catn(NULL, start, uri-start);
+
+      hname.character = string_from_gstring(g);
+      hname.length = g->ptr;

       if (uri_decode(&hname)==-1)
         {
         filter->errmsg=US"Invalid URI encoding";
       if (uri_decode(&hname)==-1)
         {
         filter->errmsg=US"Invalid URI encoding";


@@ -489,11 +490,10 @@ if (*uri=='?')
     for (start=uri; *uri && (isalnum(*uri) || strchr("$-_.+!*'(),%",*uri)); ++uri);
     if (uri>start)
       {
     for (start=uri; *uri && (isalnum(*uri) || strchr("$-_.+!*'(),%",*uri)); ++uri);
     if (uri>start)
       {

-      capacity=0;
-      hvalue.character=(uschar*)0;
-      hvalue.length=0;
-      hvalue.character=string_cat(hvalue.character,&capacity,&hvalue.length,start,uri-start);
-      hvalue.character[hvalue.length]='\0';
+      gstring * g = string_catn(NULL, start, uri-start);
+
+      hname.character = string_from_gstring(g);
+      hname.length = g->ptr;

       if (uri_decode(&hvalue)==-1)
         {
         filter->errmsg=US"Invalid URI encoding";
       if (uri_decode(&hvalue)==-1)
         {
         filter->errmsg=US"Invalid URI encoding";


@@ -529,13 +529,18 @@ if (*uri=='?')
       for (i=ignore; i<end && !eq_asciicase(&hname,i,0); ++i);
       if (i==end)
         {
       for (i=ignore; i<end && !eq_asciicase(&hname,i,0); ++i);
       if (i==end)
         {

-        if (header->length==-1) header->length=0;
-        capacity=header->length;
-        header->character=string_cat(header->character,&capacity,&header->length,hname.character,hname.length);
-        header->character=string_cat(header->character,&capacity,&header->length,CUS ": ",2);
-        header->character=string_cat(header->character,&capacity,&header->length,hvalue.character,hvalue.length);
-        header->character=string_cat(header->character,&capacity,&header->length,CUS "\n",1);
-        header->character[header->length]='\0';
+       gstring * g;
+
+        if (header->length==-1) header->length = 0;
+
+       g = string_catn(NULL, header->character, header->length);
+        g = string_catn(g, hname.character, hname.length);
+        g = string_catn(g, CUS ": ", 2);
+        g = string_catn(g, hvalue.character, hvalue.length);
+        g = string_catn(g, CUS "\n", 1);
+
+       header->character = string_from_gstring(g);
+       header->length = g->ptr;

         }
       }
     if (*uri=='&') ++uri;
         }
       }
     if (*uri=='&') ++uri;


@@ -993,10 +998,10 @@ Arguments:
 Returns:      quoted string
 */
 
 Returns:      quoted string
 */
 

-static const uschar *quote(const struct String *header)
+static const uschar *
+quote(const struct String *header)

 {
 {

-uschar *quoted=NULL;
-int size=0,ptr=0;
+gstring * quoted = NULL;

 size_t l;
 const uschar *h;
 
 size_t l;
 const uschar *h;
 


@@ -1007,26 +1012,20 @@ while (l)
   switch (*h)
     {
     case '\0':
   switch (*h)
     {
     case '\0':

-      {
-      quoted=string_cat(quoted,&size,&ptr,CUS "\\0",2);
+      quoted = string_catn(quoted, CUS "\\0", 2);

       break;
       break;

-      }

     case '$':
     case '{':
     case '}':
     case '$':
     case '{':
     case '}':

-      {
-      quoted=string_cat(quoted,&size,&ptr,CUS "\\",1);
-      }
+      quoted = string_catn(quoted, CUS "\\", 1);

     default:
     default:

-      {
-      quoted=string_cat(quoted,&size,&ptr,h,1);
-      }
+      quoted = string_catn(quoted, h, 1);

     }
   ++h;
   --l;
   }
     }
   ++h;
   --l;
   }

-quoted=string_cat(quoted,&size,&ptr,CUS "",1);
-return quoted;
+quoted = string_catn(quoted, CUS "", 1);
+return string_from_gstring(quoted);

 }
 
 
 }
 
 


@@ -1046,30 +1045,33 @@ Arguments:
 Returns:      nothing
 */
 
 Returns:      nothing
 */
 

-static void add_addr(address_item **generated, uschar *addr, int file, int maxage, int maxmessages, int maxstorage)
+static void
+add_addr(address_item **generated, uschar *addr, int file, int maxage, int maxmessages, int maxstorage)

 {
 address_item *new_addr;
 
 for (new_addr=*generated; new_addr; new_addr=new_addr->next)
 {
 address_item *new_addr;
 
 for (new_addr=*generated; new_addr; new_addr=new_addr->next)

-  {
-  if (Ustrcmp(new_addr->address,addr)==0 && (file ? testflag(new_addr, af_pfr|af_file) : 1))
+  if (  Ustrcmp(new_addr->address,addr) == 0
+     && (  !file
+       || testflag(new_addr, af_pfr)
+       || testflag(new_addr, af_file)
+       )
+     )

     {
     if ((filter_test != FTEST_NONE && debug_selector != 0) || (debug_selector & D_filter) != 0)
     {
     if ((filter_test != FTEST_NONE && debug_selector != 0) || (debug_selector & D_filter) != 0)

-      {

       debug_printf("Repeated %s `%s' ignored.\n",file ? "fileinto" : "redirect", addr);
       debug_printf("Repeated %s `%s' ignored.\n",file ? "fileinto" : "redirect", addr);

-      }
+

     return;
     }
     return;
     }

-  }

 
 if ((filter_test != FTEST_NONE && debug_selector != 0) || (debug_selector & D_filter) != 0)
 
 if ((filter_test != FTEST_NONE && debug_selector != 0) || (debug_selector & D_filter) != 0)

-  {

   debug_printf("%s `%s'\n",file ? "fileinto" : "redirect", addr);
   debug_printf("%s `%s'\n",file ? "fileinto" : "redirect", addr);

-  }
-new_addr=deliver_make_addr(addr,TRUE);
+
+new_addr = deliver_make_addr(addr,TRUE);

 if (file)
   {
 if (file)
   {

-  setflag(new_addr, af_pfr|af_file);
+  setflag(new_addr, af_pfr);
+  setflag(new_addr, af_file);

   new_addr->mode = 0;
   }
 new_addr->prop.errors_address = NULL;
   new_addr->mode = 0;
   }
 new_addr->prop.errors_address = NULL;


@@ -1472,12 +1474,14 @@ Returns:      1                success
               0                identifier not matched
 */
 
               0                identifier not matched
 */
 

-static int parse_string(struct Sieve *filter, struct String *data)
+static int
+parse_string(struct Sieve *filter, struct String *data)

 {
 {

-int dataCapacity=0;
+gstring * g = NULL;
+
+data->length = 0;
+data->character = NULL;

 
 

-data->length=0;
-data->character=(uschar*)0;

 if (*filter->pc=='"') /* quoted string */
   {
   ++filter->pc;
 if (*filter->pc=='"') /* quoted string */
   {
   ++filter->pc;


@@ -1485,11 +1489,17 @@ if (*filter->pc=='"') /* quoted string */
     {
     if (*filter->pc=='"') /* end of string */
       {
     {
     if (*filter->pc=='"') /* end of string */
       {

-      int foo=data->length;
-

       ++filter->pc;
       ++filter->pc;

+
+      if (g)
+       {
+       data->character = string_from_gstring(g);
+       data->length = g->ptr;
+       }
+      else
+       data->character = US"\0";

       /* that way, there will be at least one character allocated */
       /* 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)
 #ifdef ENCODED_CHARACTER
       if (filter->require_encoded_character
           && string_decode(filter,data)==-1)


@@ -1499,7 +1509,7 @@ if (*filter->pc=='"') /* quoted string */
       }
     else if (*filter->pc=='\\' && *(filter->pc+1)) /* quoted character */
       {
       }
     else if (*filter->pc=='\\' && *(filter->pc+1)) /* quoted character */
       {

-      data->character=string_cat(data->character,&dataCapacity,&data->length,filter->pc+1,1);
+      g = string_catn(g, filter->pc+1, 1);

       filter->pc+=2;
       }
     else /* regular character */
       filter->pc+=2;
       }
     else /* regular character */


@@ -1509,11 +1519,11 @@ if (*filter->pc=='"') /* quoted string */
 #else
       if (*filter->pc=='\n')
         {
 #else
       if (*filter->pc=='\n')
         {

-        data->character=string_cat(data->character,&dataCapacity,&data->length,US"\r",1);
+        g = string_catn(g, US"\r", 1);

         ++filter->line;
         }
 #endif
         ++filter->line;
         }
 #endif

-      data->character=string_cat(data->character,&dataCapacity,&data->length,filter->pc,1);
+      g = string_catn(g, filter->pc, 1);

       filter->pc++;
       }
     }
       filter->pc++;
       }
     }


@@ -1555,7 +1565,7 @@ else if (Ustrncmp(filter->pc,CUS "text:",5)==0) /* multiline string */
     if (*filter->pc=='\n') /* end of line */
 #endif
       {
     if (*filter->pc=='\n') /* end of line */
 #endif
       {

-      data->character=string_cat(data->character,&dataCapacity,&data->length,CUS "\r\n",2);
+      g = string_catn(g, CUS "\r\n", 2);

 #ifdef RFC_EOL
       filter->pc+=2;
 #else
 #ifdef RFC_EOL
       filter->pc+=2;
 #else


@@ -1568,10 +1578,15 @@ else if (Ustrncmp(filter->pc,CUS "text:",5)==0) /* multiline string */
       if (*filter->pc=='.' && *(filter->pc+1)=='\n') /* end of string */
 #endif
         {
       if (*filter->pc=='.' && *(filter->pc+1)=='\n') /* end of string */
 #endif
         {

-        int foo=data->length;
+       if (g)
+         {
+         data->character = string_from_gstring(g);
+         data->length = g->ptr;
+         }
+       else
+         data->character = US"\0";
+       /* that way, there will be at least one character allocated */

 
 

-        /* that way, there will be at least one character allocated */
-        data->character=string_cat(data->character,&dataCapacity,&foo,CUS "",1);

 #ifdef RFC_EOL
         filter->pc+=3;
 #else
 #ifdef RFC_EOL
         filter->pc+=3;
 #else


@@ -1587,13 +1602,13 @@ else if (Ustrncmp(filter->pc,CUS "text:",5)==0) /* multiline string */
         }
       else if (*filter->pc=='.' && *(filter->pc+1)=='.') /* remove dot stuffing */
         {
         }
       else if (*filter->pc=='.' && *(filter->pc+1)=='.') /* remove dot stuffing */
         {

-        data->character=string_cat(data->character,&dataCapacity,&data->length,CUS ".",1);
+        g = string_catn(g, CUS ".", 1);

         filter->pc+=2;
         }
       }
     else /* regular character */
       {
         filter->pc+=2;
         }
       }
     else /* regular character */
       {

-      data->character=string_cat(data->character,&dataCapacity,&data->length,filter->pc,1);
+      g = string_catn(g, filter->pc, 1);

       filter->pc++;
       }
     }
       filter->pc++;
       }
     }


@@ -1706,12 +1721,13 @@ Returns:      1                success
               -1               no string list found
 */
 
               -1               no string list found
 */
 

-static int parse_stringlist(struct Sieve *filter, struct String **data)
+static int
+parse_stringlist(struct Sieve *filter, struct String **data)

 {
 const uschar *orig=filter->pc;
 {
 const uschar *orig=filter->pc;

-int dataCapacity=0;
-int dataLength=0;
-struct String *d=(struct String*)0;
+int dataCapacity = 0;
+int dataLength = 0;
+struct String *d = NULL;

 int m;
 
 if (*filter->pc=='[') /* string list */
 int m;
 
 if (*filter->pc=='[') /* string list */


@@ -1720,20 +1736,17 @@ if (*filter->pc=='[') /* string list */
   for (;;)
     {
     if (parse_white(filter)==-1) goto error;
   for (;;)
     {
     if (parse_white(filter)==-1) goto error;

-    if ((dataLength+1)>=dataCapacity) /* increase buffer */
+    if (dataLength+1 >= dataCapacity) /* increase buffer */

       {
       struct String *new;
       {
       struct String *new;

-      int newCapacity;          /* Don't amalgamate with next line; some compilers grumble */
-      newCapacity=dataCapacity?(dataCapacity*=2):(dataCapacity=4);
-      if ((new=(struct String*)store_get(sizeof(struct String)*newCapacity))==(struct String*)0)
-        {
-        filter->errmsg=CUstrerror(errno);
-        goto error;
-        }
+
+      dataCapacity = dataCapacity ? dataCapacity * 2 : 4;
+      new = store_get(sizeof(struct String) * dataCapacity);
+

       if (d) memcpy(new,d,sizeof(struct String)*dataLength);
       if (d) memcpy(new,d,sizeof(struct String)*dataLength);

-      d=new;
-      dataCapacity=newCapacity;
+      d = new;

       }
       }

+

     m=parse_string(filter,&d[dataLength]);
     if (m==0)
       {
     m=parse_string(filter,&d[dataLength]);
     if (m==0)
       {


@@ -2124,7 +2137,7 @@ if (parse_identifier(filter,CUS "address"))
         filter->errmsg=CUS "header string expansion failed";
         return -1;
         }
         filter->errmsg=CUS "header string expansion failed";
         return -1;
         }

-      parse_allow_group = TRUE;
+      f.parse_allow_group = TRUE;

       while (*header_value && !*cond)
         {
         uschar *error;
       while (*header_value && !*cond)
         {
         uschar *error;


@@ -2170,8 +2183,8 @@ if (parse_identifier(filter,CUS "address"))
         if (saveend == 0) break;
         header_value = end_addr + 1;
         }
         if (saveend == 0) break;
         header_value = end_addr + 1;
         }

-      parse_allow_group = FALSE;
-      parse_found_group = FALSE;
+      f.parse_allow_group = FALSE;
+      f.parse_found_group = FALSE;

       }
     }
   return 1;
       }
     }
   return 1;


@@ -3105,7 +3118,7 @@ while (*filter->pc)
                 message.character=US"Notification";
                 message.length=Ustrlen(message.character);
                 }
                 message.character=US"Notification";
                 message.length=Ustrlen(message.character);
                 }

-              /* Allocation is larger than neccessary, but enough even for split MIME words */
+              /* Allocation is larger than necessary, but enough even for split MIME words */

               buffer_capacity=32+4*message.length;
               buffer=store_get(buffer_capacity);
               if (message.length!=-1) fprintf(f,"Subject: %s\n",parse_quote_2047(message.character, message.length, US"utf-8", buffer, buffer_capacity, TRUE));
               buffer_capacity=32+4*message.length;
               buffer=store_get(buffer_capacity);
               if (message.length!=-1) fprintf(f,"Subject: %s\n",parse_quote_2047(message.character, message.length, US"utf-8", buffer, buffer_capacity, TRUE));


@@ -3274,15 +3287,13 @@ while (*filter->pc)
     if (exec)
       {
       address_item *addr;
     if (exec)
       {
       address_item *addr;

-      int capacity,start;

       uschar *buffer;
       int buffer_capacity;
       uschar *buffer;
       int buffer_capacity;

-      struct String key;

       md5 base;
       uschar digest[16];
       uschar hexdigest[33];
       int i;
       md5 base;
       uschar digest[16];
       uschar hexdigest[33];
       int i;

-      uschar *once;
+      gstring * once;

 
       if (filter_personal(aliases,TRUE))
         {
 
       if (filter_personal(aliases,TRUE))
         {


@@ -3294,32 +3305,30 @@ while (*filter->pc)
           }
         /* build oncelog filename */
 
           }
         /* build oncelog filename */
 

-        key.character=(uschar*)0;
-        key.length=0;
-        capacity=0;
+        md5_start(&base);
+

         if (handle.length==-1)
           {
         if (handle.length==-1)
           {

-          if (subject.length!=-1) key.character=string_cat(key.character,&capacity,&key.length,subject.character,subject.length);
-          if (from.length!=-1) key.character=string_cat(key.character,&capacity,&key.length,from.character,from.length);
-          key.character=string_cat(key.character,&capacity,&key.length,reason_is_mime?US"1":US"0",1);
-          key.character=string_cat(key.character,&capacity,&key.length,reason.character,reason.length);
+         gstring * key = NULL;
+          if (subject.length!=-1) key =string_catn(key, subject.character, subject.length);
+          if (from.length!=-1) key = string_catn(key, from.character, from.length);
+          key = string_catn(key, reason_is_mime?US"1":US"0", 1);
+          key = string_catn(key, reason.character, reason.length);
+         md5_end(&base, key->s, key->ptr, digest);

           }
         else
           }
         else

-          key=handle;
-        md5_start(&base);
-        md5_end(&base, key.character, key.length, digest);
+         md5_end(&base, handle.character, handle.length, digest);
+

         for (i = 0; i < 16; i++) sprintf(CS (hexdigest+2*i), "%02X", digest[i]);
         for (i = 0; i < 16; i++) sprintf(CS (hexdigest+2*i), "%02X", digest[i]);

+

         if ((filter_test != FTEST_NONE && debug_selector != 0) || (debug_selector & D_filter) != 0)
         if ((filter_test != FTEST_NONE && debug_selector != 0) || (debug_selector & D_filter) != 0)

-          {

           debug_printf("Sieve: mail was personal, vacation file basename: %s\n", hexdigest);
           debug_printf("Sieve: mail was personal, vacation file basename: %s\n", hexdigest);

-          }
+

         if (filter_test == FTEST_NONE)
           {
         if (filter_test == FTEST_NONE)
           {

-          capacity=Ustrlen(filter->vacation_directory);
-          start=capacity;
-          once=string_cat(filter->vacation_directory,&capacity,&start,US"/",1);
-          once=string_cat(once,&capacity,&start,hexdigest,33);
-          once[start] = '\0';
+          once = string_cat (NULL, filter->vacation_directory);
+          once = string_catn(once, US"/", 1);
+          once = string_catn(once, hexdigest, 33);

 
           /* process subject */
 
 
           /* process subject */
 


@@ -3330,11 +3339,12 @@ while (*filter->pc)
             subject_def=expand_string(US"${if def:header_subject {true}{false}}");
             if (Ustrcmp(subject_def,"true")==0)
               {
             subject_def=expand_string(US"${if def:header_subject {true}{false}}");
             if (Ustrcmp(subject_def,"true")==0)
               {

+             gstring * g = string_catn(NULL, US"Auto: ", 6);
+

               expand_header(&subject,&str_subject);
               expand_header(&subject,&str_subject);

-              capacity=6;
-              start=6;
-              subject.character=string_cat(US"Auto: ",&capacity,&start,subject.character,subject.length);
-              subject.length=start;
+              g = string_catn(g, subject.character, subject.length);
+             subject.character = string_from_gstring(g);
+              subject.length = g->ptr;

               }
             else
               {
               }
             else
               {


@@ -3347,7 +3357,7 @@ while (*filter->pc)
 
           addr = deliver_make_addr(string_sprintf(">%.256s", sender_address), FALSE);
           setflag(addr, af_pfr);
 
           addr = deliver_make_addr(string_sprintf(">%.256s", sender_address), FALSE);
           setflag(addr, af_pfr);

-          setflag(addr, af_ignore_error);
+          addr->prop.ignore_error = TRUE;

           addr->next = *generated;
           *generated = addr;
           addr->reply = store_get(sizeof(reply_item));
           addr->next = *generated;
           *generated = addr;
           addr->reply = store_get(sizeof(reply_item));


@@ -3357,12 +3367,12 @@ while (*filter->pc)
             addr->reply->from = expand_string(US"$local_part@$domain");
           else
             addr->reply->from = from.character;
             addr->reply->from = expand_string(US"$local_part@$domain");
           else
             addr->reply->from = from.character;

-          /* Allocation is larger than neccessary, but enough even for split MIME words */
+          /* Allocation is larger than necessary, but enough even for split MIME words */

           buffer_capacity=32+4*subject.length;
           buffer=store_get(buffer_capacity);
          /* deconst cast safe as we pass in a non-const item */
           addr->reply->subject = US parse_quote_2047(subject.character, subject.length, US"utf-8", buffer, buffer_capacity, TRUE);
           buffer_capacity=32+4*subject.length;
           buffer=store_get(buffer_capacity);
          /* deconst cast safe as we pass in a non-const item */
           addr->reply->subject = US parse_quote_2047(subject.character, subject.length, US"utf-8", buffer, buffer_capacity, TRUE);

-          addr->reply->oncelog=once;
+          addr->reply->oncelog = string_from_gstring(once);

           addr->reply->once_repeat=days*86400;
 
           /* build body and MIME headers */
           addr->reply->once_repeat=days*86400;
 
           /* build body and MIME headers */


@@ -3374,27 +3384,21 @@ while (*filter->pc)
 
             for
               (
 
             for
               (

-              mime_body=reason.character,reason_end=reason.character+reason.length;
-              mime_body<(reason_end-(sizeof(nlnl)-1)) && memcmp(mime_body,nlnl,(sizeof(nlnl)-1));
+              mime_body = reason.character, reason_end = reason.character + reason.length;
+              mime_body < (reason_end-(sizeof(nlnl)-1)) && memcmp(mime_body, nlnl, (sizeof(nlnl)-1));

               ++mime_body
               );
               ++mime_body
               );

-            capacity = 0;
-            start = 0;
-            addr->reply->headers = string_cat(NULL,&capacity,&start,reason.character,mime_body-reason.character);
-            addr->reply->headers[start] = '\0';
-            capacity = 0;
-            start = 0;
+
+            addr->reply->headers = string_copyn(reason.character, mime_body-reason.character);
+

             if (mime_body+(sizeof(nlnl)-1)<reason_end) mime_body+=(sizeof(nlnl)-1);
             else mime_body=reason_end-1;
             if (mime_body+(sizeof(nlnl)-1)<reason_end) mime_body+=(sizeof(nlnl)-1);
             else mime_body=reason_end-1;

-            addr->reply->text = string_cat(NULL,&capacity,&start,mime_body,reason_end-mime_body);
-            addr->reply->text[start] = '\0';
+            addr->reply->text = string_copyn(mime_body, reason_end-mime_body);

             }
           else
             {
             }
           else
             {

-            struct String qp = { NULL, 0 };  /* Keep compiler happy (PH) */
+            struct String qp = { .character = NULL, .length = 0 };  /* Keep compiler happy (PH) */

 
 

-            capacity = 0;
-            start = reason.length;

             addr->reply->headers = US"MIME-Version: 1.0\n"
                                    "Content-Type: text/plain;\n"
                                    "\tcharset=\"utf-8\"\n"
             addr->reply->headers = US"MIME-Version: 1.0\n"
                                    "Content-Type: text/plain;\n"
                                    "\tcharset=\"utf-8\"\n"


@@ -3404,9 +3408,7 @@ while (*filter->pc)
           }
         }
         else if ((filter_test != FTEST_NONE && debug_selector != 0) || (debug_selector & D_filter) != 0)
           }
         }
         else if ((filter_test != FTEST_NONE && debug_selector != 0) || (debug_selector & D_filter) != 0)

-          {

           debug_printf("Sieve: mail was not personal, vacation would ignore it\n");
           debug_printf("Sieve: mail was not personal, vacation would ignore it\n");

-          }

       }
     }
     else break;
       }
     }
     else break;