X-Git-Url: https://git.exim.org/exim.git/blobdiff_plain/0a49a7a4f1090b6f1ce1d0f9d969804c9226b53e..43ba2742c700d625dcdcdaf7bbadc2f72776854a:/src/src/string.c

diff --git a/src/src/string.c b/src/src/string.c
index e7a3b92a5..b7e7db487 100644
--- a/src/src/string.c
+++ b/src/src/string.c
@@ -1,10 +1,8 @@
-/* $Cambridge: exim/src/src/string.c,v 1.15 2009/11/16 19:50:37 nm4 Exp $ */
-
 /*************************************************
 *     Exim - an Internet mail transport agent    *
 *************************************************/
 
-/* Copyright (c) University of Cambridge 1995 - 2009 */
+/* Copyright (c) University of Cambridge 1995 - 2015 */
 /* See the file NOTICE for conditions of use and distribution. */
 
 /* Miscellaneous string-handling functions. Some are not required for
@@ -36,7 +34,7 @@ Returns:    0 if the string is not a textual representation of an IP address
 */
 
 int
-string_is_ip_address(uschar *s, int *maskptr)
+string_is_ip_address(const uschar *s, int *maskptr)
 {
 int i;
 int yield = 4;
@@ -46,7 +44,7 @@ offset. */
 
 if (maskptr != NULL)
   {
-  uschar *ss = s + Ustrlen(s);
+  const uschar *ss = s + Ustrlen(s);
   *maskptr = 0;
   if (s != ss && isdigit(*(--ss)))
     {
@@ -167,7 +165,7 @@ Returns:      pointer to the buffer
 uschar *
 string_format_size(int size, uschar *buffer)
 {
-if (size == 0) Ustrcpy(CS buffer, "     ");
+if (size == 0) Ustrcpy(buffer, "     ");
 else if (size < 1024) sprintf(CS buffer, "%5d", size);
 else if (size < 10*1024)
   sprintf(CS buffer, "%4.1fK", (double)size / 1024.0);
@@ -212,7 +210,6 @@ return yield;
 
 
 
-#ifndef COMPILE_UTILITY
 /*************************************************
 *          Interpret escape sequence             *
 *************************************************/
@@ -227,10 +224,13 @@ Returns:   the value of the character escape
 */
 
 int
-string_interpret_escape(uschar **pp)
+string_interpret_escape(const uschar **pp)
 {
+#ifdef COMPILE_UTILITY
+const uschar *hex_digits= CUS"0123456789abcdef";
+#endif
 int ch;
-uschar *p = *pp;
+const uschar *p = *pp;
 ch = *(++p);
 if (isdigit(ch) && ch != '8' && ch != '9')
   {
@@ -244,9 +244,12 @@ if (isdigit(ch) && ch != '8' && ch != '9')
   }
 else switch(ch)
   {
+  case 'b':  ch = '\b'; break;
+  case 'f':  ch = '\f'; break;
   case 'n':  ch = '\n'; break;
   case 'r':  ch = '\r'; break;
   case 't':  ch = '\t'; break;
+  case 'v':  ch = '\v'; break;
   case 'x':
   ch = 0;
   if (isxdigit(p[1]))
@@ -261,7 +264,6 @@ else switch(ch)
 *pp = p;
 return ch;
 }
-#endif  /* COMPILE_UTILITY */
 
 
 
@@ -282,12 +284,12 @@ Arguments:
 Returns:        string with non-printers encoded as printing sequences
 */
 
-uschar *
-string_printing2(uschar *s, BOOL allow_tab)
+const uschar *
+string_printing2(const uschar *s, BOOL allow_tab)
 {
 int nonprintcount = 0;
 int length = 0;
-uschar *t = s;
+const uschar *t = s;
 uschar *ss, *tt;
 
 while (*t != 0)
@@ -302,7 +304,7 @@ if (nonprintcount == 0) return s;
 /* Get a new block of store guaranteed big enough to hold the
 expanded string. */
 
-ss = store_get(length + nonprintcount * 4 + 1);
+ss = store_get(length + nonprintcount * 3 + 1);
 
 /* Copy everying, escaping non printers. */
 
@@ -333,6 +335,73 @@ return ss;
 }
 #endif  /* COMPILE_UTILITY */
 
+/*************************************************
+*        Undo printing escapes in string         *
+*************************************************/
+
+/* This function is the reverse of string_printing2.  It searches for
+backslash characters and if any are found, it makes a new copy of the
+string with escape sequences parsed.  Otherwise it returns the original
+string.
+
+Arguments:
+  s             the input string
+
+Returns:        string with printing escapes parsed back
+*/
+
+uschar *
+string_unprinting(uschar *s)
+{
+uschar *p, *q, *r, *ss;
+int len, off;
+
+p = Ustrchr(s, '\\');
+if (!p) return s;
+
+len = Ustrlen(s) + 1;
+ss = store_get(len);
+
+q = ss;
+off = p - s;
+if (off)
+  {
+  memcpy(q, s, off);
+  q += off;
+  }
+
+while (*p)
+  {
+  if (*p == '\\')
+    {
+    *q++ = string_interpret_escape((const uschar **)&p);
+    p++;
+    }
+  else
+    {
+    r = Ustrchr(p, '\\');
+    if (!r)
+      {
+      off = Ustrlen(p);
+      memcpy(q, p, off);
+      p += off;
+      q += off;
+      break;
+      }
+    else
+      {
+      off = r - p;
+      memcpy(q, p, off);
+      q += off;
+      p = r;
+      }
+    }
+  }
+*q = '\0';
+
+return ss;
+}
+
 
 
 
@@ -347,7 +416,7 @@ Returns:  copy of string in new store
 */
 
 uschar *
-string_copy(uschar *s)
+string_copy(const uschar *s)
 {
 int len = Ustrlen(s) + 1;
 uschar *ss = store_get(len);
@@ -368,7 +437,7 @@ Returns:  copy of string in new store
 */
 
 uschar *
-string_copy_malloc(uschar *s)
+string_copy_malloc(const uschar *s)
 {
 int len = Ustrlen(s) + 1;
 uschar *ss = store_malloc(len);
@@ -388,7 +457,7 @@ Returns:  copy of string in new store, with letters lowercased
 */
 
 uschar *
-string_copylc(uschar *s)
+string_copylc(const uschar *s)
 {
 uschar *ss = store_get(Ustrlen(s) + 1);
 uschar *p = ss;
@@ -414,7 +483,7 @@ Returns:    copy of string in new store
 */
 
 uschar *
-string_copyn(uschar *s, int n)
+string_copyn(const uschar *s, int n)
 {
 uschar *ss = store_get(n + 1);
 Ustrncpy(ss, s, n);
@@ -570,9 +639,9 @@ Returns:   the new string
 */
 
 uschar *
-string_dequote(uschar **sptr)
+string_dequote(const uschar **sptr)
 {
-uschar *s = *sptr;
+const uschar *s = *sptr;
 uschar *t, *yield;
 
 /* First find the end of the string */
@@ -641,14 +710,15 @@ Returns:    pointer to fresh piece of store containing sprintf'ed string
 */
 
 uschar *
-string_sprintf(char *format, ...)
+string_sprintf(const char *format, ...)
 {
 va_list ap;
 uschar buffer[STRING_SPRINTF_BUFFER_SIZE];
 va_start(ap, format);
 if (!string_vformat(buffer, sizeof(buffer), format, ap))
   log_write(0, LOG_MAIN|LOG_PANIC_DIE,
-    "string_sprintf expansion was longer than %d", sizeof(buffer));
+    "string_sprintf expansion was longer than " SIZE_T_FMT " (%s)",
+    sizeof(buffer), format);
 va_end(ap);
 return string_copy(buffer);
 }
@@ -669,7 +739,7 @@ Returns:    < 0, = 0, or > 0, according to the comparison
 */
 
 int
-strncmpic(uschar *s, uschar *t, int n)
+strncmpic(const uschar *s, const uschar *t, int n)
 {
 while (n--)
   {
@@ -693,7 +763,7 @@ Returns:    < 0, = 0, or > 0, according to the comparison
 */
 
 int
-strcmpic(uschar *s, uschar *t)
+strcmpic(const uschar *s, const uschar *t)
 {
 while (*s != 0)
   {
@@ -798,10 +868,10 @@ Returns:     pointer to buffer, containing the next substring,
 */
 
 uschar *
-string_nextinlist(uschar **listptr, int *separator, uschar *buffer, int buflen)
+string_nextinlist(const uschar **listptr, int *separator, uschar *buffer, int buflen)
 {
-register int sep = *separator;
-register uschar *s = *listptr;
+int sep = *separator;
+const uschar *s = *listptr;
 BOOL sep_is_special;
 
 if (s == NULL) return NULL;
@@ -858,7 +928,7 @@ else
   {
   int size = 0;
   int ptr = 0;
-  uschar *ss;
+  const uschar *ss;
 
   /* We know that *s != 0 at this point. However, it might be pointing to a
   separator, which could indicate an empty string, or (if an ispunct()
@@ -897,6 +967,95 @@ return buffer;
 #endif  /* COMPILE_UTILITY */
 
 
+#ifndef COMPILE_UTILITY
+/************************************************
+*	Add element to seperated list           *
+************************************************/
+/* This function is used to build a list, returning
+an allocated null-terminated growable string. The
+given element has any embedded seperator characters
+doubled.
+
+Arguments:
+  list	points to the start of the list that is being built, or NULL
+	if this is a new list that has no contents yet
+  sep	list seperator charactoer
+  ele	new lement to be appended to the list
+
+Returns:  pointer to the start of the list, changed if copied for expansion.
+*/
+
+uschar *
+string_append_listele(uschar * list, uschar sep, const uschar * ele)
+{
+uschar * new = NULL;
+int sz = 0, off = 0;
+uschar * sp;
+
+if (list)
+  {
+  new = string_cat(new, &sz, &off, list, Ustrlen(list));
+  new = string_cat(new, &sz, &off, &sep, 1);
+  }
+
+while((sp = Ustrchr(ele, sep)))
+  {
+  new = string_cat(new, &sz, &off, ele, sp-ele+1);
+  new = string_cat(new, &sz, &off, &sep, 1);
+  ele = sp+1;
+  }
+new = string_cat(new, &sz, &off, ele, Ustrlen(ele));
+new[off] = '\0';
+return new;
+}
+
+
+static const uschar *
+Ustrnchr(const uschar * s, int c, unsigned * len)
+{
+unsigned siz = *len;
+while (siz)
+  {
+  if (!*s) return NULL;
+  if (*s == c)
+    {
+    *len = siz;
+    return s;
+    }
+  s++;
+  siz--;
+  }
+return NULL;
+}
+
+uschar *
+string_append_listele_n(uschar * list, uschar sep, const uschar * ele,
+  unsigned len)
+{
+uschar * new = NULL;
+int sz = 0, off = 0;
+const uschar * sp;
+
+if (list)
+  {
+  new = string_cat(new, &sz, &off, list, Ustrlen(list));
+  new = string_cat(new, &sz, &off, &sep, 1);
+  }
+
+while((sp = Ustrnchr(ele, sep, &len)))
+  {
+  new = string_cat(new, &sz, &off, ele, sp-ele+1);
+  new = string_cat(new, &sz, &off, &sep, 1);
+  ele = sp+1;
+  len--;
+  }
+new = string_cat(new, &sz, &off, ele, len);
+new[off] = '\0';
+return new;
+}
+#endif  /* COMPILE_UTILITY */
+
+
 
 #ifndef COMPILE_UTILITY
 /*************************************************
@@ -1039,7 +1198,7 @@ as a va_list item.
 The formats are the usual printf() ones, with some omissions (never used) and
 two additions for strings: %S forces lower case, and %#s or %#S prints nothing
 for a NULL string. Without the # "NULL" is printed (useful in debugging). There
-is also the addition of %D, which inserts the date in the form used for
+is also the addition of %D and %M, which insert the date in the form used for
 datestamped log files.
 
 Arguments:
@@ -1052,7 +1211,7 @@ Returns:       TRUE if the result fitted in the buffer
 */
 
 BOOL
-string_format(uschar *buffer, int buflen, char *format, ...)
+string_format(uschar *buffer, int buflen, const char *format, ...)
 {
 BOOL yield;
 va_list ap;
@@ -1064,17 +1223,20 @@ return yield;
 
 
 BOOL
-string_vformat(uschar *buffer, int buflen, char *format, va_list ap)
+string_vformat(uschar *buffer, int buflen, const char *format, va_list ap)
 {
-enum { L_NORMAL, L_SHORT, L_LONG, L_LONGLONG, L_LONGDOUBLE };
+/* We assume numbered ascending order, C does not guarantee that */
+enum { L_NORMAL=1, L_SHORT=2, L_LONG=3, L_LONGLONG=4, L_LONGDOUBLE=5, L_SIZE=6 };
 
 BOOL yield = TRUE;
 int width, precision;
-char *fp = format;             /* Deliberately not unsigned */
+const char *fp = format;       /* Deliberately not unsigned */
 uschar *p = buffer;
 uschar *last = buffer + buflen - 1;
 
 string_datestamp_offset = -1;  /* Datestamp not inserted */
+string_datestamp_length = 0;   /* Datestamp not inserted */
+string_datestamp_type = 0;     /* Datestamp not inserted */
 
 /* Scan the format and handle the insertions */
 
@@ -1083,8 +1245,8 @@ while (*fp != 0)
   int length = L_NORMAL;
   int *nptr;
   int slen;
-  char *null = "NULL";         /* ) These variables */
-  char *item_start, *s;        /* ) are deliberately */
+  const char *null = "NULL";   /* ) These variables */
+  const char *item_start, *s;  /* ) are deliberately */
   char newformat[16];          /* ) not unsigned */
 
   /* Non-% characters just get copied verbatim */
@@ -1134,7 +1296,7 @@ while (*fp != 0)
       }
     }
 
-  /* Skip over 'h', 'L', 'l', and 'll', remembering the item length */
+  /* Skip over 'h', 'L', 'l', 'll' and 'z', remembering the item length */
 
   if (*fp == 'h')
     { fp++; length = L_SHORT; }
@@ -1153,6 +1315,8 @@ while (*fp != 0)
       length = L_LONG;
       }
     }
+  else if (*fp == 'z')
+    { fp++; length = L_SIZE; }
 
   /* Handle each specific format type. */
 
@@ -1182,6 +1346,7 @@ while (*fp != 0)
       case L_NORMAL:   sprintf(CS p, newformat, va_arg(ap, int)); break;
       case L_LONG:     sprintf(CS p, newformat, va_arg(ap, long int)); break;
       case L_LONGLONG: sprintf(CS p, newformat, va_arg(ap, LONGLONG_T)); break;
+      case L_SIZE:     sprintf(CS p, newformat, va_arg(ap, size_t)); break;
       }
     while (*p) p++;
     break;
@@ -1229,19 +1394,31 @@ while (*fp != 0)
     *p++ = va_arg(ap, int);
     break;
 
-    case 'D':                   /* Insert datestamp for log file names */
-    s = CS tod_stamp(tod_log_datestamp);
+    case 'D':                   /* Insert daily datestamp for log file names */
+    s = CS tod_stamp(tod_log_datestamp_daily);
+    string_datestamp_offset = p - buffer;   /* Passed back via global */
+    string_datestamp_length = Ustrlen(s);   /* Passed back via global */
+    string_datestamp_type = tod_log_datestamp_daily;
+    slen = string_datestamp_length;
+    goto INSERT_STRING;
+
+    case 'M':                   /* Insert monthly datestamp for log file names */
+    s = CS tod_stamp(tod_log_datestamp_monthly);
     string_datestamp_offset = p - buffer;   /* Passed back via global */
+    string_datestamp_length = Ustrlen(s);   /* Passed back via global */
+    string_datestamp_type = tod_log_datestamp_monthly;
+    slen = string_datestamp_length;
     goto INSERT_STRING;
 
     case 's':
     case 'S':                   /* Forces *lower* case */
     s = va_arg(ap, char *);
 
-    INSERT_STRING:              /* Come to from %D above */
     if (s == NULL) s = null;
     slen = Ustrlen(s);
 
+    INSERT_STRING:              /* Come to from %D or %M above */
+
     /* If the width is specified, check that there is a precision
     set; if not, set it to the width to prevent overruns of long
     strings. */
@@ -1326,7 +1503,7 @@ Returns:        a message, in dynamic store
 */
 
 uschar *
-string_open_failed(int eno, char *format, ...)
+string_open_failed(int eno, const char *format, ...)
 {
 va_list ap;
 uschar buffer[1024];
@@ -1370,14 +1547,35 @@ static uschar *
 string_get_localpart(address_item *addr, uschar *yield, int *sizeptr,
   int *ptrptr)
 {
-if (testflag(addr, af_include_affixes) && addr->prefix != NULL)
-  yield = string_cat(yield, sizeptr, ptrptr, addr->prefix,
-    Ustrlen(addr->prefix));
-yield = string_cat(yield, sizeptr, ptrptr, addr->local_part,
-  Ustrlen(addr->local_part));
-if (testflag(addr, af_include_affixes) && addr->suffix != NULL)
-  yield = string_cat(yield, sizeptr, ptrptr, addr->suffix,
-    Ustrlen(addr->suffix));
+uschar * s;
+
+s = addr->prefix;
+if (testflag(addr, af_include_affixes) && s)
+  {
+#ifdef EXPERIMENTAL_INTERNATIONAL
+  if (testflag(addr, af_utf8_downcvt))
+    s = string_localpart_utf8_to_alabel(s, NULL);
+#endif
+  yield = string_cat(yield, sizeptr, ptrptr, s, Ustrlen(s));
+  }
+
+s = addr->local_part;
+#ifdef EXPERIMENTAL_INTERNATIONAL
+if (testflag(addr, af_utf8_downcvt))
+  s = string_localpart_utf8_to_alabel(s, NULL);
+#endif
+yield = string_cat(yield, sizeptr, ptrptr, s, Ustrlen(s));
+
+s = addr->suffix;
+if (testflag(addr, af_include_affixes) && s)
+  {
+#ifdef EXPERIMENTAL_INTERNATIONAL
+  if (testflag(addr, af_utf8_downcvt))
+    s = string_localpart_utf8_to_alabel(s, NULL);
+#endif
+  yield = string_cat(yield, sizeptr, ptrptr, s, Ustrlen(s));
+  }
+
 return yield;
 }
 
@@ -1438,10 +1636,15 @@ else
   {
   if (addr->local_part != NULL)
     {
+    const uschar * s;
     yield = string_get_localpart(addr, yield, &size, &ptr);
     yield = string_cat(yield, &size, &ptr, US"@", 1);
-    yield = string_cat(yield, &size, &ptr, addr->domain,
-      Ustrlen(addr->domain) );
+    s = addr->domain;
+#ifdef EXPERIMENTAL_INTERNATIONAL
+    if (testflag(addr, af_utf8_downcvt))
+      s = string_localpart_utf8_to_alabel(s, NULL);
+#endif
+    yield = string_cat(yield, &size, &ptr, s, Ustrlen(s) );
     }
   else
     {
@@ -1503,6 +1706,17 @@ return yield;
 #endif  /* COMPILE_UTILITY */
 
 
+#ifndef COMPILE_UTILITY
+/* qsort(3), currently used to sort the environment variables
+for -bP environment output, needs a function to compare two pointers to string
+pointers. Here it is. */
+
+int
+string_compare_by_pointer(const uschar **a, const uschar **b)
+{
+return Ustrcmp(CUS *a, CUS *b);
+}
+#endif /* COMPILE_UTILITY */