diff --git a/tools/ftp-indexer.c b/tools/ftp-indexer.c
index 5457f3d..32c6de9 100644
--- a/tools/ftp-indexer.c
+++ b/tools/ftp-indexer.c
@@ -1,1169 +1,1293 @@
 /* ftp-indexer.c - Create an HTML index file for an FTP directory
  * Copyright (C) 2017  g10 Code GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 3 of the License, or
  * (at your option) any later version.
  *
  * This program is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
  *
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, see <https://www.gnu.org/licenses/>.
  */
 
-/* How to use:
+/* The following script is triggred by a cronjob at ftp.gnupg.org to
+ * build the index pages.
+ *
+--8<---------------cut here---------------start------------->8---
+#!/bin/sh
+
+set -e
+top=/home/ftp
+scratch=/home/ftp/.scratch
+cd "$top"
+
+opt_force=no
+if [ "$1" == "--force" ]; then
+  shift
+  opt_force=yes
+fi
+
+INDEXER=/usr/local/bin/ftp-indexer
+if [ ! -x $INDEXER ]; then
+  echo "mk-ftp-index.html.sh: Index tool $INDEXER not found - aborting" >&2
+  exit 1
+fi
+INDEXER_OPTS="--reverse-ver --gpgweb --readme --index $scratch/ftp-index.new"
+INDEXER_OPTS="$INDEXER_OPTS --exclude README --exclude index.html"
+
+
+(find . -type d ! -name '\.*' ! -name dev ; echo .) |\
+ while read dir rest; do
+  dir=${dir##./}
+  if cd "$dir"; then
+    if [ "$dir" = "." ]; then
+      desc="/"
+      extraopt="--exclude dev"
+    else
+      desc="$dir/"
+      extraopt=""
+    fi
+
+    [ -f $scratch/index.html ] && rm $scratch/index.html
+    [ -f index.html ] && cat index.html >$scratch/index.html
+    $INDEXER $INDEXER_OPTS $extraopt . "$desc" >$scratch/index.html.new
+    if [ $opt_force = no -a -f $scratch/index.html ]; then
+      grep -v '^<meta name="date"' $scratch/index.html \
+          | grep -v '^Page last updated on ' >$scratch/index.html.x
+      grep -v '^<meta name="date"' $scratch/index.html.new \
+          | grep -v '^Page last updated on ' >$scratch/index.html.new.x
+      if ! cmp -s $scratch/index.html.x $scratch/index.html.new.x ; then
+         mv $scratch/index.html.new index.html
+         mv $scratch/ftp-index.new .ftp-index
+      fi
+      rm $scratch/index.html
+      [ -f $scratch/index.html.new ] && rm $scratch/index.html.new
+      [ -f $scratch/ftp-index.new ] && rm $scratch/ftp-index.new
+    else
+      mv $scratch/index.html.new index.html
+      mv $scratch/ftp-index.new .ftp-index
+    fi
+  fi
+  cd "$top"
+done
+--8<---------------cut here---------------end--------------->8---
  *
  **/
 
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
 #include <stdarg.h>
 #include <ctype.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <unistd.h>
 #include <dirent.h>
 #include <time.h>
 #include <errno.h>
 
 
 #define PGMNAME "ftp-indexer"
-
+#define VERSION "0.1"
 
 #define DIM(v)		     (sizeof(v)/sizeof((v)[0]))
 #define DIMof(type,member)   DIM(((type *)0)->member)
 #if __GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 5)
 #define ATTR_PRINTF(a,b)    __attribute__ ((format (printf,a,b)))
 #define ATTR_NR_PRINTF(a,b) __attribute__ ((noreturn,format (printf,a,b)))
 #else
 #define ATTR_PRINTF(a,b)
 #define ATTR_NR_PRINTF(a,b)
 #endif
 
 #define digitp(a) ((a) >= '0' && (a) <= '9')
 #define VALID_URI_CHARS "abcdefghijklmnopqrstuvwxyz"   \
                         "ABCDEFGHIJKLMNOPQRSTUVWXYZ"   \
                         "01234567890@"                 \
                         "!\"#$%&'()*+,-./:;<=>?[\\]^_{|}~"
 
 
 /* A simple object to keep strings in a list.  */
 struct strlist_s
 {
   struct strlist_s *next;
   char d[1];
 };
 typedef struct strlist_s *strlist_t;
 
 
 /* An object to collect information about files.  */
 struct finfo_s
 {
   struct finfo_s *next;
   unsigned int is_dir:1;
   unsigned int is_reg:1;
   time_t mtime;
   unsigned long long size;
   char name[1];
 };
 typedef struct finfo_s *finfo_t;
 
 
 static int opt_verbose;
 static int opt_debug;
 static int opt_reverse;
 static int opt_reverse_ver;
 static int opt_files_first;
 static int opt_html;
-static int opt_index;
+static const char *opt_index;
 static int opt_gpgweb;
 static int opt_readme;
 static strlist_t opt_exclude;
 
 static void die (const char *format, ...) ATTR_NR_PRINTF(1,2);
 static void err (const char *format, ...) ATTR_PRINTF(1,2);
 static void inf (const char *format, ...) ATTR_PRINTF(1,2);
 
 
 
 static void
 die (const char *fmt, ...)
 {
   va_list arg_ptr;
 
   va_start (arg_ptr, fmt);
   fputs (PGMNAME": ", stderr);
   vfprintf (stderr, fmt, arg_ptr);
   va_end (arg_ptr);
   exit (1);
 }
 
 
 static void
 err (const char *fmt, ...)
 {
   va_list arg_ptr;
 
   va_start (arg_ptr, fmt);
   fputs (PGMNAME": ", stderr);
   vfprintf (stderr, fmt, arg_ptr);
   va_end (arg_ptr);
 }
 
 
 static void
 inf (const char *fmt, ...)
 {
   va_list arg_ptr;
 
   if (!opt_verbose)
     return;
 
   va_start (arg_ptr, fmt);
   fputs (PGMNAME": ", stderr);
   vfprintf (stderr, fmt, arg_ptr);
   va_end (arg_ptr);
 }
 
 
 static void *
 xmalloc (size_t n)
 {
   void *p = malloc (n);
   if (!p)
     die ("out of core\n");
   return p;
 }
 
 
 static void *
 xcalloc (size_t n, size_t k)
 {
   void *p = calloc (n, k);
   if (!p)
     die ("out of core\n");
   return p;
 }
 
 
 static char *
 xstrdup (const char *string)
 {
   char *buf = xmalloc (strlen (string));
   strcpy (buf, string);
   return buf;
 }
 
 
 static inline char *
 my_stpcpy (char *a, const char *b)
 {
   while (*b)
     *a++ = *b++;
   *a = 0;
 
   return (char*)a;
 }
 
 
 /* If SPECIAL is NULL this function escapes in forms mode.  */
 static size_t
 escape_data (char *buffer, const void *data, size_t datalen,
              const char *special)
 {
   int forms = !special;
   const unsigned char *s;
   size_t n = 0;
 
   if (forms)
     special = "%;?&=";
 
   for (s = data; datalen; s++, datalen--)
     {
       if (forms && *s == ' ')
         {
 	  if (buffer)
 	    *buffer++ = '+';
 	  n++;
         }
       else if (forms && *s == '\n')
         {
 	  if (buffer)
 	    memcpy (buffer, "%0D%0A", 6);
 	  n += 6;
         }
       else if (forms && *s == '\r' && datalen > 1 && s[1] == '\n')
         {
 	  if (buffer)
 	    memcpy (buffer, "%0D%0A", 6);
 	  n += 6;
           s++;
           datalen--;
         }
       else if (strchr (VALID_URI_CHARS, *s) && !strchr (special, *s))
 	{
 	  if (buffer)
 	    *(unsigned char*)buffer++ = *s;
 	  n++;
 	}
       else
 	{
 	  if (buffer)
 	    {
 	      snprintf (buffer, 4, "%%%02X", *s);
 	      buffer += 3;
 	    }
 	  n += 3;
 	}
     }
   return n;
 }
 
 
 static int
 insert_escapes (char *buffer, const char *string,
 		const char *special)
 {
   return escape_data (buffer, string, strlen (string), special);
 }
 
 
 /* Allocate a new string from STRING using standard HTTP escaping as
  * well as escaping of characters given in SPECIALS.  A common pattern
  * for SPECIALS is "%;?&=". However it depends on the needs, for
  * example "+" and "/: often needs to be escaped too.  Returns NULL on
  * failure and sets ERRNO.  If SPECIAL is NULL a dedicated forms
  * encoding mode is used.  */
 static char *
 http_escape_string (const char *string, const char *specials)
 {
   int n;
   char *buf;
 
   n = insert_escapes (NULL, string, specials);
   buf = xmalloc (n+1);
   insert_escapes (buf, string, specials);
   buf[n] = 0;
   return buf;
 }
 
 
 /* Same as http_escape_string but with an explict length.  */
 static char *
 http_escape_buffer (const char *string, size_t length, const char *specials)
 {
   int n;
   char *buf;
 
   n = escape_data (NULL, string, length, specials);
   buf = xmalloc (n+1);
   escape_data (buf, string, length, specials);
   buf[n] = 0;
   return buf;
 }
 
 
 /* Percent-escape the string STR by replacing colons with '%3a'.   */
 static char *
 do_percent_escape (const char *str)
 {
   int i, j;
   char *ptr;
 
   for (i=j=0; str[i]; i++)
     if (str[i] == ':' || str[i] == '%' || str[i] == '\n')
       j++;
   ptr = xmalloc (i + 2 * j + 1);
   i = 0;
   while (*str)
     {
       if (*str == ':')
 	{
 	  ptr[i++] = '%';
 	  ptr[i++] = '3';
 	  ptr[i++] = 'a';
 	}
       else if (*str == '%')
 	{
 	  ptr[i++] = '%';
 	  ptr[i++] = '2';
 	  ptr[i++] = '5';
 	}
       else if (*str == '\n')
 	{
 	  /* The newline is problematic in a line-based format.  */
 	  ptr[i++] = '%';
 	  ptr[i++] = '0';
 	  ptr[i++] = 'a';
 	}
       else
 	ptr[i++] = *str;
       str++;
     }
   ptr[i] = '\0';
 
   return ptr;
 }
 
 
 /* Simple percent escape for colon based listings.  Returns a
  * statically allocated buffer.  */
 static char *
 percent_escape (const char *str)
 {
   static char *buffer;
 
   free (buffer);
   buffer = do_percent_escape (str);
   return buffer;
 }
 
 
 /* Escape STRING at a max length of N for use in HTML.  Returns a
  * statically allocated buffer.  */
 static const char *
 html_escape_n (const char *string, size_t length)
 {
   static char *buffer;
   char *p;
 
   /* The escaped string may be up to 6 times of STRING due to the
      expansion of '\"' to "&quot;".  */
   free (buffer);
   p = buffer = xmalloc (6 * length + 1);
   for (; *string && length; string++, length--)
     {
       switch (*string)
         {
         case '\"': p = my_stpcpy (p, "&quot;"); break;
         case '&':  p = my_stpcpy (p, "&amp;"); break;
         case '<':  p = my_stpcpy (p, "&lt;"); break;
         case '>':  p = my_stpcpy (p, "&gt;"); break;
         default:   *p++ = *string; break;
         }
     }
   *p = 0;
   return buffer;
 }
 
 
 /* Escape STRING for use in HTML.  Returns a statically allocated
  * buffer.  noet that this buffer is shared with the buffer returned
  * by html_escape_n.  */
 static const char *
 html_escape (const char *string)
 {
   return html_escape_n (string, strlen (string));
 }
 
 
 /* Escape STRING but insert <a> for one https link.  */
 static const char *
 html_escape_detect_link (const char *string)
 {
   const char *start, *s;
   char *part1, *url, *part2, *part3;
   size_t urllen;
   char *buffer, *p;
 
   start = strstr (string, "https://");
   if (!start || !start[8] || start[8] == ' ' || start[8] == '\t')
     return html_escape (string);
   if (!(start == string || start[-1] == ' ' || start[-1] == '\t'
         || start[-1] == '<'))
     return html_escape (string);
 
   urllen = 0;
   for (s = start; *s && *s != ' ' && *s != '\t' && *s != '>'; s++)
     urllen++;
 
   part1 = xstrdup (html_escape_n (string, start-string));
   url = http_escape_buffer (start, urllen, "\"");
   part2 = xstrdup (html_escape_n (start, urllen));
   part3 = xstrdup (html_escape (start + urllen));
 
   buffer = xmalloc (strlen (part1) + strlen (url)
                     + strlen (part2) + strlen (part3) + 100);
   p = my_stpcpy (buffer, part1);
   p = my_stpcpy (p, "<a href=\"");
   p = my_stpcpy (p, url);
   p = my_stpcpy (p, "\">");
   p = my_stpcpy (p, part2);
   p = my_stpcpy (p, "</a>");
   my_stpcpy (p, part3);
 
   free (part1);
   free (url);
   free (part2);
   free (part3);
   return buffer;
 }
 
 
 
 /* Escape STRING for use as a HREF attribute.  Returns a statically
  * allocated buffer.  */
 static const char *
 html_escape_href (const char *string)
 {
   static char *buffer;
 
   free (buffer);
   buffer = http_escape_string (string, "\"");
   return buffer;
 }
 
 
+/* Format T and return a statically allocated buffer.  */
+static const char *
+format_time_now (int human)
+{
+  static char buffer[40];
+  struct tm *tp;
+  time_t now;
+
+  time (&now);
+  tp = gmtime (&now);
+  if (!tp)
+    *buffer = 0;
+
+  if (human)
+    snprintf (buffer, sizeof buffer, "%04d-%02d-%02d",
+              1900 + tp->tm_year, tp->tm_mon+1, tp->tm_mday);
+  else
+    snprintf (buffer, sizeof buffer, "%04d%02d%02dT%02d%02d%02dZ",
+              1900 + tp->tm_year, tp->tm_mon+1, tp->tm_mday,
+              tp->tm_hour, tp->tm_min, tp->tm_sec);
+
+  return buffer;
+}
+
+
 /* Format T and return a statically allocated buffer.  */
 static const char *
 format_time (time_t t)
 {
   static char buffer[80];
   struct tm *tp;
 
   tp = gmtime (&t);
   if (!tp)
     *buffer = 0;
   else if (opt_html)
     snprintf (buffer, sizeof buffer,
               "<abbr title=\"%04d-%02d-%02d&nbsp;%02d:%02d:%02d UTC\">"
               "%04d-%02d-%02d</abbr>",
               1900 + tp->tm_year, tp->tm_mon+1, tp->tm_mday,
               tp->tm_hour, tp->tm_min, tp->tm_sec,
               1900 + tp->tm_year, tp->tm_mon+1, tp->tm_mday);
   else
     snprintf (buffer, sizeof buffer, "%04d%02d%02dT%02d%02d%02dZ",
               1900 + tp->tm_year, tp->tm_mon+1, tp->tm_mday,
               tp->tm_hour, tp->tm_min, tp->tm_sec);
   return buffer;
 }
 
 
 /* Format SIZE and return a statically allocated buffer.  */
 static const char *
 format_size (unsigned long long size)
 {
   static char buffer[80];
   const char *suffix;
   unsigned long long val = size;
 
   if (size < 1024)
     {
       val = size;
       suffix = "";
     }
   else if (size < 1024 * 1024)
     {
       val = size / 1024;
       suffix = "k";
     }
   else if (size < 1024 * 1024 * 1024)
     {
       val = size / (1024 * 1024);
       suffix = "M";
     }
   else
     {
       val = size / (1024 * 1024 * 1024);
       suffix = "G";
     }
 
   if (opt_html)
     snprintf (buffer, sizeof buffer,
               "<abbr title=\"%llu byte%s\">%llu%s</abbr>",
               size, size == 1? "":"s",
               val, suffix);
   else
     snprintf (buffer, sizeof buffer, "%llu%s", val, suffix);
 
   return buffer;
 }
 
 
 /* This function parses the first portion of the version number S and
  * stores it at NUMBER.  On success, this function returns a pointer
  * into S starting with the first character, which is not part of the
  * initial number portion; on failure, NULL is returned.  */
 static const char*
 parse_version_number (const char *s, int *number)
 {
   int val = 0;
 
   if (*s == '0' && digitp (s[1]))
     return NULL;  /* Leading zeros are not allowed.  */
   for (; digitp (*s); s++)
     {
       val *= 10;
       val += *s - '0';
     }
   *number = val;
   return val < 0 ? NULL : s;
 }
 
 
 /* This function breaks up the complete string-representation of the
  * version number S, which is of the following struture: <major
  * number>.<minor number>.<micro number><patch level>.  The major,
  * minor and micro number components will be stored in *MAJOR, *MINOR
  * and *MICRO.
  *
  * On success, the last component, the patch level, will be returned;
  * in failure, NULL will be returned.  */
 static const char *
 parse_version_string (const char *s, int *major, int *minor, int *micro)
 {
   s = parse_version_number (s, major);
   if (!s || *s != '.')
     return NULL;
   s++;
   s = parse_version_number (s, minor);
   if (!s || *s != '.')
     return NULL;
   s++;
   s = parse_version_number (s, micro);
   if (!s)
     return NULL;
   return s; /* patchlevel */
 }
 
 
 /* Compare function for version strings.  */
 static int
 compare_version_strings (const char *a, const char *b)
 {
   int a_major, a_minor, a_micro;
   int b_major, b_minor, b_micro;
   const char *a_plvl, *b_plvl;
 
   a_plvl = parse_version_string (a, &a_major, &a_minor, &a_micro);
   if (!a_plvl)
     a_major = a_minor = a_micro = 0;
 
   b_plvl = parse_version_string (b, &b_major, &b_minor, &b_micro);
   if (!b_plvl)
     b_major = b_minor = b_micro = 0;
 
   if (!a_plvl && !b_plvl)
     return -1;  /* Put invalid strings at the end.  */
   if (a_plvl && !b_plvl)
     return 1;
   if (!a_plvl && b_plvl)
     return -1;
 
   if (a_major > b_major)
     return 1;
   if (a_major < b_major)
     return -1;
 
   if (a_minor > b_minor)
     return 1;
   if (a_minor < b_minor)
     return -1;
 
   if (a_micro > b_micro)
     return 1;
   if (a_micro < b_micro)
     return -1;
 
   if (opt_reverse_ver && !opt_reverse)
     {
       /* We may only compare up to the next dot and the swicth back to
        * regular order.  */
       for (; *a_plvl && *b_plvl; a_plvl++, b_plvl++)
         {
           if (*a_plvl == '.' && *b_plvl == '.')
             return 0 - strcmp (a_plvl, b_plvl);
           else if (*a_plvl == '.')
             return 1;  /* B is larger but we need to reverse. */
           else if (*b_plvl == '.')
             return -1; /* A is larger but we need to reverse. */
           else if (*a_plvl != *b_plvl)
             break;
         }
       if (*a_plvl == *b_plvl)
         return 0;
       else
         return (*(signed char *)b_plvl - *(signed char *)a_plvl);
     }
   else
     return strcmp (a_plvl, b_plvl);
 }
 
 
 /* If string looks like a file name with a version nuymber, return a
  * pointer to the version number part; else return NULL.  */
 static const char *
 find_version_string (const char *s)
 {
   do
     {
       s = strchr (s, '-');
       if (!s)
         return NULL; /* Version string must be prefixed with a dash.  */
       s++;
     }
   while (!digitp (*s));
 
   return s;
 }
 
 
 /* Sort function for the directory listing.  */
 static int
 sort_finfo (const void *arg_a, const void *arg_b)
 {
   const finfo_t *a = arg_a;
   const finfo_t *b = arg_b;
   const char *astr, *bstr;
   const char *aver, *bver;
 
   if (opt_reverse)
     {
       astr = (*b)->name;
       bstr = (*a)->name;
     }
   else
     {
       astr = (*a)->name;
       bstr = (*b)->name;
     }
 
   aver = find_version_string (astr);
   bver = aver? find_version_string (bstr) : NULL;
 
   if (aver && bver
       && (aver - astr) == (bver - bstr)
       && !memcmp (astr, bstr, (aver - astr)))
     {
       if (opt_reverse_ver)
         return 0 - compare_version_strings (aver, bver);
       else
         return compare_version_strings (aver, bver);
     }
 
   return strcmp(astr, bstr);
 }
 
 
 /* Note: This function assumes that the CWD is the listed directory.  */
 static void
 print_header (const char *title)
 {
   const char *esc_title;
 
   if (!opt_html)
     return;
 
   esc_title = html_escape (title);
 
   if (opt_gpgweb)
     {
       FILE *readme;
       char line[256];
       char *p;
       int c;
 
       fputs ("<!--?xml version=\"1.0\" encoding=\"utf-8\"?-->\n"
              "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Strict//EN\""
              " \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd\">\n"
              "<html xmlns=\"http://www.w3.org/1999/xhtml\""
              " xml:lang=\"en\" lang=\"en\">\n", stdout);
       printf("<head>\n"
-             "<title>ftp.gnupg.org - %s</title>\n",
+             "<title>ftp.gnupg.org:%s</title>\n",
              esc_title);
       fputs ("<meta http-equiv=\"Content-Type\""
-             " content=\"text/html; charset=UTF-8\">\n"
+             " content=\"text/html; charset=UTF-8\"/>\n", stdout);
+      printf("<meta name=\"date\" content=\"%s\"/>\n", format_time_now (0));
+      fputs ("<meta name=\"generator\" content=\""PGMNAME" v"VERSION"\"/>\n"
              "<meta name=\"viewport\""
-             " content=\"width=device-width, initial-scale=1\">\n"
+             " content=\"width=device-width, initial-scale=1\"/>\n"
              "<link rel=\"stylesheet\" href=\"/share/site.css\""
-             " type=\"text/css\">\n"
+             " type=\"text/css\"/>\n"
              "</head>\n", stdout);
 
       fputs ("<body>\n"
              "<div id=\"wrapper\">\n"
              "<div id=\"header\"><a href=\"https://gnupg.org/index.html\""
              " class=\"logo\">"
              "<img src=\"/share/logo-gnupg-light-purple-bg.png\"></a>&nbsp;"
              "</div>\n", stdout);
 
       printf("<main>\n"
              "<div id=\"content\">\n"
-             "<h2>%s</h2>\n"
+             "<h2>ftp.gnupg.org:%s</h2>\n"
              "<div class=\"outline-text-2\" id=\"text-1\">\n",
              esc_title);
 
       readme = fopen ("README", "r");
       if (opt_readme && (readme = fopen ("README", "r")))
         {
           fputs ("<pre class=\"ftp-readme\">\n", stdout);
           while (fgets (line, sizeof line, readme))
             {
               int no_lf = 0;
               /* Eat up the rest of an incomplete line.  */
               if (!*line)
                 no_lf = 1;
               else if (line[strlen (line)-1] != '\n')
                 {
                   no_lf = 1;
                   while ((c = getc (readme)) != EOF && c != '\n')
                     ;
                 }
 
               /* Replace empty lines with a leading doc by an empty
                * line.  These lines are used on FTP servers to avoid
                * problems with broken FTP cleints.  */
               if (*line == '.')
                 {
                   for (p=line+1; (*p == ' ' || *p == '\t' || *p == '\n'); p++)
                     ;
                   if (!*p)
                     {
                       putchar ('\n');
                       *line = 0;
                     }
                 }
 
               if (*line)
                 fputs (html_escape_detect_link (line), stdout);
               if (no_lf)
                 putchar ('\n');
             }
           fputs ("</pre>\n", stdout);
           fclose (readme);
         }
       fputs ("</div>\n", stdout);
 
    }
   else
     {
       printf ("<html>\n"
               "<head>\n"
               "<title>Index of %s</title>\n"
               "</head>\n"
               "<body bgcolor=\"#ffffff\">\n"
               "<h2>Index of %s</h2>\n"
-              "<table>\n",
+              "<table class=\"ftp\">\n",
               esc_title, esc_title);
     }
 }
 
 
 static void
 print_footer (void)
 {
   if (!opt_html)
     return;
 
   if (opt_gpgweb)
     {
       fputs ("</div><!-- end content -->\n"
              "</main>\n"
              "<div id=\"footer\">\n", stdout);
       fputs ("<div id=\"nav_bottom\">\n"
              "<ul>\n"
              "<li><a href=\"/privacy-policy.html\">Privacy&nbsp;Policy</a>"
              "</li>\n"
              "<li><a href=\"/imprint.html\">Imprint</a>"
              "</li>\n"
-             "<li><a href=\"/misc/index.html\">Archive</a>"
-             "</li>\n"
-             "<li><a href=\"/sitemap.html\">Sitemap</a>"
-             "</li>\n"
              "<li><a href=\"/blog/index.html\">Blog</a>"
              "</li>\n"
+             "<li><a href=\"/index.html\">Web</a>"
+             "</li>\n"
              "</ul>\n"
              "</div>\n", stdout);
 
       fputs ("<div class=\"footerbox\">\n"
              "<a><img src=\"/share/traueranzeige-g10_v2015.png\""
              " alt=\"Traueranzeige: Wir nehmen Abschied von einem"
              " sicher geglaubten Freund, dem | Fernmeldegeheimniss"
              " | (Artikel 10 Grundgesetz) | * 23. Mai 1949,"
              " + 18. Dezember 2015\" title=\"Article 10 of the"
              " German constitution (communication privacy) is"
              " not anymore with us.\" height=\"73px\" width=\"200px\"></a>\n"
              "<p></p>\n"
              "</div>\n", stdout);
 
       fputs ("<div id=\"cpyright\">\n"
              "<a rel=\"license\""
              " href=\"https://creativecommons.org/licenses/by-sa/4.0/\">"
              "<img alt=\"CC BY-SA 4.0\" style=\"border: 0\""
              " src=\"/share/cc-by-sa_80x15.png\"></a>&nbsp;"
              "This web page is Copyright 2017 GnuPG e.V. and"
              " licensed under a <a rel=\"license\""
              " href=\"https://creativecommons.org/licenses/by-sa/4.0/\">"
              "Creative Commons Attribution-ShareAlike 4.0 International"
              " License</a>.  See <a href=\"https://gnupg.org/copying.html\">"
              "copying</a> for details.\n", stdout);
-      printf("Page last updated on 2017-xx-xx.\n");
+      printf("Page last updated on %s.\n", format_time_now (1));
       fputs ("</div>\n"
              "</div>\n"
              "</div><!-- end wrapper -->\n"
              "</body>\n"
              "</html>\n", stdout);
     }
   else
     {
       printf ("</table>\n"
               "</body>\n"
               "</html>\n");
     }
 }
 
 
 /* Print COUNT directories from the array SORTED. */
 static void
-print_dirs (finfo_t *sorted, int count, const char *directory)
+print_dirs (finfo_t *sorted, int count, int at_root)
 {
   int idx;
   finfo_t fi;
   int any = 0;
 
   for (idx=0; idx < count; idx++)
     {
       fi = sorted[idx];
       if (!fi->is_dir)
         continue;
 
       if (!any && opt_html)
         {
           any = 1;
 
           if (opt_gpgweb)
             {
               fputs ("<h3>Directories</h3>\n"
                      "<div class=\"outline-text-3\">\n"
-                     "<table>\n", stdout);
+                     "<table class=\"ftp\">\n", stdout);
 
-              if (strcmp (directory, "/"))
+              if (!at_root)
                 fputs ("<tr><td><img src=\"/share/up.png\""
                        " width=\"22\" height=\"22\"/></td>"
                        "<td><a href=\"../\">"
                        "Parent Directory</a></td></tr>\n", stdout);
             }
           else
             {
               fputs ("<tr><td>&nbsp</td>"
                      "<td colspan=3><h3>Directories</h3></td></tr>\n",
                      stdout);
-              if (strcmp (directory, "/"))
+              if (!at_root)
                 fputs ("<tr><td><a href=\"../\">"
                        "Parent Directory</a></td></tr>\n", stdout);
             }
         }
 
       if (opt_gpgweb)
         printf ("<tr><td><img src=\"/share/folder.png\""
                 " width=\"22\" height=\"22\"/></td>"
                 "<td><a href=\"%s\">%s</a></td></tr>\n",
                 html_escape_href (fi->name), html_escape (fi->name));
       else if (opt_html)
         printf ("<tr><td width=\"40%%\"><a href=\"%s\">%s</a></td></tr>\n",
                 html_escape_href (fi->name), html_escape (fi->name));
       else
         printf ("D %s\n", fi->name);
     }
 
   if (any && opt_gpgweb)
     {
       fputs ("</table>\n"
              "</div>\n\n", stdout);
+    }
+  else if (opt_gpgweb && !at_root)
+    {
+      /* !any - need to print an UP link */
+      fputs ("<div class=\"outline-text-3\">\n"
+             "<table class=\"ftp\">\n"
+             "<tr><td><img src=\"/share/up.png\""
+             " width=\"22\" height=\"22\"/></td>"
+             "<td><a href=\"../\">"
+             "Parent Directory</a></td></tr>\n"
+             "</table>\n"
+             "</div>\n", stdout);
+
+
     }
 }
 
 
 /* Print COUNT files from the array SORTED. */
 static void
 print_files (finfo_t *sorted, int count)
 {
   int idx;
   finfo_t fi;
   int any = 0;
 
   for (idx=0; idx < count; idx++)
     {
       fi = sorted[idx];
       if (!fi->is_reg)
         continue;
 
       if (!any && opt_html)
         {
           any = 1;
           if (opt_gpgweb)
             {
               fputs ("<h3>Files</h3>\n"
                      "<div class=\"outline-text-3\">\n"
-                     "<table>\n", stdout);
+                     "<table class=\"ftp\">\n", stdout);
             }
           else
             fputs ("<tr><td colspan=3><h3>Files</h3></td></tr>\n",
                    stdout);
 
         }
 
       if (opt_gpgweb)
         printf ("<tr><td><img src=\"/share/%s.png\""
                 " width=\"22\" height=\"22\"/></td>"
                 "<td><a href=\"%s\">%s</a></td>"
                 "<td align=\"right\">%s</td><td align=\"right\">%s</td></tr>\n",
                 strstr (fi->name, ".sig")? "document":
                 strstr (fi->name, ".tar")? "tar" : "document",
                 html_escape_href (fi->name), html_escape (fi->name),
                 format_time (fi->mtime), format_size (fi->size));
       else if (opt_html)
         printf ("<tr><td width=\"50%%\"><a href=\"%s\">%s</a></td>"
                 "<td align=\"right\">%s</td><td align=\"right\">%s</td></tr>\n",
                 html_escape_href (fi->name), html_escape (fi->name),
                 format_time (fi->mtime), format_size (fi->size));
       else
         printf ("F %s\n", fi->name);
     }
 
   if (any && opt_gpgweb)
     {
       fputs ("</table>\n"
              "</div>\n\n", stdout);
     }
 }
 
 
 /* Scan DIRECTORY and print an index.
  * FIXME: This does a chdir and does not preserve the old PWD.
  *        The fix is to build the full filename beofre stat'ing.
  */
 static void
 scan_directory (const char *directory, const char *title)
 {
   DIR *dir;
   struct dirent *dentry;
   finfo_t fi;
   finfo_t finfo = NULL;
   finfo_t *sorted;
   int count = 0;
   int idx;
   size_t len;
   strlist_t sl;
+  int at_root = 0;
+
+  if (opt_gpgweb)
+    {
+      if (!strcmp (title, "/"))
+        at_root = 1;
+    }
+  else if (!strcmp (directory, "/"))
+    at_root = 1;
 
   dir = opendir (directory);
   if (!dir)
     {
       err ("can't open directory '%s': %s\n", directory, strerror (errno));
       return;
     }
 
   while (errno=0,(dentry = readdir (dir)))
     {
       if (*dentry->d_name == '.')
         continue;  /* Skip self, parent, and hidden directories.  */
       len = strlen (dentry->d_name);
       if (!len)
         continue;  /* Empty filenames should actually not exist.  */
       if (dentry->d_name[len-1] == '~')
         continue;  /* Skip backup files.  */
       for (sl = opt_exclude; sl; sl = sl->next)
         if (!strcmp (sl->d, dentry->d_name))
           break;
       if (sl)
         continue; /* Skip excluded names.  */
       fi = xcalloc (1, sizeof *fi + strlen (dentry->d_name));
       strcpy (fi->name, dentry->d_name);
       fi->next = finfo;
       finfo = fi;
       count++;
     }
   if (errno)
     die ("error reading directory '%s': %s\n", directory, strerror (errno));
   closedir (dir);
 
   sorted = xcalloc (count, sizeof *sorted);
   for (fi=finfo, idx=0; fi; fi = fi->next)
     sorted[idx++] = fi;
 
   inf ("directory '%s' has %d files\n", directory, count);
   qsort (sorted, count, sizeof *sorted, sort_finfo);
 
   if (chdir (directory))
     die ("cannot chdir to '%s': %s\n", directory, strerror (errno));
 
   for (idx=0; idx < count; idx++)
     {
       struct stat sb;
 
       fi = sorted[idx];
       if (stat (fi->name, &sb))
         {
           err ("cannot stat '%s': %s\n", fi->name, strerror (errno));
           continue;
         }
 
       fi->is_dir = !!S_ISDIR(sb.st_mode);
       fi->is_reg = !!S_ISREG(sb.st_mode);
       fi->size = fi->is_reg? sb.st_size : 0;
       fi->mtime = sb.st_mtime;
     }
 
   print_header (title);
   if (opt_files_first)
     {
       print_files (sorted, count);
-      print_dirs (sorted, count, directory);
+      print_dirs (sorted, count, at_root);
     }
   else
     {
-      print_dirs (sorted, count, directory);
+      print_dirs (sorted, count, at_root);
       print_files (sorted, count);
     }
   print_footer ();
 
   /* We create the index file in the current directory.  */
   if (opt_index)
     {
-      FILE *indexfp = fopen (".ftp-index", "w");
+      FILE *indexfp = fopen (opt_index, "w");
       if (!indexfp)
-        die ("error creating .ftp-index in '%s': %s\n",
-             directory, strerror (errno));
+        die ("error creating '%s' for '%s': %s\n",
+             opt_index, directory, strerror (errno));
 
       for (idx=0; idx < count; idx++)
         {
           fi = sorted[idx];
           fprintf (indexfp, "%s:%c:%llu:%lu:\n",
                    percent_escape (fi->name),
                    fi->is_dir? 'd':
                    fi->is_reg? 'r': '?',
                    fi->size,
                    (unsigned long)fi->mtime);
         }
       if (ferror (indexfp))
-        die ("error writing .ftp-index in '%s': %s\n",
-             directory, strerror (errno));
+        die ("error writing '%s' for '%s': %s\n",
+             opt_index, directory, strerror (errno));
       /* Fixme: Check for close errors.  */
       fclose (indexfp);
     }
 
   free (sorted);
   while ((fi = finfo))
     {
       fi = finfo->next;
       free (finfo);
       finfo = fi;
     }
 }
 
 
 int
 main (int argc, char **argv)
 {
   int last_argc = -1;
   strlist_t sl;
 
   if (argc < 1)
     die ("Hey, read up on how to use exec(2)\n");
   argv++; argc--;
 
   while (argc && last_argc != argc )
     {
       last_argc = argc;
       if (!strcmp (*argv, "--"))
         {
           argc--; argv++;
           break;
         }
+      else if (!strcmp (*argv, "--version"))
+        {
+          fputs (PGMNAME " " VERSION "\n"
+                 "Copyright (C) 2017 g10 Code GmbH\n"
+                 "License GPLv3+: GNU GPL version 3 or later"
+                 " <https://gnu.org/licenses/gpl.html>\n"
+                 "This is free software: you are free to change"
+                 " and redistribute it.\n"
+                 "There is NO WARRANTY, to the extent permitted by law.\n",
+                 stdout);
+          exit (0);
+        }
       else if (!strcmp (*argv, "--help"))
         {
           fputs ("usage: " PGMNAME " [options] directory [title]\n"
                  "Print an index for an FTP directory.\n\n"
                  "Options:\n"
+                 "  --version       print program version\n"
                  "  --verbose       verbose diagnostics\n"
                  "  --debug         flyswatter\n"
                  "  --reverse       reverse sort order\n"
                  "  --reverse-ver   reverse only the version number order\n"
                  "  --files-first   print files before directories\n"
                  "  --html          output HTML\n"
                  "  --gpgweb        output HTML as used at gnupg.org\n"
                  "  --readme        include README file\n"
-                 "  --index         create an .ftp-index file\n"
+                 "  --index FILE    create index FILE\n"
                  "  --exclude NAME  ignore file NAME\n"
                  , stdout);
           exit (0);
         }
       else if (!strcmp (*argv, "--verbose"))
         {
           opt_verbose++;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--debug"))
         {
           opt_debug++;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--reverse"))
         {
           opt_reverse = 1;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--reverse-ver"))
         {
           opt_reverse_ver = 1;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--files-first"))
         {
           opt_files_first = 1;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--readme"))
         {
           opt_readme = 1;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--html"))
         {
           opt_html = 1;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--index"))
         {
-          opt_index = 1;
+          argc--; argv++;
+          if (!argc || !**argv)
+            die ("argument missing for option '%s'\n", argv[-1]);
+          opt_index = *argv;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--gpgweb"))
         {
           opt_gpgweb = opt_html = 1;
           argc--; argv++;
         }
       else if (!strcmp (*argv, "--exclude"))
         {
           argc--; argv++;
           if (!argc || !**argv)
             die ("argument missing for option '%s'\n", argv[-1]);
           sl = xmalloc (sizeof *sl + strlen (*argv));
           strcpy (sl->d, *argv);
           sl->next = opt_exclude;
           opt_exclude = sl;
           argc--; argv++;
         }
       else if (!strncmp (*argv, "--", 2))
         die ("unknown option '%s' (use --help)\n", *argv);
     }
 
   if (argc < 1 || argc > 2)
     die ("usage: " PGMNAME " [options] directory [title]\n");
 
 
   scan_directory (argv[0], argv[1]? argv[1]:argv[0]);
 
 
   return 0;
 }
 
 /*
 Local Variables:
 compile-command: "cc -Wall -g -o ftp-indexer ftp-indexer.c"
 End:
 */
diff --git a/web/share/gpgweb.el b/web/share/gpgweb.el
index 9a719f2..6539206 100644
--- a/web/share/gpgweb.el
+++ b/web/share/gpgweb.el
@@ -1,524 +1,527 @@
 ;;; gpgweb.el --- elisp helper code for the GnuPG web pages
 
 (if (< (string-to-number emacs-version) 24)
     (require 'org-exp))
 
 ;; makeindex disabled because the generated file is created in the
 ;; source directory.
 (defun gpgweb-setup-project ()
   "Set up an org-publish project for the gnupg.org website."
   (progn
    (require 'ox-gpgweb (concat gpgweb-root-dir "share/ox-gpgweb.el"))
    (aput 'org-publish-project-alist "gpgweb-org"
    `(:base-directory ,gpgweb-root-dir
      :base-extension "org"
      :language "en"
      :html-extension "html"
      :recursive t
      :publishing-directory ,gpgweb-stage-dir
      :publishing-function gpgweb-org-to-html
      :body-only t
      :section-numbers nil
      :tags nil
      :with-toc nil
      :makeindex nil
      :auto-sitemap nil
      :sitemap-title "GnuPG - Sitemap"
      :sitemap-sort-folders "last"
      :sitemap-file-entry-format "%t  @@html:<span id=\"smallnote\">@@(%d)@@html:</span>@@"
      :style-include-default nil
      :timestamp-file nil
      :html-head "<link rel=\"stylesheet\" href=\"gnupg.css\" type=\"text/css\" />"
      :html-head-include-scripts nil))
 
    (aput 'org-publish-project-alist "gpgweb-other"
    `(:base-directory ,gpgweb-root-dir
      :base-extension "jpg\\|png\\|css\\|txt\\|rss\\|lst\\|sig\\|js\\|map\\|eot\\|ttf\\|woff\\|woff2\\|svg"
      :recursive t
      :publishing-directory ,gpgweb-stage-dir
      :publishing-function org-publish-attachment
      :completion-function gpgweb-upload))
 
    (aput 'org-publish-project-alist "gpgweb"
    '(:components ("gpgweb-org" "gpgweb-other")))
 
    (add-hook 'org-export-before-processing-hook 'gpgweb-preprocess)))
 
 
 (defun gpgweb-preprocess (backend)
   "Insert certain stuff before processing."
   (let ()
     (goto-char (point-min))
     (when (re-search-forward
            "^#\\+GPGWEB-NEED-SWDB\\b" 2048 t)
       (beginning-of-line)
       (kill-line 1)
       (insert (org-file-contents (concat gpgweb-root-dir "swdb.mac")
                                  'noerror)))))
 
 
 (defun gpgweb-insert-header (title committed-at custom)
   "Insert the header.
 
 COMMITTED-AT is the commit date string of the source file or nil
 if not available.  If CUSTOM is true only a minimal header is set."
   (goto-char (point-min))
   (insert "<?xml version=\"1.0\" encoding=\"utf-8\"?>
 <!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Strict//EN\"
                \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd\">
 <html xmlns=\"http://www.w3.org/1999/xhtml\" lang=\"en\" xml:lang=\"en\">
 <head>
 <title>" title "</title>
 <meta http-equiv=\"Content-Type\" content=\"text/html;charset=utf-8\" />
 ")
   (when (and committed-at (>= (length committed-at) 10))
       (insert "<meta name=\"DC.Date\" content=\""
               (substring committed-at 0 10) "\" />\n"))
   (insert "<meta name=\"DC.Language\" content=\"en\" />
 <meta name=\"DC.Title\" content=\"" title "\" />
 <meta name=\"DC.Description\"
  content=\"GnuPG is a free implementation of OpenPGP\" />
 <meta name=\"DC.Creator\" content=\"The People of the GnuPG Project\" />
 <meta name=\"DC.Publisher\" content=\"The GnuPG Project\" />
 <meta name=\"DC.Identifier\" content=\"https://gnupg.org/\" />
 <meta name=\"DC.Rights\" content=\"https://gnupg.org/copying.html\" />
 <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\" />
 ")
 (unless custom
  (insert "<link rel=\"stylesheet\" href=\"/share/site.css\" type=\"text/css\" />
 </head>
 <body>
 ")))
 
 (defconst gpgweb-gnupg-menu-alist
   '(("/index.html"
      "Home"
      (("/index.html"                       "Home")
       ("/news.html"                        "News")
       ("/people/index.html"                "People")
       ("/verein/index.html"                "Verein")
       ("/documentation/sites.html"         "Sites")))
     ("/donate/index.html"
      "Donate"
      (("/donate/index.html"                "Donate")
       ("/donate/kudos.html"                "List of Donors")))
     ("/software/index.html"
      "Software"
      (("/software/index.html"              "GnuPG")
       ("/software/frontends.html"          "Frontends")
       ("/software/tools.html"              "Tools")
       ("/software/libraries.html"          "Libraries")
       ("/software/swlist.html"             "All")))
     ("/download/index.html"
      "Download"
      (("/download/index.html"              "Download")
       ("/download/integrity_check.html"    "Integrity&nbsp;Check")
       ("/download/supported_systems.html"  "Supported&nbsp;Systems")
       ("/download/release_notes.html"      "Release&nbsp;Notes")
       ("/download/mirrors.html"            "Mirrors")
       ("/download/git.html"                "GIT")))
     ("/documentation/index.html"
      "Documentation"
      (("/documentation/howtos.html"        "HOWTOs")
       ("/documentation/manuals.html"       "Manuals")
       ("/documentation/guides.html"        "Guides")
       ("/documentation/faqs.html"          "FAQs")
       ("/documentation/mailing-lists.html" "Mailing&nbsp;Lists")
       ("/service.html"                     "3rd Party Support")
       ("/documentation/bts.html"           "Bug&nbsp;Tracker")
       ("/documentation/security.html"      "Security")))
     ("/blog/index.html"
      "Blog"))
   "The definition of the gnupg.org menu structure.")
 
 (defconst gpgweb-gnupg-bottom-menu-alist
   '(("/privacy-policy.html"
      "Privacy&nbsp;Policy"
      ())
     ("/imprint.html"
      "Imprint"
      ())
     ("/misc/index.html"
      "Archive"
      ())
     ("/sitemap.html"
      "Sitemap"
      ())
     ("/blog/index.html"
      "Blog"
+     ())
+    ("/ftp/index.html"
+     "Files"
      ()))
   "The definition of the gnupg.org bottom menu structure.")
 
 
 (defun gpgweb--any-selected-menu-p (menu selected-file)
   "Return t if any item in MENU has been selected."
   (let ((item (car menu))
         res)
     (when menu
       (when item
         (when (string= (car item) selected-file)
             (setq res t))
         (when (caddr item)
           (when (gpgweb--any-selected-menu-p (caddr item) selected-file)
             (setq res t))))
       (when (gpgweb--any-selected-menu-p (cdr menu) selected-file)
         (setq res t)))
     res))
 
 
 (defun gpgweb--selected-top-menu (menu selected-file)
   "Return the selected top menu or nil."
   (when menu
     (let ((item (car menu)))
       (if (and item
                (or (string= (car item) selected-file)
                    (gpgweb--any-selected-menu-p (caddr item) selected-file)))
           menu
         (gpgweb--selected-top-menu (cdr menu) selected-file)))))
 
 (defun gpgweb--insert-menu (menu lvl selected-file)
   "Helper function to insert the menu."
   (when menu
     (let ((item (car menu)))
       (when item
         (dotimes (i (1+ lvl)) (insert "  "))
         (insert "  <li><a href=\"" (car item) "\"")
         (when (or (string= (car item) selected-file)
                   (gpgweb--any-selected-menu-p (caddr item) selected-file))
           (insert " class=\"selected\""))
         (insert  ">" (cadr item) "</a>\n")
         (when (caddr item)
           (dotimes (i (1+ lvl)) (insert "  "))
           (insert "  <ul class=\"sub-menu\">\n")
           (gpgweb--insert-menu (caddr item) (1+ lvl) selected-file)
           (dotimes (i (1+ lvl)) (insert "  "))
           (insert "  </ul>\n"))
         (dotimes (i (1+ lvl)) (insert "  "))
         (insert "  </li>\n")))
     (gpgweb--insert-menu (cdr menu) lvl selected-file)))
 
 
 (defun gpgweb--insert-submenu (menu selected-file)
    "Helper function to insert the sub-menu."
    (when menu
      (let ((item (car menu)))
        (when item
          (insert "    <li><a href=\"" (car item) "\"")
          (when (or (string= (car item) selected-file)
                    (gpgweb--any-selected-menu-p (caddr item) selected-file))
            (insert " class=\"selected\""))
          (insert ">" (cadr item) "</a></li>\n")))
      (gpgweb--insert-submenu (cdr menu) selected-file)))
 
 
 (defun gpgweb-insert-menu (selected-file)
   "Insert the menu structure into the HTML file."
   (goto-char (point-min))
   (when (re-search-forward "^<body>\n" nil t)
     (insert "<div id=\"wrapper\">
 <div id=\"header\"><a href=\"/index.html\" class=\"logo\"
      ><img src=\"/share/logo-gnupg-light-purple-bg.png\"></a>&nbsp;</div>
 <nav>
   <ul>
 ")
     (gpgweb--insert-menu gpgweb-gnupg-menu-alist 0 selected-file)
     (insert "  </ul>
 </nav>
 ")
     (let ((m (caddr (car (gpgweb--selected-top-menu
                           gpgweb-gnupg-menu-alist selected-file)))))
       (when m
           (insert "<nav class=\"subnav\">\n  <ul>\n")
           (gpgweb--insert-submenu m selected-file)
           (insert "  </ul>\n</nav>\n")))
     (insert "<main>
 <div id=\"content\">
 ")))
 
 
 (defun gpgweb-blog-index (orgfile filelist)
   "Return the index of ORGFILE in FILELIST or nil if not found."
   (let (found
         (i 0))
     (while (and filelist (not found))
       (if (string= orgfile (car filelist))
           (setq found i))
       (setq i (1+ i))
       (setq filelist (cdr filelist)))
     found))
 
 (defun gpgweb-blog-prev (fileidx filelist)
   "Return the chronological previous file at FILEIDX from FILELIST
 with the suffixed replaced by \"html\"."
   (if (> fileidx 1)
       (concat (file-name-sans-extension (nth (1- fileidx) filelist)) ".html")))
 
 (defun gpgweb-blog-next (orgfile filelist)
   "Return the chronological next file at FILEIDX from FILELIST
 with the suffixed replaced by \"html\"."
   (if (< fileidx (1- (length filelist)))
       (concat (file-name-sans-extension (nth (1+ fileidx) filelist)) ".html")))
 
 (defun gpgweb-fixup-blog (info orgfile filelist)
   "Insert the blog specific content.  INFO is the usual
 plist. ORGFILE is the name of the current source file without the
 directory part.  If FILELIST is a list it has an ordered list of
 org filenames."
   (let ((authorstr (car (plist-get info :author)))
         (datestr   (car (plist-get info :date))))
     (goto-char (point-min))
     (if (re-search-forward "^<main>" nil t)
         (let* ((indexp (string= orgfile "index.org"))
                (fileidx (if (listp filelist)
                             (if indexp
                                 (1- (length filelist))
                               (gpgweb-blog-index orgfile filelist))))
                (prevfile (if fileidx
                              (gpgweb-blog-prev fileidx filelist)))
                (nextfile (if (and fileidx (not indexp))
                            (gpgweb-blog-next fileidx filelist))))
           (move-beginning-of-line nil)
           (insert "<nav class=\"subnav\">\n  <ul>\n")
           (if prevfile
               (insert "    <li><a href=\"" prevfile "\">Previous</a></li>\n"))
           (insert
            "    <li><a href=\"/blog/index.html#blogindex\">Index</a></li>\n")
           (if nextfile
               (insert "    <li><a href=\"" nextfile "\">Next</a></li>\n"))
           (insert "  </ul>\n</nav>\n")))
     (if (and datestr authorstr)
         (if (re-search-forward "^<h2 id=.*\n" nil t)
             (insert "<p class=\"postdate\">Posted "
                     datestr
                     " by "
                     authorstr
                     "</p>\n")))))
 
 
 (defun gpgweb-insert-footer (htmlfile committed-at blogmode)
   "Insert the footer.
 
 HTMLFILE is HTML file name and COMMITTED-AT is the commit date
 string of the source file or nil if not available."
   (let ((srcfile (concat "https://git.gnupg.org/cgi-bin/gitweb.cgi?"
                          "p=gnupg-doc.git;a=blob;f="
                          (if blogmode "misc/blog.gnupg.org" "web/")
                          ;; The replace below is a hack to cope with
                          ;; blogmode where HTMLFILE is like "./foo.html".
                          (replace-regexp-in-string
                           "^\\./" "/"
                           (file-name-sans-extension htmlfile) t)
                          ".org"))
         (changed (if (and committed-at (>= (length committed-at) 10))
                      (substring committed-at 0 10)
                      "[unknown]")))
     (goto-char (point-max))
     (insert "</div><!-- end content -->
 </main>
 <div id=\"footer\">
   <div id=\"nav_bottom\">
   <ul>
 ")
     (gpgweb--insert-menu gpgweb-gnupg-bottom-menu-alist 0 nil)
     (insert "    </ul>
   </div>
 ")
     (insert "  <div class=\"footerbox\">
   <a><img src=\"/share/traueranzeige-g10_v2015.png\"
           width=\"200px\" height=\"73px\"
           alt=\"Traueranzeige: Wir nehmen Abschied von einem sicher geglaubten Freund, dem | Fernmeldegeheimniss | (Artikel 10 Grundgesetz) | * 23. Mai 1949, + 18. Dezember 2015\"
           title=\"Article 10 of the German constitution (communication privacy) is not anymore with us.\" /></a>
   <p></p>
   </div>
 ")
     (goto-char (point-min))
     (unless (search-forward "<!--disable-copyright-footer-->" nil t)
       (goto-char (point-max))
       (if (string-prefix-p "verein/" htmlfile)
           (insert "  <div id=\"cpyright\">
     <a rel=\"license\" href=\"https://creativecommons.org/licenses/by-sa/4.0/\"
       ><img alt=\"CC BY-SA 4.0\" style=\"border: 0\"
             src=\"/share/cc-by-sa_80x15.png\"/></a>&nbsp;
     This web page is
     Copyright 2017 GnuPG e.V. and licensed under a
     <a rel=\"license\" href=\"https://creativecommons.org/licenses/by-sa/4.0/\"
     >Creative Commons Attribution-ShareAlike 4.0 International License</a>.  See
     <a href=\"/copying.html\">copying</a> for details.
     Page <a href=\"" srcfile "\">source</a> last changed on " changed ".
   </div>\n")
           (insert "  <div id=\"cpyright\">
     <a rel=\"license\" href=\"https://creativecommons.org/licenses/by-sa/3.0/\"
       ><img alt=\"CC BY-SA 3.0\" style=\"border: 0\"
             src=\"/share/cc-by-sa_80x15.png\"/></a>&nbsp;
     These web pages are
     Copyright 1998--2017 The GnuPG Project and licensed under a
     <a rel=\"license\" href=\"https://creativecommons.org/licenses/by-sa/3.0/\"
     >Creative Commons Attribution-ShareAlike 3.0 Unported License</a>.  See
     <a href=\"/copying.html\">copying</a> for details.
     Page <a href=\"" srcfile "\">source</a> last changed on " changed ".
   </div>\n")))
   (goto-char (point-max))
   (insert "</div>
 </div><!-- end wrapper -->
 </body>
 </html>")))
 
 
 (defun gpgweb-publish-find-title (file &optional reset)
   "Find the title of FILE in project.
 This is a copy of org-publish-find-title which switches the
 buffer into read-write mode so that it works with read-only files."
   (or
    (and (not reset) (org-publish-cache-get-file-property file :title nil t))
    (let* ((org-inhibit-startup t)
 	  (visiting (find-buffer-visiting file))
 	  (buffer (or visiting (find-file-noselect file))))
      (with-current-buffer buffer
        (toggle-read-only 0)
        (let ((title
 	      (let ((property
 		     (plist-get
 		      ;; protect local variables in open buffers
 		      (if visiting
 			  (org-export-with-buffer-copy (org-export-get-environment))
 			(org-export-get-environment))
 		      :title)))
 		(if property
 		    (org-no-properties (org-element-interpret-data property))
 		  (file-name-nondirectory (file-name-sans-extension file))))))
 	 (unless visiting (kill-buffer buffer))
 	 (org-publish-cache-set-file-property file :title title)
 	 title)))))
 
 
 (defun gpgweb-want-custom-page-p ()
   "Return true if the current buffer indicated that it wants to
 be a custom page."
   (let ((savepoint (point))
         (result))
     (goto-char (point-min))
     (setq result (not (not (search-forward "<!--custom-page-->" nil t))))
     (goto-char savepoint)
     result))
 
 
 (defun gpgweb-postprocess-html (plist orgfile htmlfile blogmode)
   "Post-process the generated HTML file
 
   - Insert header and footer
   - Insert \"class=selected\" into the active menu entry
   - Fixup sitemap.
 
 If blogmode is not nil the output is rendered as a blog.  BLOGMODE
 may then contain an ordered list of org file names which are used
 to create the previous and Next links for an entry."
   (let* ((visitingp (find-buffer-visiting htmlfile))
 	 (work-buffer (or visitingp (find-file-noselect htmlfile)))
          (committed-at (shell-command-to-string
                         (concat "git"
                                 (if blogmode (concat " -C " gpgweb-blog-dir))
                                 " log -1 --format='%ci' -- " orgfile))))
     (prog1 (with-current-buffer work-buffer
              (let ((fname (file-name-nondirectory htmlfile))
                    (fname-2 (replace-regexp-in-string
                              ".*/gnupg-doc-stage/web/\\(.*\\)$" "\\1"
                              htmlfile t))
                    (title (gpgweb-publish-find-title orgfile))
                    (custom (gpgweb-want-custom-page-p)))
                ;; Insert header, menu, and footer.
                (gpgweb-insert-header title committed-at custom)
                (unless custom
                  (goto-char (point-min))
                  (unless (search-forward "<!--disable-menu-->" nil t)
                    (gpgweb-insert-menu fname-2))
                  (if blogmode
                      (gpgweb-fixup-blog plist
                                         (file-name-nondirectory orgfile)
                                         blogmode))
                  (gpgweb-insert-footer fname-2 committed-at blogmode))
 
                ; Fixup the sitemap
                (when (string-equal fname "sitemap.html")
                  (goto-char (point-min))
                  (while (re-search-forward
                          "^.*<li>.*>\\(GnuPG - \\).*<span.*$" nil t)
                    (replace-match "" t nil nil 1)))
 
                ; Due to a problem with the current org exporter (cases
                ; were we link to file mapped via a webserver alias) we
                ; have to use a full URL at some places in the org
                ; source.  We fix that up here.
                (goto-char (point-min))
                (while (re-search-forward
                        "href=\"\\(https://www.gnupg.org\\)/.*\"" nil t)
                  (replace-match "" t t nil 1))
 
                ; If the wideright flag is used, change <td> and <th>
                ; attributes.
                (goto-char (point-min))
                (when (search-forward "<!--table_data_wideright-->" nil t)
                  (goto-char (point-min))
                  (while (re-search-forward
                          "^<t[hd].*class=\"\\(right\\)\".*$" nil t)
                    (replace-match "right wideright" t nil nil 1)))
 
                ; And save the changes
                (basic-save-buffer))
       (unless visitingp (kill-buffer work-buffer))))))
 
 
 (defun gpgweb-org-to-html (plist filename pub-dir)
   "The publishing function used by the HTML exporter"
   (gpgweb-postprocess-html plist
                            filename
                            (org-gpgweb-publish-to-html plist filename pub-dir)
                            nil))
 
 
 (defun gpgweb-render-blog (&optional filelist)
   "Turn the current buffer which has an org-mode blog entry into its
 rendered form and save it with the suffix .html."
   (interactive)
   (let* ((extplist '(:language "en"
                      :section-numbers nil
                      :tags nil
                      :with-toc nil))
          (orgfile (buffer-file-name))
          (plist (org-export-get-environment 'gpgweb nil extplist))
          (htmlfile (org-gpgweb-export-to-html nil nil nil t extplist)))
     (gpgweb-postprocess-html plist orgfile htmlfile (if filelist filelist t))))
 
 
 (defun gpgweb-publish-blogs ()
   "Publish all blog entries in the current directory"
   (interactive)
   (let ((orgfiles (directory-files gpgweb-blog-dir nil "^2[0-9]+-.*\.org$")))
     (dolist (file (cons "index.org" orgfiles))
       (let* ((file2 (concat gpgweb-blog-dir file))
              (visitingp (find-buffer-visiting file2))
              (work-buffer (or visitingp (find-file-noselect file2))))
         (with-current-buffer work-buffer
           (setq default-directory gpgweb-stage-dir)
           (toggle-read-only 0)
           (gpgweb-render-blog orgfiles)
           (basic-save-buffer))
         (unless visitingp
           (kill-buffer work-buffer))))))
 
 
 (defun gpgweb-upload ()
   "We don't do an upload directly.  Instead we only print the
 commands to do that.  In reality a cron jobs syncs the stage dir."
   (let ((stagedir (plist-get project-plist :publishing-directory)))
     (message "gpgweb  rootdir '%s'" gpgweb-root-dir)
     (message "gpgweb stagedir '%s'" stagedir)
     (message
      (concat "cd " gpgweb-root-dir " && cd " stagedir
              " && echo rsync -rlt --exclude \"*~\" ./ "
              "werner@trithemius.gnupg.org:"
              "/var/www/www/www.gnupg.org/htdocs/ ;"
              " echo ssh werner@trithemius.gnupg.org"
              " touch /var/www/www/www.gnupg.org/htdocs/donate/donors.dat"))
 ))
 
 (provide 'gpgweb)