allow 0-9 in words as well

README.md

@@ -56,6 +56,8 @@ modify `$PREFIX` during installation time like this:
                               /usr/share/dict/american-english)
     -l --minlen <count>       Minimum word len (default: 5)
     -m --maxlen <count>       Maximum word len (default: 10)
+    -n --dontjump             Use all words in the dict file, e.g.
+                              if it is an original diceware list   
     -d --debug                Enable debug output
     -v --version              Print program version
     -h -? --help              Print this help screen

dicepwgen.1

@@ -133,7 +133,7 @@
 .\" ========================================================================
 .\"
 .IX Title "DICEPWGEN 1"
-.TH DICEPWGEN 1 "2015-08-02" "perl v5.18.2" "User Contributed Perl Documentation"
+.TH DICEPWGEN 1 "2015-08-03" "perl v5.18.2" "User Contributed Perl Documentation"
 .\" For nroff, turn off justification.  Always turn off hyphenation; it makes
 .\" way too many mistakes in technical documents.
 .if n .ad l
@@ -142,7 +142,7 @@
 dicepwgen \- A diceware password generator
 .SH "SYNOPSIS"
 .IX Header "SYNOPSIS"
-.Vb 11
+.Vb 10
 \&    Usage: dice [\-tcfvhd]
 \&    Options:
 \&    \-t \-\-humantoss            Asks interactively for rolled dices
@@ -151,6 +151,8 @@ dicepwgen \- A diceware password generator
 \&                              /usr/share/dict/american\-english)
 \&    \-l \-\-minlen <count>       Minimum word len (default: 5)
 \&    \-m \-\-maxlen <count>       Maximum word len (default: 10)
+\&    \-n \-\-dontjump             Use all words in the dict file, e.g.
+\&                              if it is an original diceware list
 \&    \-d \-\-debug                Enable debug output
 \&    \-v \-\-version              Print program version
 \&    \-h \-? \-\-help              Print this help screen
@@ -170,11 +172,35 @@ minimum and maximum word length.
 You can tell dicepwgen to use another dictionary file with
 the option \fB\-f\fR.
 .PP
+If you're using a precomputed diceware list, use the parameter
+\&\fB\-n\fR, in which case dicepwgen will use all entries in the file.
+.PP
 The program only uses words which contain 7bit \s-1ASCII\s0 letters
-(a\-zA-Z), which are easier for password usage anyway.
+(a\-zA\-Z0\-9), which are easier for password usage anyway.
 .SH "FILES"
 .IX Header "FILES"
 \&\fB/usr/share/dict/american\-english\fR: default dictionary file.
+.PP
+You can use almost any dictionary file you want. Dicepwgen expects
+the file to be in the following format: one word per line.
+.PP
+You can also use precomputed diceware word lists by adding the
+option \fB\-n\fR to the commandline. If the file already contains
+numbers, you have to remove them before using. Say, the file
+looks like this:
+.PP
+.Vb 4
+\& 11126   abase
+\& 11131   abash
+\& 11132   abate
+\& 11133   abbas
+.Ve
+.PP
+Prepare the file with this shell oneliner:
+.PP
+.Vb 1
+\& grep "^[0\-9]" dicewarelist.txt | awk \*(Aq{print $2}\*(Aq >> newlist.txt
+.Ve
 .SH "SEE ALSO"
 .IX Header "SEE ALSO"
 <http://world.std.com/~reinhold/diceware.html>
@@ -185,6 +211,15 @@ The program only uses words which contain 7bit \s-1ASCII\s0 letters
 In order to report a bug, unexpected behavior, feature requests
 or to submit a patch, please open an issue on github:
 <https://github.com/TLINDEN/diceware/issues>.
+.PP
+Current known issues:
+.IP "\(bu" 4
+The program does not count the lines in a dictionary file before
+using it. If the file contains a low number of entries (e.g. because
+it is a precomputed diceware word list and \fB\-n\fR has not been
+specified), it seems to \*(L"hang\*(R". In fact it runs very long because it
+jumps over a random number of entries and restarts from the beginning
+again and again til the required number of words has been collected (7776).
 .SH "LICENSE"
 .IX Header "LICENSE"
 This software is licensed under the \s-1GNU GENERAL PUBLIC LICENSE\s0 version 3.

dicepwgen.c

@@ -32,6 +32,8 @@ int usage() {
 	  "                          /usr/share/dict/american-english)\n"
 	  "-l --minlen <count>       Minimum word len (default: 5)\n"
 	  "-m --maxlen <count>       Maximum word len (default: 10)\n"
+	  "-n --dontjump             Use all words in the dict file, e.g.\n"
+	  "                          if it is an original diceware list\n"
 	  "-d --debug                Enable debug output\n"
 	  "-v --version              Print program version\n"
 	  "-h -? --help              Print this help screen\n"
@@ -46,8 +48,7 @@ int main (int argc, char **argv)  {
 
   WMIN = 6;
   WMAX = 10;
-  humantoss = 0;
+  humantoss = verbose = dontjump = 0;
-  verbose = 0;
   
   static struct option longopts[] = {
     { "wordcount", required_argument, NULL,           'c' },
@@ -55,12 +56,13 @@ int main (int argc, char **argv)  {
     { "maxlen",    required_argument, NULL,           'm' },
     { "humantoss", required_argument, NULL,           't' },    
     { "dictfile",  required_argument, NULL,           'f' },
+    { "dontjump",  no_argument,       NULL,           'n' },   
     { "version",   no_argument,       NULL,           'v' },
     { "help",      no_argument,       NULL,           'h' },
     { "debug",     no_argument,       NULL,           'd' },
   };
 
-   while ((opt = getopt_long(argc, argv, "l:m:tf:c:vh?d", longopts, NULL)) != -1) {
+   while ((opt = getopt_long(argc, argv, "l:m:tf:c:vh?dn", longopts, NULL)) != -1) {
      switch (opt) {
      case 'v':
        fprintf(stderr, "This is %s version %s\n", argv[0], VERSION);
@@ -85,6 +87,9 @@ int main (int argc, char **argv)  {
      case 'd':
        verbose++;
        break;
+     case 'n':
+       dontjump = 1;
+       break;
      case 'f':
        dictfile = malloc(strlen(optarg));
        strncpy(dictfile, optarg, strlen(optarg));
@@ -99,6 +104,11 @@ int main (int argc, char **argv)  {
      dictfile =  STRINGIZE_VALUE_OF(DICTFILE);
    }
 
+   if(dontjump) {
+     WMIN = 0;
+     WMAX = 128;
+   }
+   
    debug("     using dictfile: %s", dictfile);
    debug("minimum word length: %d", WMIN);
    debug("maximum word length: %d", WMAX);

dicepwgen.h

@@ -42,6 +42,7 @@
 
 int humantoss;
 int verbose;
+int dontjump;
 int WMIN;
 int WMAX;
 

dicepwgen.pod

@@ -12,6 +12,8 @@ dicepwgen - A diceware password generator
                               /usr/share/dict/american-english)
     -l --minlen <count>       Minimum word len (default: 5)
     -m --maxlen <count>       Maximum word len (default: 10)
+    -n --dontjump             Use all words in the dict file, e.g.
+                              if it is an original diceware list
     -d --debug                Enable debug output
     -v --version              Print program version
     -h -? --help              Print this help screen
@@ -31,13 +33,33 @@ minimum and maximum word length.
 You can tell dicepwgen to use another dictionary file with
 the option B<-f>.
 
+If you're using a precomputed diceware list, use the parameter
+B<-n>, in which case dicepwgen will use all entries in the file.
+
 The program only uses words which contain 7bit ASCII letters
-(a-zA-Z), which are easier for password usage anyway.
+(a-zA-Z0-9), which are easier for password usage anyway.
 
 =head1 FILES
 
 B</usr/share/dict/american-english>: default dictionary file.
 
+You can use almost any dictionary file you want. Dicepwgen expects
+the file to be in the following format: one word per line.
+
+You can also use precomputed diceware word lists by adding the
+option B<-n> to the commandline. If the file already contains
+numbers, you have to remove them before using. Say, the file
+looks like this:
+
+ 11126   abase
+ 11131   abash
+ 11132   abate
+ 11133   abbas
+
+Prepare the file with this shell oneliner:
+
+ grep "^[0-9]" dicewarelist.txt | awk '{print $2}' >> newlist.txt
+
 =head1 SEE ALSO
 
 L<http://world.std.com/~reinhold/diceware.html>
@@ -50,6 +72,21 @@ In order to report a bug, unexpected behavior, feature requests
 or to submit a patch, please open an issue on github:
 L<https://github.com/TLINDEN/diceware/issues>.
 
+Current known issues:
+
+=over
+
+=item *
+
+The program does not count the lines in a dictionary file before
+using it. If the file contains a low number of entries (e.g. because
+it is a precomputed diceware word list and B<-n> has not been
+specified), it seems to "hang". In fact it runs very long because it
+jumps over a random number of entries and restarts from the beginning
+again and again til the required number of words has been collected (7776).
+
+=back
+
 =head1 LICENSE
 
 This software is licensed under the GNU GENERAL PUBLIC LICENSE version 3.

dictfile.c

@@ -106,18 +106,20 @@ char **fetch_dict(char *dictfile) {
   
  LOOP:
   while ((linelen = getline(&line, &len, DICT)) != -1) {
-    if(jump > 0) {
+    if(! dontjump) {
-      jump--;
+      if(jump > 0) {
-      continue;
+	jump--;
-    }
+	continue;
-    else {
+      }
-      jump = rand_lim(32);
+      else {
+	jump = rand_lim(32);
+      }
     }
     if(linelen >= WMIN+1 && linelen <= WMAX+1) {
       line[linelen-1] = '\0'; /* remove newline */
 
       for(i=0; i<linelen-1; i++) {
-	if(isalpha((int)line[i]) == 0) {
+	if(isalnum((int)line[i]) == 0) {
 	  next = 1;
 	  break;
 	}

dictfile.h

@@ -36,6 +36,7 @@
 
 extern int WMIN;
 extern int WMAX;
+extern int dontjump;
 
 int *incr_dicedigit(int *digits);
 int get_dicenum(int *digits);