# HG changeset patch
# User Bernhard Reiter <bernhard@intevation.de>
# Date 1475767726 -7200
# Node ID 85c65a59742033dc2ddb17af600ac162d3ee6097
# Parent  757625ec8364edb8bb0dc97e2244b63ea4230f7c
Improves: command line options and code style.

Adds a command line option for dumping the dictionary into a file.
Improves source code style to better conform to pep8.
Calls it BETA now.

diff -r 757625ec8364 -r 85c65a597420 ppgen.py
--- a/ppgen.py	Mon May 02 09:26:29 2016 +0200
+++ b/ppgen.py	Thu Oct 06 17:28:46 2016 +0200
@@ -1,11 +1,11 @@
 #!/usr/bin/env python3
-"""Create a passphrase from a few random words. DRAFT
+"""Create a random passphrase from a dictionary of words. BETA
 
 Relies on the entropy of python's
     random.SystemRandom class
         which (according to the documentation) calls os.urandom()
         which (according to the documentation) calls the operating system
-           specific randomness source which "should be unpredictable 
+           specific randomness source which "should be unpredictable
             enough for cryptographic applications"
 
 Requires:
@@ -28,6 +28,7 @@
 comes without any warranty (to extend permissible under applicable law).
 """
 
+import argparse
 import math
 import re
 import sys
@@ -37,13 +38,15 @@
 
 tainted = False   # to be set if we find a hint that the passphrase may be weak
 
-def buildDictionary():
+
+def buildDictionary(options):
     """Build up a dictionary of unique words, calculate stats."""
     global tainted
     d = []
 
     # dictionary for testing
-    #d = ["abc", "aBc", "cde", "efg", "hij", "blubber", "jikf", "zug", "lmf", "opq"]
+    #d = ["abc", "aBc", "cde", "efg", "hij", "blubber",
+    #      "jikf", "zug", "lmf", "opq"]
     # second test dictionary to show that different string functions are used.
     #d =  [''.join('A' * 1000) for _ in range(1000)]
 
@@ -51,11 +54,11 @@
     d = readDingDict(filename="/usr/share/trans/de-en", useLeft=True)
 
     ## for debugging purpuses, dump dictionary
-    #dumpfilename = "ddump.txt"
-    #print("Writing out {}.".format(dumpfilename))
-    #with open(dumpfilename, "w") as f:
-    #    for i in d:
-    #        f.write("{}\n".format(i))
+    if options.ddump_filename:
+        print("Writing out dictionary in '{}'.".format(options.ddump_filename))
+        with open(options.ddump_filename, "w") as f:
+            for i in d:
+                f.write("{}\n".format(i))
 
     # Print some stats on the dictionary to be used
     dl = len(d)
@@ -65,13 +68,13 @@
         tainted = True
 
     print("|= Number of words |= possibilities |")
-    for i in range(1,5):
+    for i in range(1, 5):
         print("|               {:2d} |    2^{:4.1f}      |".format(
-                                    i, math.log(dl**i,2)))
+              i, math.log(dl**i, 2)))
     return d
 
 
-def readDingDict(filename = "/usr/share/trans/de-en", useLeft=False):
+def readDingDict(filename="/usr/share/trans/de-en", useLeft=False):
     """Read dictionary with unique words from file in Ding format.
 
     useLeft: Boolean to control which language to use
@@ -79,19 +82,20 @@
     TODO: add option to use both languages for people that speak them both?
     """
 
-    dset = set() #using the datatype 'set' to aviod duplicates
+    dset = set()  # using the datatype 'set' to aviod duplicates
 
     splitter = re.compile(r"""\ \|\  # first pattern  ' | '
                            |;\       # second pattern '; '
                            |(?<=\S)/(?=\S)  # 3.:    '\' surrounded by chars
                            |\s+      # by whitespace
-                           """,re.VERBOSE)
+                           """, re.VERBOSE)
 
     print("Reading entries from {}.".format(filename), end='')
-    counter = 0 # for progress or stopping early
+    counter = 0  # for progress or stopping early
     with open(filename, "r") as f:
         for line in f:
-            if line[0] == '#': continue
+            if line[0] == '#':
+                continue
 
             # languages are separated by " :: "
             p = line.partition(" :: ")
@@ -114,9 +118,16 @@
 
     return list(dset)
 
+
 def main():
     global tainted
-    dictionary = buildDictionary()
+
+    parser = argparse.ArgumentParser(description=__doc__.splitlines()[0])
+    parser.add_argument('--ddump-filename',
+                        help='filename to dump the dictionary to')
+    options = parser.parse_args()
+
+    dictionary = buildDictionary(options)
 
     howMany = 4
 
@@ -128,7 +139,7 @@
     words = {}
     for x in range(howMany):
         word = _srandom.choice(dictionary)
-        words[word.lower()]= True
+        words[word.lower()] = True
         print(word, end='\n    ')
     print("\n")