calculate letter frequency from wordlist

[hexbog.git] / freq.coffee

diff --git a/freq.coffee b/freq.coffee
new file mode 100644 (file)
index 0000000..bf4e497
--- /dev/null
+++ b/freq.coffee
@@ -0,0 +1,65 @@
+#!/usr/bin/coffee
+
+# this script tries to calculate an optimal letter distrobution.
+
+fs = require 'fs'
+
+
+fs.readFile 'wordlist.txt', 'utf8', (err, data) ->
+       if err?
+               console.log "Error reading #{orig_file}: #{err}"
+               process.exit(1)
+       else
+               weights = {
+                       a: 0
+                       b: 0
+                       c: 0
+                       d: 0
+                       e: 0
+                       f: 0
+                       g: 0
+                       h: 0
+                       i: 0
+                       j: 0
+                       k: 0
+                       l: 0
+                       m: 0
+                       n: 0
+                       o: 0
+                       p: 0
+                       q: 0
+                       r: 0
+                       s: 0
+                       t: 0
+                       u: 0
+                       v: 0
+                       w: 0
+                       x: 0
+                       y: 0
+                       z: 0
+                       qu: 0
+               }
+               words = data.split '\n'
+               for word in words
+                       weight = 1.0 * Math.pow .6, (Math.abs(word.length - 4))
+                       i = 0
+                       while i < word.length
+                               if word[i] is 'q' and i < word.length and word[i+1] is 'u'
+                                       weights['qu'] += weight
+                                       i += 2
+                               else
+                                       #if word[i] is 'u' and i > 0 and word[i-1] is 'q'
+                                       #       console.log "skipping the u of a qu didn't work"
+                                       weights[word[i]] += weight
+                                       i += 1
+
+               total = 0
+               for letter, weight of weights
+                       total += Math.round(weight)
+
+               for letter, weight of weights
+                       weight = Math.round(weight)
+                       #console.log "#{letter}: #{Math.round(weight / total * 1000)} (1 in #{Math.round(total / weight)})"
+                       console.log "#{weight} # #{letter}"
+
+               console.log "total: #{total}"