Closing out several small problems

README.txt

@@ -1,13 +1,19 @@
-Installer: http://MrFeinberg.com/haikufinder-1.1.zip
+Files: http://MrFeinberg.com/haikufinder-1.1.zip or use github repo.
 
-* 
-*	REQUIRES NLTK, THE NATURAL LANGUAGE TOOLKIT
-*	http://www.nltk.org/download
-*
+Installation requires NLTK and its punkt package. First-time steps
+Unzip main file.
+pip install nltk
+python
+import nltk
+nltk.download('punkt')
+exit()
+python setup.py install
+
+A script to begin processing is located in \scripts
 
 A Python hack to find "haikus" in English text. For the purposes of this
 module, a "haiku" is one or more complete sentences that, together, can be
-broken into groups of 5, 7, and 5 syllables. Each canididate haiku line, 
+broken into groups of 5, 7, and 5 syllables. Each candidate haiku line, 
 and then the entire haiku, has to make it through a few heuristics to 
 filter out constructions that are likely to scan awkwardly (like verb 
 phrases split across lines). Since this code doesn't really try to 

haikufinder/__init__.py

@@ -71,8 +71,8 @@ def read_alternates(which):
     syllables = pickle.load(p)
 with open(file('cmudict/custom.dict'), 'r') as p:
     for line in p.readlines():
-    	if not len(line):
-    		continue
+        if not len(line):
+            continue
         (word, count) = line.split()
         syllables[word] = int(count)
 

scripts/findhaikus → scripts/findhaikus.py

@@ -34,7 +34,7 @@
     text = open(sys.argv[1], "r").read()
 except:
     text = sys.stdin.read()
-for haiku in HaikuFinder(text.decode('utf-8')).find_haikus():
+for haiku in HaikuFinder(unicode(text, errors='replace')).find_haikus():
     print(haiku[0])
     print("    %s" % haiku[1])
     print(haiku[2])