import nltk

def countWords(searchWords, textWords):
	tempCount = 0
	for aWord in searchWords:
		tempCount += textWords.count( aWord )
	return tempCount
	
wordQuery = [ "loves", "love", "loved", "loveth", "lov", "loue", "loues" ]
wordQuery = [ "kill", "killed", "kills", "murder", "murdered", "killeth", "killes" ]

print "<?xml version=\"1.0\" encoding=\"utf-8\"?>"
print "<data>"
for book in nltk.corpus.gutenberg.items:
	vWords = nltk.corpus.gutenberg.words(book)
	wordNum = countWords(wordQuery, vWords)
	print "<record book=\""+book+"\" len=\""+str(len(vWords))+"\" count=\""+str(wordNum)+"\" />"
print "</data>"