Files
exercises-in-programming-style/16-reflective/tf-16.py
2013-12-29 09:55:02 -08:00

49 lines
1.5 KiB
Python
Executable File

#!/usr/bin/env python
import sys, re, operator, string, os
#
# Two down-to-earth things
#
stops = set(open("../stop_words.txt").read().split(",") + list(string.ascii_lowercase))
def frequencies_imp(word_list):
word_freqs = {}
for w in word_list:
if w in word_freqs:
word_freqs[w] += 1
else:
word_freqs[w] = 1
return word_freqs
#
# Let's write our functions as strings.
#
if len(sys.argv) > 1:
extract_words_func = "lambda name : [x.lower() for x in re.split('[^a-zA-Z]+', open(name).read()) if len(x) > 0 and x.lower() not in stops]"
frequencies_func = "lambda wl : frequencies_imp(wl)"
sort_func = "lambda word_freq: sorted(word_freq.iteritems(), key=operator.itemgetter(1), reverse=True)"
filename = sys.argv[1]
else:
extract_words_func = "lambda x: []"
frequencies_func = "lambda x: []"
sort_func = "lambda x: []"
filename = os.path.basename(__file__)
#
# So far, this program isn't much about term-frequency. It's about
# a bunch of strings that look like functions.
# Let's add our functions to the "base" program, dynamically.
#
exec('extract_words = ' + extract_words_func)
exec('frequencies = ' + frequencies_func)
exec('sort = ' + sort_func)
#
# The main function. This would work just fine:
# word_freqs = sort(frequencies(extract_words(filename)))
#
word_freqs = locals()['sort'](locals()['frequencies'](locals()['extract_words'](filename)))
for (w, c) in word_freqs[0:25]:
print w, ' - ', c