Better spreadsheet style, more uniform across columns

This commit is contained in:
Crista Lopes
2013-10-21 19:21:40 -07:00
parent 5cde2691ea
commit 4576b3c5e8

View File

@@ -6,28 +6,34 @@ import sys, re, itertools, operator
#
all_words = [(), None]
stop_words = [(), None]
non_stop_words = [(), lambda w : w if w not in stop_words[0] else '']
unique_words = [(), lambda word_list: set([w for w in word_list if w != ''])]
counts = [(), lambda w, word_list : word_list.count(w)]
sorted_data = [(), lambda z : sorted(z, key=operator.itemgetter(1), reverse=True)]
non_stop_words = [(), lambda : map(lambda w : w if w not in stop_words[0] else '', \
all_words[0])]
unique_words = [(), lambda : set([w for w in non_stop_words[0] if w != ''])]
counts = [(), lambda : map(lambda w, word_list : word_list.count(w), \
unique_words[0], \
itertools.repeat(non_stop_words[0], len(unique_words[0])))]
sorted_data = [(), lambda : sorted(zip(list(unique_words[0]), counts[0]),
key=operator.itemgetter(1), reverse=True)]
# The entire spreadsheet
all_columns = [all_words, stop_words, non_stop_words, unique_words, counts, sorted_data]
#
# The active procedure over the columns of data.
# Call this everytime the input data changes, or periodically.
#
def update():
global non_stop_words
global unique_words
global counts
global sorted_data
# Apply the formulas to the 4 last columns
non_stop_words[0] = map(non_stop_words[1], all_words[0])
unique_words[0] = unique_words[1](non_stop_words[0])
counts[0] = map(counts[1], unique_words[0], itertools.repeat(non_stop_words[0], len(unique_words[0])))
sorted_data[0] = sorted_data[1](zip(list(unique_words[0]), counts[0]))
global all_columns
# Apply the formula in each column
for c in all_columns:
if c[1] != None:
c[0] = c[1]()
#
# The Main program
#
# Load the fixed data into the first 2 columns
#
all_words[0] = re.findall('[a-z]{2,}', open(sys.argv[1]).read().lower())
stop_words[0] = set(open('../stop_words.txt').read().split(','))
# Update the columns with formulas