|
1
|
|
|
#!/usr/bin/env python |
|
2
|
|
|
# encoding=utf8 |
|
3
|
|
|
""" |
|
4
|
|
|
Get top word charts from a CSV file of logged tweets: |
|
5
|
|
|
word_charts.py -n 100 |
|
6
|
|
|
Get top in a given year: |
|
7
|
|
|
word_charts.py -y 2014 -n 100 |
|
8
|
|
|
Get top in a given year that aren't in the previous year. |
|
9
|
|
|
word_charts.py -y 2014 -n 100 --diff |
|
10
|
|
|
Get top for a search term: |
|
11
|
|
|
word_charts.py -s favorite |
|
12
|
|
|
word_charts.py -s favourite |
|
13
|
|
|
""" |
|
14
|
|
|
from __future__ import print_function |
|
15
|
|
|
from most_frequent_words import most_frequent_words_and_counts, commafy |
|
16
|
|
|
import argparse |
|
17
|
|
|
import csv |
|
18
|
|
|
|
|
19
|
|
|
|
|
20
|
|
|
def filter_year(tweets, desired_year): |
|
21
|
|
|
found = [] |
|
22
|
|
|
for tweet in tweets: |
|
23
|
|
|
year = int(tweet["created_at"][-4:]) |
|
24
|
|
|
if desired_year == year: |
|
25
|
|
|
found.append(tweet) |
|
26
|
|
|
print("Total in " + str(desired_year) + ":\t" + commafy(len(found))) |
|
27
|
|
|
return found |
|
28
|
|
|
|
|
29
|
|
|
|
|
30
|
|
|
def filter_search_term(tweets, desired_search_term): |
|
31
|
|
|
found = [] |
|
32
|
|
|
for tweet in tweets: |
|
33
|
|
|
if desired_search_term in tweet["search_term"]: |
|
34
|
|
|
found.append(tweet) |
|
35
|
|
|
print("Total " + desired_search_term + ":\t" + commafy(len(found))) |
|
36
|
|
|
return found |
|
37
|
|
|
|
|
38
|
|
|
|
|
39
|
|
|
def print_chart(top): |
|
40
|
|
|
""" |
|
41
|
|
|
param top: a list of (word, count) tuples. |
|
42
|
|
|
""" |
|
43
|
|
|
if args.html: |
|
44
|
|
|
print("<ol>") |
|
45
|
|
|
|
|
46
|
|
|
for i, (word, count) in enumerate(top): |
|
47
|
|
|
if args.html: |
|
48
|
|
|
print("<li>{0} ({1})</li>".format(word, commafy(count))) |
|
49
|
|
|
else: |
|
50
|
|
|
print(str(i+1) + ". " + word + " (" + commafy(count) + ")") |
|
51
|
|
|
|
|
52
|
|
|
if args.html: |
|
53
|
|
|
print("</ol>") |
|
54
|
|
|
|
|
55
|
|
|
|
|
56
|
|
|
def print_top(tweets, number=10, year=None, search_term=None): |
|
57
|
|
|
words = [] |
|
58
|
|
|
|
|
59
|
|
|
if search_term: |
|
60
|
|
|
tweets = filter_search_term(tweets, search_term) |
|
61
|
|
|
|
|
62
|
|
|
if year: |
|
63
|
|
|
tweets = filter_year(tweets, year) |
|
64
|
|
|
|
|
65
|
|
|
for tweet in tweets: |
|
66
|
|
|
words.append(tweet['word']) |
|
67
|
|
|
|
|
68
|
|
|
print() |
|
69
|
|
|
title = "# Top " + str(number) |
|
70
|
|
|
if year: |
|
71
|
|
|
title += " (" + str(year) + ")" |
|
72
|
|
|
print(title) |
|
73
|
|
|
print() |
|
74
|
|
|
top = most_frequent_words_and_counts(words, number) |
|
75
|
|
|
print_chart(top) |
|
76
|
|
|
return top |
|
77
|
|
|
|
|
78
|
|
|
|
|
79
|
|
|
# cmd.exe cannot do Unicode so encode first |
|
80
|
|
|
def print_it(text): |
|
81
|
|
|
print(text.encode('utf-8')) |
|
82
|
|
|
|
|
83
|
|
|
|
|
84
|
|
|
def load_csv(filename): |
|
85
|
|
|
# with codecs.open(filename, mode='rb', encoding='cp1252') as fd: |
|
86
|
|
|
with open(filename, mode='rb') as fd: |
|
87
|
|
|
data = csv.DictReader(fd) |
|
88
|
|
|
rows = [] |
|
89
|
|
|
seen = set() # avoid duplicates |
|
90
|
|
|
for row in data: |
|
91
|
|
|
if row['id_str'] not in seen and row['word'] not in ['actually']: |
|
92
|
|
|
seen.add(row['id_str']) |
|
93
|
|
|
rows.append(row) |
|
94
|
|
|
return rows |
|
95
|
|
|
|
|
96
|
|
|
|
|
97
|
|
|
if __name__ == "__main__": |
|
98
|
|
|
parser = argparse.ArgumentParser( |
|
99
|
|
|
description="Get top word charts from a CSV file of logged tweets.", |
|
100
|
|
|
formatter_class=argparse.ArgumentDefaultsHelpFormatter) |
|
101
|
|
|
parser.add_argument( |
|
102
|
|
|
'-c', '--csv', default='M:/bin/data/favibot.csv', |
|
103
|
|
|
help='Input CSV file') |
|
104
|
|
|
parser.add_argument( |
|
105
|
|
|
'-n', '--top-number', type=int, default=10, |
|
106
|
|
|
help="Show top X") |
|
107
|
|
|
parser.add_argument( |
|
108
|
|
|
'-y', '--year', type=int, default=None, |
|
109
|
|
|
help="Only from this year") |
|
110
|
|
|
parser.add_argument( |
|
111
|
|
|
'-s', '--search_term', |
|
112
|
|
|
help="Only for this search term") |
|
113
|
|
|
parser.add_argument( |
|
114
|
|
|
'--diff', action='store_true', |
|
115
|
|
|
help="Compare a year to the previous year") |
|
116
|
|
|
parser.add_argument( |
|
117
|
|
|
'--html', action='store_true', |
|
118
|
|
|
help="Output with html markup") |
|
119
|
|
|
args = parser.parse_args() |
|
120
|
|
|
|
|
121
|
|
|
tweets = load_csv(args.csv) |
|
122
|
|
|
print("Total tweets:\t" + commafy(len(tweets))) |
|
123
|
|
|
|
|
124
|
|
|
if args.diff and args.year: |
|
125
|
|
|
last_year = print_top(tweets, number=args.top_number, year=args.year-1, |
|
126
|
|
|
search_term=args.search_term) |
|
127
|
|
|
this_year = print_top(tweets, number=args.top_number, year=args.year, |
|
128
|
|
|
search_term=args.search_term) |
|
129
|
|
|
last_years_words = [e[0] for e in last_year] |
|
130
|
|
|
this_years_words = [e[0] for e in this_year] |
|
131
|
|
|
|
|
132
|
|
|
set1 = set(last_years_words) |
|
133
|
|
|
diff = [x for x in this_years_words if x not in set1] |
|
134
|
|
|
|
|
135
|
|
|
top = [] |
|
136
|
|
|
for word in diff: |
|
137
|
|
|
for e in this_year: |
|
138
|
|
|
if e[0] == word: |
|
139
|
|
|
top.append(e) |
|
140
|
|
|
continue |
|
141
|
|
|
print() |
|
142
|
|
|
top_x = " top " + str(args.top_number) |
|
143
|
|
|
print("# New entries in the " + str(args.year) + top_x + |
|
144
|
|
|
" which weren't in the " + str(args.year-1) + top_x) |
|
145
|
|
|
print() |
|
146
|
|
|
print_chart(top) |
|
147
|
|
|
|
|
148
|
|
|
else: |
|
149
|
|
|
print_top(tweets, number=args.top_number, year=args.year, |
|
150
|
|
|
search_term=args.search_term) |
|
151
|
|
|
|
|
152
|
|
|
|
|
153
|
|
|
# End of file |
|
154
|
|
|
|
hugovk /
word-tools
