|
1
|
|
|
#!/usr/bin/env python |
|
2
|
|
|
# encoding=utf8 |
|
3
|
|
|
""" |
|
4
|
|
|
Show the number of mentions per month for a word: |
|
5
|
|
|
word_usage -w twizzle |
|
6
|
|
|
word_usage -w bae -y 2014 |
|
7
|
|
|
word_usage -w twizzle -s favorite |
|
8
|
|
|
word_usage -w twizzle -s favourite |
|
9
|
|
|
""" |
|
10
|
|
|
from __future__ import print_function |
|
11
|
|
|
from most_frequent_words import commafy |
|
12
|
|
|
import argparse |
|
13
|
|
|
from word_charts import load_csv, filter_year, filter_search_term |
|
14
|
|
|
from collections import OrderedDict |
|
15
|
|
|
|
|
16
|
|
|
|
|
17
|
|
|
def filter_word(tweets, desired_word): |
|
18
|
|
|
found = [] |
|
19
|
|
|
for tweet in tweets: |
|
20
|
|
|
if desired_word == tweet["word"]: |
|
21
|
|
|
found.append(tweet) |
|
22
|
|
|
print("Total " + desired_word + ":\t" + commafy(len(found))) |
|
23
|
|
|
return found |
|
24
|
|
|
|
|
25
|
|
|
|
|
26
|
|
|
def print_per_month(tweets): |
|
27
|
|
|
counts = OrderedDict() |
|
28
|
|
|
for tweet in tweets: |
|
29
|
|
|
year = tweet["created_at"][-4:] |
|
30
|
|
|
month = tweet["created_at"][4:7] |
|
31
|
|
|
month_year = month + " " + year |
|
32
|
|
|
if month_year not in counts: |
|
33
|
|
|
counts[month_year] = 0 |
|
34
|
|
|
counts[month_year] += 1 |
|
35
|
|
|
|
|
36
|
|
|
# from pprint import pprint |
|
37
|
|
|
xs, ys = [], [] |
|
38
|
|
|
for month_year in counts: |
|
39
|
|
|
print(month_year, counts[month_year]) |
|
40
|
|
|
xs.append(month_year) |
|
41
|
|
|
ys.append(counts[month_year]) |
|
42
|
|
|
|
|
43
|
|
|
return xs, ys |
|
44
|
|
|
|
|
45
|
|
|
|
|
46
|
|
|
if __name__ == "__main__": |
|
47
|
|
|
parser = argparse.ArgumentParser( |
|
48
|
|
|
description="Show the number of mentions per month for a word.", |
|
49
|
|
|
formatter_class=argparse.ArgumentDefaultsHelpFormatter) |
|
50
|
|
|
parser.add_argument( |
|
51
|
|
|
'-c', '--csv', default='M:/bin/data/all.csv', |
|
52
|
|
|
help='Input CSV file') |
|
53
|
|
|
parser.add_argument( |
|
54
|
|
|
'-w', '--word', default='bae', |
|
55
|
|
|
help='Word to search for') |
|
56
|
|
|
parser.add_argument( |
|
57
|
|
|
'-y', '--year', type=int, default=None, |
|
58
|
|
|
help="Only from this year") |
|
59
|
|
|
parser.add_argument( |
|
60
|
|
|
'-s', '--search_term', |
|
61
|
|
|
help="Only for this search term") |
|
62
|
|
|
parser.add_argument( |
|
63
|
|
|
'-p', '--plot', action='store_true', |
|
64
|
|
|
help="Create a bar graph of results (requires Plotly authentication: " |
|
65
|
|
|
"https://plot.ly/python/getting-started/") |
|
66
|
|
|
args = parser.parse_args() |
|
67
|
|
|
|
|
68
|
|
|
tweets = load_csv(args.csv) |
|
69
|
|
|
print("Total tweets:\t" + commafy(len(tweets))) |
|
70
|
|
|
|
|
71
|
|
|
tweets = filter_word(tweets, args.word) |
|
72
|
|
|
|
|
73
|
|
|
if args.search_term: |
|
74
|
|
|
tweets = filter_search_term(tweets, args.search_term) |
|
75
|
|
|
|
|
76
|
|
|
if args.year: |
|
77
|
|
|
tweets = filter_year(tweets, args.year) |
|
78
|
|
|
|
|
79
|
|
|
# Sort by ID = sort chronologically |
|
80
|
|
|
tweets = sorted(tweets, key=lambda k: k['id_str']) |
|
81
|
|
|
|
|
82
|
|
|
xs, ys = print_per_month(tweets) |
|
83
|
|
|
|
|
84
|
|
|
if args.plot: |
|
85
|
|
|
print("Plotting...") |
|
86
|
|
|
import plotly.plotly as py # pip install plotly + needs auth |
|
87
|
|
|
import plotly.graph_objs as go |
|
88
|
|
|
layout = go.Layout( |
|
89
|
|
|
title=args.word |
|
90
|
|
|
) |
|
91
|
|
|
trace0 = go.Scatter( |
|
92
|
|
|
x=xs, |
|
93
|
|
|
y=ys, |
|
94
|
|
|
mode='lines+markers', |
|
95
|
|
|
name=args.word |
|
96
|
|
|
) |
|
97
|
|
|
data = [trace0] |
|
98
|
|
|
fig = go.Figure(data=data, layout=layout) |
|
99
|
|
|
plot_url = py.plot(fig, filename='word_usage_' + args.word) |
|
100
|
|
|
print(plot_url) |
|
101
|
|
|
|
|
102
|
|
|
# End of file |
|
103
|
|
|
|
hugovk /
word-tools
