def word_frequency(query_name):
data_container = []
query_name = query_name
spans = [{
'from_parameter': '2018-01-01',
'to': '2018-01-15'
}, {
'from_parameter': '2018-01-16',
'to': '2018-01-31'
}, {
'from_parameter': '2018-02-01',
'to': '2018-02-15'
}, {
'from_parameter': '2018-02-15',
'to': '2018-02-28'
}, {
'from_parameter': '2018-03-01',
'to': '2018-03-15'
}, {
'from_parameter': '2018-03-16',
'to': '2018-03-31'
}]
for span in spans:
news = News()
data = news.fetch_news(q=query_name,
language='en',
from_parameter=span['from_parameter'],
to=span['to'])
# print(data)
analysis = news.analyze_count(data)
data_container.append(analysis)
wordset = set()
combined_data_container = {}
for analysis in data_container:
for key in analysis.keys():
if key in wordset:
combined_data_container[key] = {
**combined_data_container[key],
**analysis[key]
}
else:
wordset.add(key)
combined_data_container[key] = analysis[key]
# var trace1 = {
# x: [1, 2, 3, 4],
# y: [10, 15, 13, 17],
# mode: 'markers',
# name: 'Scatter'
# };
js_data_container = []
for word in combined_data_container:
x = list(combined_data_container[word].keys())
x.sort()
y = []
for date in x:
y.append(int(combined_data_container[word][date]))
trace = {'x': x, 'y': y, 'name': word}
if sum(trace['y']) > 100 and trace['name'] != query_name.lower():
js_data_container.append(trace)
layout = {
'title':
'Other Words Found in Headlines that include "{}" (2018-YTD)'.format(
query_name),
'xaxis': {
'title': 'Week Number'
},
'yaxis': {
'title': 'Count of Appearances in Headlines'
}
}
# save raw js to the database with id=query_name above
return [js_data_container, layout]
