# Go through each file, open it, and add its content to the list
for myFile in filesList:
with open(os.path.join(directory, myFile), 'r') as f:
# Here, the JSON is converted back to a Python object
transcript = json.load(f)
transcripts.append(transcript)
# Create lists for the years and the length of the text for each year.
years = []
lengths = []
# Go through each transcript
for transcript in transcripts:
# Get the date - converting the ISO date back into a datetime.date object
date = cf.iso_to_datetime(transcript['date'])
# Convert the year into a campaign year
year = cf.campaign_year_from_year(date.year)
years.append(year)
# Create a string for all of the text in the debate
allText = ""
# Add all the text spoken by speakers to that string
for speaker in transcript['text_by_speakers']:
allText += (" " + speaker['text'])
# removes punctuation, digits, splits text into words
# remove words shorter than 3 characters and suffixes
with open(os.path.join(directory, myFile), 'r') as f:
# Here, the JSON is converted back to a Python object
transcript = json.load(f)
transcripts.append(transcript)
all_list = []
# Loop through the three 'parties': 'r' - republican; 'd' - democrat; 't' - anything else
for winnerloser in ['w', 'l']:
transcript_results = []
# Go through each transcript
for transcript in transcripts:
# Get the date - converting the ISO date back into a datetime.date object
year = cf.iso_to_datetime(transcript['date']).year
# Create a string for all of the text in the debate
allText = ""
# Add all the text spoken by speakers to that string
for speaker in transcript['text_by_speakers']:
if 'winnerloser' in speaker:
if speaker['winnerloser'] == winnerloser or (
(speaker['winnerloser'] == 'n' or speaker['winnerloser']
== 'd' or speaker['winnerloser'] == 'r')
and winnerloser == 'l'):
print winnerloser
print cf.unicode_to_ascii(transcript['description'])
allText += (" " + speaker['text'])
username = '******'
else:
import_filename = 'realDonaldTrumpTweetsResults.json'
export_filename = 'images/twitter-analysis-sentiment-time-trump.svg'
username = '******'
export_filename = os.path.join(root_directory, export_filename)
title = 'Positive and negative words in tweets - @' + username
with open(import_filename, 'r') as f:
statuses = json.load(f)
results = {}
for status in statuses:
month = cf.iso_to_datetime(status['date']).month
year = cf.iso_to_datetime(status['date']).year
if year == 2016:
if month not in results:
results[month] = [status]
else:
results[month].append(status)
positive_results = [
(month_no,
cf.mean([status['total_pos_words'] for status in results[month_no]]))
for month_no in results
]
negative_results = [
(month_no,
cf.mean([status['total_neg_words'] for status in results[month_no]]))