Question

Greetings fellow enthusiasts! I am working on a project and I am utilizing python to interact with the twitter api.

Goal: Extract the location, text of tweet, created at, and user id from the raw data you see returned in the code from the “pprint.pprint(datares)” into a specified format in csv file.

Problem: How can I get the info I am returning into a csv file so that every row in the file would display as follows:

Row: tweet text, it’s location, created at, user id

The following is my code and shows what I am able to return so far.

import urllib2, json, pprint, codecs, unicodedata

u = urllib2.urlopen('http://search.twitter.com/search.json?geocode=29.762778,-95.383056,25.0mi&page=1&rpp=20')
datares = json.load(u)
##raw data returned
pprint.pprint(datares)

##open csv file
with codecs.open('Geotweets.csv',mode='w', encoding='utf-8',errors='replace') as cache:
##need to save tweets,date,area,id to file
    for tweet in datares['results']:
        print tweet['text']
        archive=tweet['text']
        unicodedata.normalize('NFKD', archive).encode('ascii','ignore')
        cache.write(archive)


for date in datares['results']:
    print date['created_at']
for area in datares['results']:
    print area['location']
for id in datares['results']:
    print id['from_user']

Answer 1

If you want to make a CSV file, use the csv module!

E.g:

with codecs.open('Geotweets.csv',mode='w', encoding='utf-8',errors='replace') as cache:
    writer = csv.writer(cache)
    for tweet in datares['results']:
        writer.writerow([tweet['text'], tweet['area'], tweet['date'], tweet['id']])

Or alternatively:

with codecs.open('Geotweets.csv',mode='w', encoding='utf-8',errors='replace') as cache:
    writer = csv.DictWriter(cache, ["text", "area", "date", "id"])
    for tweet in datares['results']:
        writer.writerow(tweet)

Obviously, you could also use writerows() to simplify this further:

with codecs.open('Geotweets.csv',mode='w', encoding='utf-8',errors='replace') as cache:
    writer = csv.DictWriter(cache, ["text", "area", "date", "id"])
    writer.writerows(datares['results'])