def parse_post_per_word_face(collect, parameters, RECENT, LIMIT, SKIP):
word = parameters[0]['$match'].pop('word')
output = []
# MongoDB find
db_cursor = collect.aggregate(parameters)
print('\nData acquired.\n')
for doc in db_cursor:
# print(doc)
text = doc['message']
if not text:
text = ''
tmp_text = filtered(text)
tmp_text = tmp_text.translate(punct_tab)
tmp_text = tmp_text.split(' ')
if any(filtered(word) == filtered(w) for w in tmp_text):
output.append(doc)
output = sorted(output, key=itemgetter('like_count'), reverse=True)
# +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
output = output[SKIP:SKIP + LIMIT] # pagination implementation
return output
def parse_post_per_word_no_rt(collect, parameters):
"""
Returns list of tweets per word.
"""
word = parameters[0]['$match'].pop('word')
output = []
# MongoDB find
db_cursor = collect.aggregate(parameters)
print('\nData acquired.\n')
for doc in db_cursor:
text = doc['status']['text']
tmp_text = filtered(text)
tmp_text = tmp_text.translate(punct_tab)
tmp_text = tmp_text.split(' ')
if any(filtered(word) == filtered(w) for w in tmp_text):
output.append(doc)
return output
def parse_word_face(collect, FILTER, SKIP, LIMIT, parameters, RECENT):
word_count = {}
top = []
# MongoDB find
db_cursor = collect.aggregate(parameters)
print('\nPosts Acquired.\n')
for doc in db_cursor:
text = doc['message']
if not text:
text = ''
tmp = clear_text(text)
# creates a word count
for word in tmp:
try:
word_count[filtered(word)] += 1
except KeyError:
word_count[filtered(word)] = 1
# CREATES LIST WITH COUNT
for word in word_count:
top.append([word, word_count[word]])
top = sorted(top, key=lambda x: x[1], reverse=True)
# +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
top = top[SKIP:SKIP + LIMIT] # pagination implementation
top_list = []
for item in top:
top_list.append({'word': item[0], 'count': item[1]})
return top_list
def parse_post_per_word(collect, FILTER, projection, SKIP,LIMIT, parameters, RECENT):
"""
Returns list of tweets per word.
"""
word = parameters[0]['$match'].pop('word')
output = []
# MongoDB aggreg
db_cursor = collect.aggregate(parameters)
print('\nRetweets acquired.\n')
for doc in db_cursor:
# print(doc)
# input()
text = doc['status']['retweeted_status']['text']
tmp_text = filtered(text)
tmp_text = tmp_text.translate(punct_tab)
tmp_text = tmp_text.split(' ')
if any(filtered(word)==filtered(w) for w in tmp_text):
output.append(doc)
# +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
# find for original tweets
if len(output) < LIMIT+SKIP and not RECENT:
FILTER['status.retweeted_status'] = {'$exists':False}
db_cursor = collect.find(FILTER,projection)
print('\nTweets acquired.\n')
for doc in db_cursor:
text = doc['status']['text']
tmp_text = filtered(text)
tmp_text = tmp_text.translate(punct_tab)
tmp_text = tmp_text.split(' ')
if any(filtered(word)==filtered(w) for w in tmp_text):
doc['count'] = 0
doc['status']['id'] = doc['status']['id_str']
output.append(doc)
db_cursor.close()
output = sorted(output, key=itemgetter('count'), reverse=True)
output = output[SKIP:SKIP+LIMIT] # pagination implementation
return output
def parse_word(collect, FILTER, projection, SKIP, LIMIT, parameters, RECENT):
"""
Returns list of top words plus count.
"""
word_count = {}
top = []
# ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
db_cursor = collect.aggregate(parameters)
print('\nRetweets\' texts acquired.\n')
for doc in db_cursor:
text = doc['retweeted_status']['text']
rt_count = doc['count']
tmp = clear_text(text)
# creates a word count
temp_words = []
for word in tmp:
if word not in temp_words:
temp_words.append(word)
try:
word_count[filtered(word)] += rt_count
except KeyError:
word_count[filtered(word)] = rt_count
else:
# count words only once
pass
if not RECENT:
# ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
# MongoDB find
FILTER['status.retweeted_status'] = {'$exists': False}
db_cursor = collect.find(FILTER, projection)
print('\nTweets\' texts acquired.\n')
for doc in db_cursor:
text = doc['status']['text']
tmp = clear_text(text)
temp_words = []
for word in tmp:
if word not in temp_words:
temp_words.append(word)
try:
word_count[filtered(word)] += 1
except KeyError:
word_count[filtered(word)] = 1
else:
# count words only once
pass
db_cursor.close()
# CREATES LIST WITH COUNT
for word in word_count:
tmp_tweets = []
top.append([word, word_count[word]])
top = sorted(top, key=lambda x: x[1], reverse=True)
# +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
top = top[SKIP:SKIP + LIMIT] # pagination implementation
top_list = []
for item in top:
top_list.append({'word': item[0], 'count': item[1]})
return top_list