def count_word_occurance():
if request.method == 'POST':
session_id = request.args.get("session_id")
word = request.form['word']
word_stem = indeed_scrape.stemmer.stem(word)
sess_dict = get_sess(session_id)
df = load_csv(session_id)
count = 0
for doc in df['summary_stem']:
tokens = np.array(indeed_scrape.toker(doc))
test = np.where(tokens == word_stem)[0].shape[0]
if test > 0:
count += 1
number = df.shape[0]
percent = 100 * np.round(float(count) / number, decimals=1)
percent = str(percent)
session_string = "?session_id=%s" % session_id
return render_template('word_count.html',
count=count,
session_id=session_string,
word=word,
number=number,
percent=percent)
if request.method == 'GET':
session_id = request.args.get("session_id")
session_string = "?session_id=%s" % session_id
return render_template('word_count.html', session_id=session_string)
def normalize_titles(self, titles):
ind = indeed_scrape.Indeed("kw")
titles = map(lambda x: x.replace("/", " "), titles)
titles = map(lambda x: x.replace("-", " "), titles)
titles = map(lambda x: re.sub("^\s+", "", x), titles)
titles = map(lambda x: re.sub("$\s+", "", x), titles)
titles = map(lambda x: re.sub("\s{2,}", " ", x), titles)
out = []
for title in titles:
#temp_list = ind._decode(title)
#if temp_list is None:
#continue
temp_list = indeed_scrape.toker(title)
temp_list = ind.len_tester(temp_list)
temp_string = " ".join(temp_list)
temp_string = re.sub('\s+amp\s+', ' ', temp_string)
temp_string = re.sub('\s+and\s+', ' ', temp_string)
out.append(temp_string)
out = map(lambda x: x.lower(), out)
return out
