You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
79 lines
2.4 KiB
79 lines
2.4 KiB
import sys, os
|
|
from nltk import sent_tokenize, word_tokenize
|
|
from nltk import everygrams
|
|
from nltk import FreqDist
|
|
import json
|
|
import re
|
|
|
|
"""
|
|
PART 1
|
|
We create the dictionary and save it.
|
|
"""
|
|
|
|
stopws = [",", ".", "?","!",":","(",")",">","<","@","#","``","/","–","''","‘","-","’", "DOCTYPE", "html", "!", "'", "<br>", "<br />", "/body", "/html", "/head", "h2", "/h2", "h1", "/h1","”","“", "o", "Ca", "/"]
|
|
|
|
path = "static/files/"
|
|
for path, subdirs, files in os.walk(path):
|
|
for name in files:
|
|
if name.endswith('html'):
|
|
file = os.path.join(path, name)
|
|
total = open("allhtml.txt", "a")
|
|
with open(file, 'r+') as f:
|
|
content = f.read()
|
|
total.write(content)
|
|
total.close()
|
|
|
|
keyword_list = []
|
|
|
|
|
|
with open('allhtml.txt') as f:
|
|
content = f.read()
|
|
tokens = word_tokenize(content)
|
|
tokens = [token for token in tokens if token not in stopws]
|
|
keyword_list = list(set(tokens))
|
|
# print(tokens)
|
|
# print(keyword_list)
|
|
|
|
"""
|
|
PART 2
|
|
We iterate through the entire collection of html files, tokenize the words, and check to see whether any of them is in the keyword_list. If they are, then we generate a json file.
|
|
"""
|
|
|
|
# wordlist = {}
|
|
# avoiding_repetition = []
|
|
|
|
|
|
sentences_w_word = {}
|
|
|
|
def analysis(the_word, file_name):
|
|
id = file_name[13:15]
|
|
with open(file_name, 'r+') as f:
|
|
content = f.read()
|
|
sent_tokens = sent_tokenize(content)
|
|
new_sent_tokens = []
|
|
# the_word = the_word.lower()
|
|
re_word = r"\b" + re.escape(the_word) + r"\b"
|
|
# print(re_word)
|
|
# print(the_word)
|
|
for sent_token in sent_tokens:
|
|
if re.search(re_word, sent_token):
|
|
new_sent_tokens.append({'id': id, 'sentence': sent_token.replace('\n', ' ').strip("'<>()“”")})
|
|
if the_word in sentences_w_word: # if this is not the first iteration
|
|
previous_sent_tokens = sentences_w_word[the_word]
|
|
full_sent_tokens = previous_sent_tokens + new_sent_tokens
|
|
else:
|
|
full_sent_tokens = new_sent_tokens
|
|
sentences_w_word[the_word] = full_sent_tokens
|
|
|
|
# maybe ISO-8859-1 instead of utf8??
|
|
|
|
path = "static/files/"
|
|
for path, subdirs, files in os.walk(path):
|
|
for name in files:
|
|
if name.endswith('html'):
|
|
file = os.path.join(path, name)
|
|
for word in keyword_list:
|
|
analysis(word, file)
|
|
|
|
with open('wordlist.json', 'w') as outfile:
|
|
json.dump(sentences_w_word, outfile, ensure_ascii=False)
|
|
|