Add: source lib 추가

-# OS Generated files
-*.DS_Store
-*.DS_Store?
-._*
\ No newline at end of file
+**/__pycache__

+from nltk.tokenize import word_tokenize
+import nltk
+import re
+from bs4 import BeautifulSoup
+import requests
+
+
+def get_HTML_from_url(url):
+    return requests.get(url).text
+
+
+def get_text_from_HTML(html):
+    soup = BeautifulSoup(html)
+    script_tag = soup.find_all(['script', 'style', 'header', 'footer', 'form'])
+
+    for script in script_tag:
+        script.extract()
+    content = soup.get_text('\n', strip=True)
+    return content
+
+
+# def get_HTML_from_regexp_url(url_pattern):
+
+
+def is_string(target):
+    return type(target) == str
+
+
+def cut_corpus(corpus):
+    if not is_string(corpus):
+        return []
+    return corpus.split('.')[:-1]
+
+
+def postag_sentence(sentence):
+    if not is_string(sentence):
+        return []
+    tags = word_tokenize(sentence)
+    return nltk.pos_tag(tags)
+
+
+# verb의 index를 return 합니다.
+# 만약, 존재하지 않는다면, -1을 return 합니다.
+def find_verb_idx(tags):
+    idx = 0
+    for tag in tags:
+        if tag[0][1] == 'V':
+            return idx
+    return -1
+
+
+def make_be_verb(subj):
+    if subj == 'I':
+        return 'am'
+    elif subj in ['You', 'you']:
+        return 'are'
+    else:
+        return 'is'
+
+
+def cut_quot(sentence):
+    return re.sub("[\'\"\`]", '', sentence)
+
+
+# 예외
+# 1. brace가 닫히지 않음
+# 2. target_str가 없음
+def make_brace_triple(target_str, brace_tags):
+    if target_str == '':
+        return []
+    idx = find_verb_idx(brace_tags)
+    subj = target_str
+    pred = ''
+    if idx != -1:
+        pred = brace_tags[idx]
+        obj = ' '.join([value for value, _ in brace_tags[idx:]])
+    else:
+        pred = make_be_verb(subj)
+        obj = ' '.join([value for value, _ in brace_tags])
+    return [subj, pred, obj]