[Update] subtitle 메소드화

@@ -19,7 +19,9 @@
     <select />
     <select />
   </component>
   </component>
   <component name="ChangeListManager">
   <component name="ChangeListManager">
-    <list default="true" id="b9decb0c-dc9e-4239-bdad-09ea8dd5179d" name="Default Changelist" comment="" />
+    <list default="true" id="b9decb0c-dc9e-4239-bdad-09ea8dd5179d" name="Default Changelist" comment="">
+      <change beforePath="$PROJECT_DIR$/subtitle.py" beforeDir="false" afterPath="$PROJECT_DIR$/subtitle.py" afterDir="false" />
+    </list>
     <option name="SHOW_DIALOG" value="false" />
     <option name="SHOW_DIALOG" value="false" />
     <option name="HIGHLIGHT_CONFLICTS" value="true" />
     <option name="HIGHLIGHT_CONFLICTS" value="true" />
     <option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
     <option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />

@@ -11,47 +11,56 @@ def komoran_tokenize(sent):
     words = [w for w in words if ('/NN' in w or '/XR' in w or '/VA' in w or '/VV' in w)]
     words = [w for w in words if ('/NN' in w or '/XR' in w or '/VA' in w or '/VV' in w)]
     return words
     return words
 
 
-#youtube url의 자막 -> xml으로 가져오기
-topk_size = 30
-video_url = 'https://www.youtube.com/watch?v=ecUWKU_v318'
-yt = YouTube(video_url)
-title = yt.title
-description = yt.description
-caption = yt.captions.get_by_language_code('ko')
-caption_xml = caption.xml_captions
+def subtitle(video_url, topk_size):
+    #youtube url의 자막 -> xml으로 가져오기
+    topk_size = 30
+    video_url = 'https://www.youtube.com/watch?v=ecUWKU_v318'
+    yt = YouTube(video_url)
+    title = yt.title
+    description = yt.description
+    caption = yt.captions.get_by_language_code('ko')
+    caption_xml = caption.xml_captions
 
 
-#xml -> string list로 파싱
-root = ElementTree.fromstring(caption_xml)
-texts = []
-texts.append(title)
-for child in root.findall("text"):
+    #xml -> string list로 파싱
+    root = ElementTree.fromstring(caption_xml)
+    texts = []
+    texts.append(title)
+    for child in root.findall("text"):
         text = child.text.replace('\n', ' ')
         text = child.text.replace('\n', ' ')
         texts.append(text)
         texts.append(text)
-topk_size = texts.size() * 100 // topk_size
+    topk_size = texts.size() * 100 // topk_size
 
 
-#Komoran을 통해 형태소 단위로 분리 후 태깅
-komoran = Komoran('STABLE')
-sents = []
-for text in texts:
+    #Komoran을 통해 형태소 단위로 분리 후 태깅
+    komoran = Komoran('STABLE')
+    sents = []
+    for text in texts:
         tokened_text = komoran.get_plain_text(text)
         tokened_text = komoran.get_plain_text(text)
         sents.append(tokened_text)
         sents.append(tokened_text)
 
 
-keyword_extractor = KeywordSummarizer(
+    keyword_extractor = KeywordSummarizer(
         tokenize = komoran_tokenize,
         tokenize = komoran_tokenize,
         window = -1,
         window = -1,
         verbose = False
         verbose = False
-)
-keywords = keyword_extractor.summarize(sents, topk=30)
+    )
+    keywords = keyword_extractor.summarize(sents, topk=30)
 
 
-summarizer = KeysentenceSummarizer(
+    summarizer = KeysentenceSummarizer(
         tokenize = lambda x:x.split(),
         tokenize = lambda x:x.split(),
         min_sim = 0.5,
         min_sim = 0.5,
         verbose = False
         verbose = False
-)
-bias = np.ones(len(texts))
-bias[0] = 5
-keysents = summarizer.summarize(texts, topk=topk_size, bias=bias)
-keysents.sort(key=itemgetter(0))
-for _, _, sent in keysents:
+    )
+    bias = np.ones(len(texts))
+    bias[0] = 5
+    keysents = summarizer.summarize(texts, topk=topk_size, bias=bias)
+    keysents.sort(key=itemgetter(0))
+    first = True
+    ret = ''
+    for _, _, sent in keysents:
         sent = sent.replace(''', "'")
         sent = sent.replace(''', "'")
-    print(sent)
+        ret = ret + sent
+        if first:
+            ret += '\n'
+            first = False
+        else:
+            ret += ' '
+    return ret;