Update app.py

app.py

@@ -12,36 +12,6 @@ logging.basicConfig(filename='youtube_script_extractor.log', level=logging.DEBUG
 
 openai.api_key = os.getenv("OPENAI_API_KEY")
 
-# 문장 구분 함수
-def split_sentences(text):
-    # 주어진 구분 기준을 사용한 정규 표현식
-    sentence_end = r'(니다|에요|구나|해요|군요|겠어요|시오|해라|예요|아요|데요|대요|세요|어요|게요|구요|고요|나요|하죠)(?![\w])'
-    segments = re.split(f'({sentence_end})', text)
-    
-    combined_sentences = []
-    current_sentence = ""
-    
-    for i in range(0, len(segments), 2):
-        segment = segments[i]
-        if i + 1 < len(segments):
-            segment += segments[i + 1]
-        
-        if len(current_sentence) + len(segment) > 100:
-            if current_sentence:
-                combined_sentences.append(current_sentence.strip())
-            current_sentence = segment.strip()
-        else:
-            current_sentence += (' ' if current_sentence else '') + segment.strip()
-        
-        if re.search(sentence_end, segment):
-            combined_sentences.append(current_sentence.strip())
-            current_sentence = ""
-    
-    if current_sentence:
-        combined_sentences.append(current_sentence.strip())
-    
-    return combined_sentences
-
 def parse_api_response(response):
     try:
         if isinstance(response, str):
@@ -126,8 +96,8 @@ def analyze(url, progress=gr.Progress()):
         script_sentences = split_sentences(script)
         script_content = f"# {title}\n\n" + "\n".join(script_sentences)
         
-        # 원문 스크립트 먼저 출력
-        yield script_content, None
+        # 원문 스크립트 먼저 출력하고 진행
+        yield script_content, "요약 준비 중..."
         
         progress(0.5, desc="요약 생성 중...")
         summary = summarize_text(title, description, "\n".join(script_sentences))
@@ -142,6 +112,34 @@ def analyze(url, progress=gr.Progress()):
         logging.exception(error_msg)
         yield error_msg, error_msg
 
+def split_sentences(text):
+    sentence_end = r'(니다|에요|구나|해요|군요|겠어요|시오|해라|예요|아요|데요|대요|세요|어요|게요|구요|고요|나요|하죠)(?![\w])'
+    segments = re.split(f'({sentence_end})', text)
+    
+    combined_sentences = []
+    current_sentence = ""
+    
+    for i in range(0, len(segments), 2):
+        segment = segments[i]
+        if i + 1 < len(segments):
+            segment += segments[i + 1]
+        
+        if len(current_sentence) + len(segment) > 100:
+            if current_sentence:
+                combined_sentences.append(current_sentence.strip())
+            current_sentence = segment.strip()
+        else:
+            current_sentence += (' ' if current_sentence else '') + segment.strip()
+        
+        if re.search(sentence_end, segment):
+            combined_sentences.append(current_sentence.strip())
+            current_sentence = ""
+    
+    if current_sentence:
+        combined_sentences.append(current_sentence.strip())
+    
+    return combined_sentences
+
 # Gradio 인터페이스
 with gr.Blocks() as demo:
     gr.Markdown("## YouTube 스크립트 추출 및 요약 도구")
@@ -157,7 +155,8 @@ with gr.Blocks() as demo:
     analyze_button.click(
         analyze,
         inputs=[youtube_url_input],
-        outputs=[script_output, summary_output]
+        outputs=[script_output, summary_output],
+        show_progress=True
     )
 
 if __name__ == "__main__":