Update gen_dag.py

gen_dag.py

@@ -12,7 +12,7 @@ def load_config():
     """
     base_dir = os.path.dirname(os.path.abspath(__file__))
     
-    # 获取由 app.py 注入的专属输出文件夹，如果没有则回退到本地默认文件夹（方便本地调试）
+    # 获取由 app.py 注入的专属输出文件夹，如果没有则回退到本地默认文件夹（方便本地单独运行调试）
     user_root_folder = os.environ.get("USER_OUTPUT_DIR", os.path.join(base_dir, "mineru_outputs"))
     
     # 获取由 app.py 注入的 API 凭证
@@ -50,16 +50,18 @@ def load_prompt(prompt_path="prompt.json", prompt_name="poster_prompt"):
         data = json.load(f)
     return data.get(prompt_name, "")
 
-# 应用配置 (此时 config 已经在内存中组装好了)
+# 应用配置
 config = load_config()
+
 model_name = config['model_settings']['generation_model']
 
 # ========== 主流程 ==========
 def main():
-    # 这里的 root_folder 现在指向的是当前用户的专属 mineru_outputs 文件夹
+    # 输入总文件夹路径（包含多个论文子文件夹）此时指向用户的专属文件夹
     root_folder = config['path']['root_folder']  
     prompt_path = "prompt.json"
 
+
     print("📘 Loading prompts...")
     # dag prompt:
     section_split_prompt = load_prompt(prompt_path, prompt_name="section_split_prompt")
@@ -86,6 +88,9 @@ def main():
     add_title_and_hashtag_prompt = load_prompt(prompt_path, prompt_name="add_title_and_hashtag_prompt")
     pr_refinement_prompt = load_prompt(prompt_path, prompt_name="pr_refinement_prompt")
 
+
+
+
     # === 遍历每个子文件夹 ===
     for subdir in os.listdir(root_folder):
         subdir_path = os.path.join(root_folder, subdir)
@@ -124,64 +129,64 @@ def main():
         output_html = os.path.join(auto_path, "poster.html")
         graph_json_path = os.path.join(auto_path, "graph.json")
 
-        # # === 清理 markdown ===                                                   去除无意义的段落，如relative work，reference，appendix等等
-        # print("🧹 Cleaning markdown before splitting...")
-        # cleaned_md_path = clean_paper(md_path, clean_prompt, model="gemini-3-pro-preview", config=config)
+        # === 清理 markdown ===                                                   去除无意义的段落，如relative work，reference，appendix等等
+        print("🧹 Cleaning markdown before splitting...")
+        cleaned_md_path = clean_paper(md_path, clean_prompt, model="gemini-3-pro-preview", config=config)
 
-        # # === 利用gpt将论文分段 === 
-        # paths = split_paper(cleaned_md_path, section_split_prompt, model="gemini-3-pro-preview" ,config=config)
+        # === 利用gpt将论文分段 === 
+        paths = split_paper(cleaned_md_path, section_split_prompt, model="gemini-3-pro-preview" ,config=config)
         
-        # # === 利用gpt初始化dag === 
-        # dag = initialize_dag(markdown_path=cleaned_md_path,initialize_dag_prompt=initialize_dag_prompt,model="gemini-3-pro-preview", config=config)
+        # === 利用gpt初始化dag === 
+        dag = initialize_dag(markdown_path=cleaned_md_path,initialize_dag_prompt=initialize_dag_prompt,model="gemini-3-pro-preview", config=config)
         dag_path = os.path.join(auto_path, "dag.json")
 
 
-        # # === 生成visual_dag === 
-        # visual_dag_path=os.path.join(auto_path, "visual_dag.json")
-        # extract_and_generate_visual_dag(markdown_path=cleaned_md_path,prompt_for_gpt=visual_dag_prompt,output_json_path=visual_dag_path,model="gemini-3-pro-preview", config=config)
-        # add_resolution_to_visual_dag(auto_path, visual_dag_path)
+        # === 生成visual_dag === 
+        visual_dag_path=os.path.join(auto_path, "visual_dag.json")
+        extract_and_generate_visual_dag(markdown_path=cleaned_md_path,prompt_for_gpt=visual_dag_prompt,output_json_path=visual_dag_path,model="gemini-3-pro-preview", config=config)
+        add_resolution_to_visual_dag(auto_path, visual_dag_path)
 
-        # # === 生成section_dag ===
-        # section_split_output_path=os.path.join(subdir_path, "section_split_output")
-        # build_section_dags(folder_path=section_split_output_path,base_prompt=section_dag_generation_prompt,model="gemini-3-pro-preview", config=config)
+        # === 生成section_dag ===
+        section_split_output_path=os.path.join(subdir_path, "section_split_output")
+        build_section_dags(folder_path=section_split_output_path,base_prompt=section_dag_generation_prompt,model="gemini-3-pro-preview", config=config)
 
 
-        # # === 向dag.json添加section_dag ===
-        # section_dag_path=os.path.join(subdir_path, "section_dag")
-        # merged_path = add_section_dag(section_dag_folder=section_dag_path,main_dag_path=dag_path,output_path=None)
+        # === 向dag.json添加section_dag ===
+        section_dag_path=os.path.join(subdir_path, "section_dag")
+        merged_path = add_section_dag(section_dag_folder=section_dag_path,main_dag_path=dag_path,output_path=None)
 
-        # # === 向dag.json添加visual_dag ===
-        # add_visual_dag(dag_path=dag_path,visual_dag_path=visual_dag_path)
+        # === 向dag.json添加visual_dag ===
+        add_visual_dag(dag_path=dag_path,visual_dag_path=visual_dag_path)
 
-        # # ===  完善dag中每一个结点的visual_node ===
-        # refine_visual_node(dag_path)
+        # ===  完善dag中每一个结点的visual_node ===
+        refine_visual_node(dag_path)
 
 
 
 
-        # =============================  PPT部分  ================================
+        # # =============================  PPT部分  ================================
 
-        # ===  按照算法选出结点，以便后续生成outline ===
-        selected_node_path=os.path.join(auto_path, "selected_node.json")
-        generate_selected_nodes(dag_json_path=dag_path, max_len=15,output_path=selected_node_path)
+        # # ===  按照算法选出结点，以便后续生成outline ===
+        # selected_node_path=os.path.join(auto_path, "selected_node.json")
+        # generate_selected_nodes(dag_json_path=dag_path, max_len=15,output_path=selected_node_path)
 
-        # ===  初始化ouline ===
-        outline_path= os.path.join(auto_path, "outline.json")
-        outline = outline_initialize(dag_json_path=dag_path,outline_initialize_prompt=outline_initialize_prompt,model=model_name, config=config)
+        # # ===  初始化ouline ===
+        # outline_path= os.path.join(auto_path, "outline.json")
+        # outline = outline_initialize(dag_json_path=dag_path,outline_initialize_prompt=outline_initialize_prompt,model=model_name, config=config)
         
-        # ===  生成完整ouline ===
-        outline_data=generate_complete_outline(selected_node_path,outline_path,generate_complete_outline_prompt,model=model_name, config=config)
+        # # ===  生成完整ouline ===
+        # outline_data=generate_complete_outline(selected_node_path,outline_path,generate_complete_outline_prompt,model=model_name, config=config)
         
-        # ===  配模板 ===
-        arrange_template(outline_path,arrange_template_prompt,model=model_name, config=config)
+        # # ===  配模板 ===
+        # arrange_template(outline_path,arrange_template_prompt,model=model_name, config=config)
         
-        # ===  生成最终的PPT ===
-        ppt_template_path="./ppt_template"
-        generate_ppt_prompt = {"role":"system","content":"You are given (1) a slide node (JSON) and (2) an HTML slide template. Your task: revise the HTML template to produce the final slide HTML using the node content. Node fields: text (slide textual content), figure (list of images to display, each has name, caption, resolution), formula (list of formula images to display, each has name, caption, resolution), template (template filename). IMPORTANT RULES: 1) Only modify places in the HTML that are marked by comments like . 2) For 'Subjects' sections: replace the placeholder title with ONE concise summary sentence for this slide. 3) For 'Image' sections (<img ...>): replace src with the relative path extracted from node.figure/formula[i].name; the node image name may be markdown like '![](images/abc.jpg)', use only 'images/abc.jpg'. 4) For 'Text' sections: replace the placeholder text with the node.text content, formatted cleanly in HTML; keep it readable and you may use <p>, <ul><li>, <br/> appropriately. 5) If the template expects more images/text blocks than provided by the node, leave the missing positions unchanged and do not invent content. 6) If the node provides more images than the template has slots, fill slots in order and ignore the rest. 7) Preserve all other HTML, CSS, and structure exactly. OUTPUT FORMAT: Return ONLY the revised HTML as plain text. Do NOT wrap it in markdown fences. Do NOT add explanations."}
-        generate_ppt(outline_path,ppt_template_path,generate_ppt_prompt,model=model_name, config=config)
+        # # ===  生成最终的PPT ===
+        # ppt_template_path="./ppt_template"
+        # generate_ppt_prompt = {"role":"system","content":"You are given (1) a slide node (JSON) and (2) an HTML slide template. Your task: revise the HTML template to produce the final slide HTML using the node content. Node fields: text (slide textual content), figure (list of images to display, each has name, caption, resolution), formula (list of formula images to display, each has name, caption, resolution), template (template filename). IMPORTANT RULES: 1) Only modify places in the HTML that are marked by comments like . 2) For 'Subjects' sections: replace the placeholder title with ONE concise summary sentence for this slide. 3) For 'Image' sections (<img ...>): replace src with the relative path extracted from node.figure/formula[i].name; the node image name may be markdown like '![](images/abc.jpg)', use only 'images/abc.jpg'. 4) For 'Text' sections: replace the placeholder text with the node.text content, formatted cleanly in HTML; keep it readable and you may use <p>, <ul><li>, <br/> appropriately. 5) If the template expects more images/text blocks than provided by the node, leave the missing positions unchanged and do not invent content. 6) If the node provides more images than the template has slots, fill slots in order and ignore the rest. 7) Preserve all other HTML, CSS, and structure exactly. OUTPUT FORMAT: Return ONLY the revised HTML as plain text. Do NOT wrap it in markdown fences. Do NOT add explanations."}
+        # generate_ppt(outline_path,ppt_template_path,generate_ppt_prompt,model=model_name, config=config)
 
-        # ===  Refiner  ===                
-        refinement_ppt(input_index=auto_path, prompts=[commenter_prompt, reviser_prompt], model=model_name, max_iterations=3, config=config)
+        # # ===  Refiner  ===                
+        # refinement_ppt(input_index=auto_path, prompts=[commenter_prompt, reviser_prompt], model=model_name, max_iterations=3, config=config)
 
         # # =============================  Poster部分  ================================
         
@@ -262,7 +267,7 @@ def main():
 
         # =============================================================
         # 在auto目录下创建success.txt作为标志
-        success_file_path = os.path.join(auto_path, "success_ppt.txt")
+        success_file_path = os.path.join(auto_path, "success_dag.txt")
         open(success_file_path, "w").close()