refactor: Enhance FFMPEG command generation with step-by-step thinking and improved command extraction

app.py

@@ -132,6 +132,7 @@ You are given:
 Your objective is to generate the SIMPLEST POSSIBLE single ffmpeg command to create the requested video.
 
 Key requirements:
+    - First, think step-by-step about what the user is asking for and reformulate it into a clear technical specification
     - Use the absolute minimum number of ffmpeg options needed
     - Avoid complex filter chains or filter_complex if possible
     - Prefer simple concatenation, scaling, and basic filters
@@ -147,15 +148,19 @@ Remember: Simpler is better. Only use advanced ffmpeg features if absolutely nec
         },
         {
             "role": "user",
-            "content": f"""Always output the media as video/mp4 and output file with "output.mp4". Provide only the shell command without any explanations.
-The current assets and objective follow. Reply with the FFMPEG command:
+            "content": f"""Always output the media as video/mp4 and output file with "output.mp4".
+The current assets and objective follow.
 
 AVAILABLE ASSETS LIST:
 
 {files_info_string}
 
 OBJECTIVE: {prompt} and output at "output.mp4"
-YOUR FFMPEG COMMAND:
+
+First, think step-by-step about what I'm asking for and reformulate it into a clear technical specification.
+Then provide the FFMPEG command that will accomplish this task.
+
+YOUR RESPONSE:
          """,
         },
     ]
@@ -188,13 +193,23 @@ YOUR FFMPEG COMMAND:
             # Find content between ```sh or ```bash and the next ```
             import re
 
-            command = re.search(r"```(?:sh|bash)?\n(.*?)\n```", content, re.DOTALL)
-            if command:
-                command = command.group(1).strip()
+            command_match = re.search(r"```(?:sh|bash)?\n(.*?)\n```", content, re.DOTALL)
+            if command_match:
+                command = command_match.group(1).strip()
             else:
-                command = content.replace("\n", "")
+                # Try to find a line that starts with ffmpeg
+                ffmpeg_lines = [line.strip() for line in content.split('\n') if line.strip().startswith('ffmpeg')]
+                if ffmpeg_lines:
+                    command = ffmpeg_lines[0]
+                else:
+                    command = content.replace("\n", "")
         else:
-            command = content.replace("\n", "")
+            # Try to find a line that starts with ffmpeg
+            ffmpeg_lines = [line.strip() for line in content.split('\n') if line.strip().startswith('ffmpeg')]
+            if ffmpeg_lines:
+                command = ffmpeg_lines[0]
+            else:
+                command = content.replace("\n", "")
 
         # remove output.mp4 with the actual output file path
         command = command.replace("output.mp4", "")
@@ -270,7 +285,11 @@ def update(
                 f"\n=== EXECUTING FFMPEG COMMAND ===\nffmpeg {' '.join(final_command[1:])}\n"
             )
             subprocess.run(final_command, cwd=temp_dir)
-            generated_command = f"### Generated Command\n```bash\nffmpeg {' '.join(args[1:])} -y output.mp4\n```"
+            # Store the full model response for display
+            model_response = content.strip()
+            # Extract just the command for display
+            command_for_display = f"ffmpeg {' '.join(args[1:])} -y output.mp4"
+            generated_command = f"### Model Analysis\n{model_response}\n\n### Generated Command\n```bash\n{command_for_display}\n```"
             return output_file_path, gr.update(value=generated_command)
         except Exception as e:
             attempts += 1