Allow Melody in MCP call

app.py

@@ -33,7 +33,7 @@ import librosa
 import modules.user_history
 from modules.version_info import versions_html, commit_hash, get_xformers_version
 from modules.gradio import *
-from modules.file_utils import get_file_parts, get_filename_from_filepath, convert_title_to_filename, get_unique_file_path, delete_file, download_and_save_image
+from modules.file_utils import get_file_parts, get_filename_from_filepath, convert_title_to_filename, get_unique_file_path, delete_file, download_and_save_image, download_and_save_file
 from modules.constants import IS_SHARED_SPACE, HF_REPO_ID, TMPDIR, HF_API_TOKEN
 from modules.storage import upload_files_to_repo
 
@@ -294,11 +294,17 @@ def predict(model, text, melody_filepath = None, duration=10, dimension=2, topk=
     if melody_filepath in ["None", ""]:
         melody_filepath = None
 
+    
+    # if melody_filepath is a url string, download it using download_and_save_file
+    if melody_filepath and melody_filepath.startswith(("http://", "https://")):
+        username = profile if isinstance(profile, str) else profile.value.username if hasattr(profile.value, 'username') else "default_user" if (profile is None) else profile
+        melody_filepath = download_and_save_file(melody_filepath, Path(TMPDIR) / str(username), HF_API_TOKEN)
+
     #if background is a url string, download it using download_and_save_image
     if background is None or background in ["None", ""]:
         background = load_background_filepath(video_orientation)
 
-    if background.startswith("http://") or background.startswith("https://"):
+    if background.startswith(("http://", "https://")):
         username = profile if isinstance(profile, str) else profile.value.username if hasattr(profile.value, 'username') else "default_user" if (profile is None) else profile
         background = download_and_save_image(background, Path(TMPDIR) / str(username), HF_API_TOKEN)
     
@@ -615,13 +621,14 @@ def fix_path(path: str) -> str:
         return "./" + path[index:].replace("\\", "/")
     return path
 # Add this wrapper function above the gr.api definitions
-def predict_simple(model: str, text: str, duration: int = 10, dimension: int = 2, topk: int = 200, topp: float = 0.01, temperature: float = 1.0, cfg_coef: float = 4.0, background: str = "./assets/background.png", title: str = "UnlimitedMusicGen", settings_font: str = "./assets/arial.ttf", settings_font_color: str = "#c87f05", seed: int = -1, overlap: int = 1, prompt_index: int = -1, include_title: bool = True, include_settings: bool = True, profile: str = "Satoshi Nakamoto", segment_length: int = 30, settings_font_size: int = 28, settings_animate_waveform: bool = False, video_orientation: str = "Landscape",  return_history_json: bool = False) -> tp.List[tp.Tuple[str, str, str]]:
+def predict_simple(model: str, text: str,  melody_filepath: str = None, duration: int = 10, dimension: int = 2, topk: int = 200, topp: float = 0.01, temperature: float = 1.0, cfg_coef: float = 4.0, background: str = "./assets/background.png", title: str = "UnlimitedMusicGen", settings_font: str = "./assets/arial.ttf", settings_font_color: str = "#c87f05", seed: int = -1, overlap: int = 1, prompt_index: int = -1, include_title: bool = True, include_settings: bool = True, harmony_only: bool = False, profile: str = "Satoshi Nakamoto", segment_length: int = 30, settings_font_size: int = 28, settings_animate_waveform: bool = False, video_orientation: str = "Landscape",  return_history_json: bool = False) -> tp.List[tp.Tuple[str, str, str]]:
     """
         Generate music and video based on the provided parameters and model.
 
         Args:
             model (str): Model name to use for generation.
             text (str): Prompt describing the music.
+            melody_filepath (str, optional): Path to melody conditioning file. Default to None.
             duration (int): Total duration in seconds.
             dimension (int): Audio stacking/concatenation dimension. 
             topk (int): Top-k sampling value.
@@ -637,6 +644,7 @@ def predict_simple(model: str, text: str, duration: int = 10, dimension: int = 2
             prompt_index (int, optional): Melody segment index. Default to -1.
             include_title (bool, optional): Whether to add title to video. Default to True.
             include_settings (bool, optional): Whether to add settings to video. Default to True.
+            harmony_only (bool, optional): Whether to use harmony only. Default to False.
             profile (str, optional): User profile.
             segment_length (int, optional): Segment length in seconds.
             settings_font_size (int, optional): Font size for settings text.
@@ -664,7 +672,7 @@ def predict_simple(model: str, text: str, duration: int = 10, dimension: int = 2
         elif isinstance(actual_profile_data, str) and actual_profile_data: # string username
             profile_username_to_send = actual_profile_data
 
-    UMG_result = predict(model, text, melody_filepath=None, duration=duration, dimension=dimension, topk=topk, topp=topp, temperature=temperature, cfg_coef=cfg_coef, background=background, title=title, settings_font=settings_font, settings_font_color=settings_font_color, seed=seed, overlap=overlap, prompt_index=prompt_index, include_title=include_title, include_settings=include_settings, harmony_only=False, profile=profile, segment_length=segment_length, settings_font_size=settings_font_size, settings_animate_waveform=settings_animate_waveform, video_orientation=video_orientation, excerpt_duration=3.5,  return_history_json=return_history_json)
+    UMG_result = predict(model, text, melody_filepath=melody_filepath, duration=duration, dimension=dimension, topk=topk, topp=topp, temperature=temperature, cfg_coef=cfg_coef, background=background, title=title, settings_font=settings_font, settings_font_color=settings_font_color, seed=seed, overlap=overlap, prompt_index=prompt_index, include_title=include_title, include_settings=include_settings, harmony_only=harmony_only, profile=profile, segment_length=segment_length, settings_font_size=settings_font_size, settings_animate_waveform=settings_animate_waveform, video_orientation=video_orientation, excerpt_duration=3.5,  return_history_json=return_history_json)
 
     # upload to storage and return urls
     folder_name = f"user_uploads/{convert_title_to_filename(profile_username_to_send)}/{convert_title_to_filename(title)}"
@@ -868,6 +876,7 @@ def ui(**kwargs):
         launch_kwargs['allowed_paths'] = ["assets", "./assets", "images", "./images", 'e:/TMP']
         launch_kwargs['favicon_path'] = "./assets/favicon.ico"
         launch_kwargs['mcp_server'] = True
+        launch_kwargs['ssr_mode'] = False
 
         gr.api(ping, api_name="ping")
         gr.api(predict_simple)

modules/file_utils.py

@@ -153,4 +153,37 @@ def download_and_save_image(url: str, dst_folder: Path, token: str = None) -> Pa
     dst = Path(unique_filepath_str)
     dst_folder.mkdir(parents=True, exist_ok=True)
     pil_image.save(dst)
+    return dst
+
+def download_and_save_file(url: str, dst_folder: Path, token: str = None) -> Path:
+    """
+    Downloads a binary file (e.g., audio or video) from a URL with authentication if a token is provided,
+    and saves it in dst_folder with a unique filename.
+
+    Args:
+        url (str): The file URL.
+        dst_folder (Path): The destination folder for the file.
+        token (str, optional): A valid Bearer token.
+
+    Returns:
+        Path: The saved file's path.
+    """
+    headers = {}
+    if token:
+        headers["Authorization"] = f"Bearer {token}"
+    
+    response = requests.get(url, headers=headers)
+    response.raise_for_status()
+    
+    parsed_url = urlparse(url)
+    original_filename = os.path.basename(parsed_url.path)
+    base, ext = os.path.splitext(original_filename)
+    
+    unique_filepath_str = get_unique_file_path(str(dst_folder), base, ext)
+    dst = Path(unique_filepath_str)
+    dst_folder.mkdir(parents=True, exist_ok=True)
+    
+    with open(dst, "wb") as f:
+        f.write(response.content)
+    
     return dst

requirements.txt

@@ -38,4 +38,6 @@ spacy==3.7.6
 sentencepiece
 num2words
 numpy<1.26.4
-matplotlib
+matplotlib
+fastapi>=0.115.0
+starlette>=0.40.0