Upload app.py

app.py

@@ -5,6 +5,7 @@ import re
 import os
 import tempfile
 import shutil
+import urllib
 from huggingface_hub import whoami, HfApi, hf_hub_download
 from huggingface_hub.utils import build_hf_headers, hf_raise_for_status
 from gradio_huggingfacehub_search import HuggingfaceHubSearch
@@ -14,7 +15,7 @@ ENDPOINT = "https://huggingface.co"
 
 REPO_TYPES = ["model", "dataset", "space"]
 HF_REPO = os.environ.get("HF_REPO") if os.environ.get("HF_REPO") else "" # set your default repo
-
+REGEX_HF_REPO = r'^[\w_\-\.]+/[\w_\-\.]+$'
 
 def duplicate(source_repo, dst_repo, repo_type, private, overwrite, auto_dir, oauth_token: gr.OAuthToken | None, progress=gr.Progress(track_tqdm=True)):
     print(oauth_token.token)
@@ -29,7 +30,12 @@ def duplicate(source_repo, dst_repo, repo_type, private, overwrite, auto_dir, oa
         raise gr.Error(f"""Oops, you forgot to login. Please use the loggin button on the top left to migrate your repo {e}""")
     
     try:
-        if re.fullmatch(r'^[\w_\-\.]+/[\w_\-\.]+$', dst_repo): subfolder = ""
+        if re.fullmatch(REGEX_HF_REPO, source_repo): target = ""
+        else:
+            source_repo, target = re.findall(r'^(?:http.+\.co/)?(?:datasets)?(?:spaces)?([\w_\-\.]+/[\w_\-\.]+)/?(?:blob/main/)?(?:resolve/main/)?(.+)?$', source_repo)[0]
+            target = urllib.parse.unquote(target.removesuffix("/"))
+
+        if re.fullmatch(REGEX_HF_REPO, dst_repo): subfolder = ""
         else:
             dst_repo, subfolder = re.findall(r'^([\w_\-\.]+/[\w_\-\.]+)/?(.+)?$', dst_repo)[0]
             subfolder = subfolder.removesuffix("/")
@@ -39,6 +45,7 @@ def duplicate(source_repo, dst_repo, repo_type, private, overwrite, auto_dir, oa
             temp_dir = tempfile.mkdtemp()
             api.create_repo(repo_id=dst_repo, repo_type=repo_type, private=private, exist_ok=True, token=hf_token)
             for path in api.list_repo_files(repo_id=source_repo, repo_type=repo_type, token=hf_token):
+                if target and target not in path: continue
                 file = hf_hub_download(repo_id=source_repo, filename=path, repo_type=repo_type, local_dir=temp_dir, token=hf_token)
                 if not Path(file).exists(): continue
                 if Path(file).is_dir(): # unused for now