Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -52,6 +52,55 @@ def get_all_files(owner, repo, path="", is_hf=False):
|
|
| 52 |
except Exception as e:
|
| 53 |
return None
|
| 54 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
def get_repo_contents(url):
|
| 56 |
"""Parse URL and fetch repository contents."""
|
| 57 |
try:
|
|
@@ -67,7 +116,7 @@ def get_repo_contents(url):
|
|
| 67 |
# Add more known paths or implement HTML scraping if needed
|
| 68 |
]
|
| 69 |
#files = get_all_files(owner, repo, "", True) or known_files
|
| 70 |
-
files =
|
| 71 |
return owner, repo, files, True
|
| 72 |
else: # Assume GitHub URL
|
| 73 |
parts = url.rstrip('/').split('/')
|
|
|
|
| 52 |
except Exception as e:
|
| 53 |
return None
|
| 54 |
|
| 55 |
+
def get_hf_files(repo,name):
|
| 56 |
+
api = HfApi()
|
| 57 |
+
f_ist = (api.list_repo_files(repo_id=f'{repo}/{name}', repo_type="space"))
|
| 58 |
+
print (f_ist)
|
| 59 |
+
file_list = []
|
| 60 |
+
#file_out = []
|
| 61 |
+
|
| 62 |
+
if not os.path.exists(name):
|
| 63 |
+
os.makedirs(name)
|
| 64 |
+
for d_app in f_ist:
|
| 65 |
+
if "/" in d_app:
|
| 66 |
+
dir_1=d_app.split("/",1)[0]
|
| 67 |
+
rem_1=d_app.split("/",1)[1]
|
| 68 |
+
if not os.path.exists(f'{name}/{dir_1}'):
|
| 69 |
+
os.makedirs(f'{name}/{dir_1}')
|
| 70 |
+
|
| 71 |
+
if "/" in rem_1:
|
| 72 |
+
dir_2=rem_1.split("/",1)[0]
|
| 73 |
+
rem_2=rem_1.split("/",1)[1]
|
| 74 |
+
if not os.path.exists(f'{name}/{dir_1}/{dir_2}'):
|
| 75 |
+
os.makedirs(f'{name}/{dir_1}/{dir_2}')
|
| 76 |
+
sf=rem_2.split(".",1)[1]
|
| 77 |
+
pf=rem_2.split(".",1)[0]
|
| 78 |
+
f_name=f'{dir_1}/{dir2}/{pf}.{sf}'
|
| 79 |
+
|
| 80 |
+
else:
|
| 81 |
+
sf=rem_1.split(".",1)[1]
|
| 82 |
+
pf=rem_1.split(".",1)[0]
|
| 83 |
+
f_name=f'{dir_1}/{pf}.{sf}'
|
| 84 |
+
print(f_name)
|
| 85 |
+
|
| 86 |
+
else:
|
| 87 |
+
|
| 88 |
+
sf=d_app.split(".",1)[1]
|
| 89 |
+
pf=d_app.split(".",1)[0]
|
| 90 |
+
f_name=f'{pf}.{sf}'
|
| 91 |
+
pass
|
| 92 |
+
|
| 93 |
+
r = requests.get(f'https://huggingface.co/spaces/{repo}/{name}/raw/main/{d_app}')
|
| 94 |
+
print(d_app)
|
| 95 |
+
#print (r.text)
|
| 96 |
+
uid = uuid.uuid4()
|
| 97 |
+
file = open(f'{name}/{f_name}','wb')
|
| 98 |
+
file.write(r.content)
|
| 99 |
+
file.close()
|
| 100 |
+
file_list.append(Path(f'{name}/{f_name}'))
|
| 101 |
+
print(file_list)
|
| 102 |
+
return file_list
|
| 103 |
+
|
| 104 |
def get_repo_contents(url):
|
| 105 |
"""Parse URL and fetch repository contents."""
|
| 106 |
try:
|
|
|
|
| 116 |
# Add more known paths or implement HTML scraping if needed
|
| 117 |
]
|
| 118 |
#files = get_all_files(owner, repo, "", True) or known_files
|
| 119 |
+
files = get_hf_files(owner, repo)
|
| 120 |
return owner, repo, files, True
|
| 121 |
else: # Assume GitHub URL
|
| 122 |
parts = url.rstrip('/').split('/')
|