Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -31,84 +31,92 @@ def generate_file_tree(paths):
|
|
| 31 |
def get_all_files(owner, repo, path="", is_hf=False):
|
| 32 |
"""Recursively fetch all files from a repository."""
|
| 33 |
if is_hf:
|
| 34 |
-
|
| 35 |
-
api_url = f"https://huggingface.co/spaces/{owner}/{repo}/tree/main/{path}".rstrip('/')
|
| 36 |
else:
|
| 37 |
api_url = f"{GITHUB_API}{owner}/{repo}/contents/{path}".rstrip('/')
|
| 38 |
|
| 39 |
try:
|
| 40 |
-
response = requests.get(api_url)
|
| 41 |
-
print(response.content)
|
| 42 |
response = requests.get(api_url, headers={"Accept": "application/json"})
|
| 43 |
-
print(str(response.json()))
|
| 44 |
response.raise_for_status()
|
| 45 |
-
items = response.json()
|
| 46 |
|
| 47 |
-
#
|
| 48 |
-
if
|
| 49 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
|
| 51 |
files = []
|
| 52 |
for item in items:
|
| 53 |
-
if item
|
| 54 |
files.append(item)
|
| 55 |
-
elif item
|
| 56 |
-
|
| 57 |
-
|
|
|
|
| 58 |
return files
|
| 59 |
|
| 60 |
-
except
|
|
|
|
| 61 |
return None
|
| 62 |
|
|
|
|
| 63 |
def get_hf_files(repo, name, path=""):
|
| 64 |
api = HfApi()
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
# Create base directory if it doesn't exist
|
| 70 |
-
if not os.path.exists(name):
|
| 71 |
-
os.makedirs(name)
|
| 72 |
-
|
| 73 |
-
for file_path in file_list:
|
| 74 |
-
# Handle nested directories
|
| 75 |
-
if "/" in file_path:
|
| 76 |
-
# Split into directory and remainder
|
| 77 |
-
dir_part, file_part = file_path.split("/", 1)
|
| 78 |
-
# Ensure directory exists
|
| 79 |
-
dir_path = os.path.join(name, dir_part)
|
| 80 |
-
if not os.path.exists(dir_path):
|
| 81 |
-
os.makedirs(dir_path)
|
| 82 |
-
# Recursively handle subdirectories if needed
|
| 83 |
-
if "/" in file_part:
|
| 84 |
-
processed_files.extend(get_hf_files(repo, name, dir_part))
|
| 85 |
-
continue
|
| 86 |
-
|
| 87 |
-
# Safely split filename into prefix and extension
|
| 88 |
-
filename = os.path.basename(file_path)
|
| 89 |
-
if "." in filename:
|
| 90 |
-
pf, sf = filename.rsplit(".", 1) # Safely split on last period
|
| 91 |
-
f_name = f"{pf}.{sf}"
|
| 92 |
-
else:
|
| 93 |
-
pf = filename
|
| 94 |
-
sf = ""
|
| 95 |
-
f_name = pf
|
| 96 |
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
# Ensure subdirectory exists for nested files
|
| 100 |
-
os.makedirs(os.path.dirname(local_path), exist_ok=True)
|
| 101 |
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
|
| 108 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 109 |
|
| 110 |
-
|
| 111 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 112 |
|
| 113 |
def get_repo_contents(url):
|
| 114 |
"""Parse URL and fetch repository contents."""
|
|
|
|
| 31 |
def get_all_files(owner, repo, path="", is_hf=False):
|
| 32 |
"""Recursively fetch all files from a repository."""
|
| 33 |
if is_hf:
|
| 34 |
+
api_url = f"https://huggingface.co/api/spaces/{owner}/{repo}/tree/main/{path}".rstrip('/')
|
|
|
|
| 35 |
else:
|
| 36 |
api_url = f"{GITHUB_API}{owner}/{repo}/contents/{path}".rstrip('/')
|
| 37 |
|
| 38 |
try:
|
|
|
|
|
|
|
| 39 |
response = requests.get(api_url, headers={"Accept": "application/json"})
|
|
|
|
| 40 |
response.raise_for_status()
|
|
|
|
| 41 |
|
| 42 |
+
# Check if the response is JSON
|
| 43 |
+
if response.headers.get('Content-Type', '').startswith('application/json'):
|
| 44 |
+
items = response.json()
|
| 45 |
+
else:
|
| 46 |
+
# If not JSON, it might be HTML (e.g., error page)
|
| 47 |
+
print(f"Received non-JSON response: {response.text[:100]}...")
|
| 48 |
+
return None
|
| 49 |
|
| 50 |
files = []
|
| 51 |
for item in items:
|
| 52 |
+
if isinstance(item, dict) and item.get('type') == 'file':
|
| 53 |
files.append(item)
|
| 54 |
+
elif isinstance(item, dict) and item.get('type') == 'dir':
|
| 55 |
+
sub_files = get_all_files(owner, repo, item['path'], is_hf)
|
| 56 |
+
if sub_files:
|
| 57 |
+
files.extend(sub_files)
|
| 58 |
return files
|
| 59 |
|
| 60 |
+
except requests.exceptions.RequestException as e:
|
| 61 |
+
print(f"Error fetching repository contents: {str(e)}")
|
| 62 |
return None
|
| 63 |
|
| 64 |
+
|
| 65 |
def get_hf_files(repo, name, path=""):
|
| 66 |
api = HfApi()
|
| 67 |
+
try:
|
| 68 |
+
file_list = api.list_repo_files(repo_id=f'{repo}/{name}', repo_type="space")
|
| 69 |
+
print(f"Files in {repo}/{name}: {file_list}")
|
| 70 |
+
processed_files = []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
|
| 72 |
+
if not os.path.exists(name):
|
| 73 |
+
os.makedirs(name)
|
|
|
|
|
|
|
| 74 |
|
| 75 |
+
for file_path in file_list:
|
| 76 |
+
# Handle nested directories
|
| 77 |
+
if "/" in file_path:
|
| 78 |
+
dir_part, file_part = file_path.split("/", 1)
|
| 79 |
+
dir_path = os.path.join(name, dir_part)
|
| 80 |
+
if not os.path.exists(dir_path):
|
| 81 |
+
os.makedirs(dir_path)
|
| 82 |
+
if "/" in file_part:
|
| 83 |
+
processed_files.extend(get_hf_files(repo, name, dir_part))
|
| 84 |
+
continue
|
| 85 |
+
|
| 86 |
+
# Fetch raw file content
|
| 87 |
+
raw_url = f"https://huggingface.co/spaces/{repo}/{name}/raw/main/{file_path}"
|
| 88 |
+
response = requests.get(raw_url, timeout=10)
|
| 89 |
+
response.raise_for_status()
|
| 90 |
+
|
| 91 |
+
# Ensure we get raw content, not HTML
|
| 92 |
+
if response.headers.get('Content-Type', '').startswith('text/html'):
|
| 93 |
+
print(f"Received HTML instead of raw content for {file_path}: {response.text[:100]}...")
|
| 94 |
+
continue
|
| 95 |
+
|
| 96 |
+
# Process file
|
| 97 |
+
filename = os.path.basename(file_path)
|
| 98 |
+
if "." in filename:
|
| 99 |
+
pf, sf = filename.rsplit(".", 1)
|
| 100 |
+
f_name = f"{pf}.{sf}"
|
| 101 |
+
else:
|
| 102 |
+
pf = filename
|
| 103 |
+
sf = ""
|
| 104 |
+
f_name = pf
|
| 105 |
+
|
| 106 |
+
local_path = os.path.join(name, file_path)
|
| 107 |
+
os.makedirs(os.path.dirname(local_path), exist_ok=True)
|
| 108 |
+
|
| 109 |
+
with open(local_path, 'wb') as file:
|
| 110 |
+
file.write(response.content)
|
| 111 |
+
|
| 112 |
+
processed_files.append({"path": file_path})
|
| 113 |
|
| 114 |
+
print(f"Processed files: {processed_files}")
|
| 115 |
+
return processed_files
|
| 116 |
+
|
| 117 |
+
except Exception as e:
|
| 118 |
+
print(f"Error processing Hugging Face files: {str(e)}")
|
| 119 |
+
return []
|
| 120 |
|
| 121 |
def get_repo_contents(url):
|
| 122 |
"""Parse URL and fetch repository contents."""
|