Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -56,56 +56,6 @@ def create_excel_if_missing(filename: str):
|
|
| 56 |
print(f"Created empty Excel file: {filename}")
|
| 57 |
|
| 58 |
|
| 59 |
-
# def save_to_excel(df, user_state):
|
| 60 |
-
# """
|
| 61 |
-
# Saves the user's data into a unique Excel file named:
|
| 62 |
-
# Cond3_<ProlificID>_<timestamp>.xlsx
|
| 63 |
-
# If a file with that name already exists (e.g., user
|
| 64 |
-
# revisits or calls this function multiple times quickly),
|
| 65 |
-
# we merge the new data with the existing file.
|
| 66 |
-
# Afterwards, uploads the file to Hugging Face under the same unique name.
|
| 67 |
-
# """
|
| 68 |
-
# hf_token = os.getenv("tiliah_write3")
|
| 69 |
-
|
| 70 |
-
# dataset_repo_id = "Tiliah/user_data"
|
| 71 |
-
|
| 72 |
-
# # 1) Generate a unique filename from Prolific ID and timestamp
|
| 73 |
-
# local_file = user_state.get("excel_file")
|
| 74 |
-
|
| 75 |
-
# try:
|
| 76 |
-
# with file_lock:
|
| 77 |
-
# # 2) Merge existing data if the same file name exists
|
| 78 |
-
# if os.path.exists(local_file):
|
| 79 |
-
# existing = pd.read_excel(local_file, engine="openpyxl")
|
| 80 |
-
# merged = pd.concat([existing, df], ignore_index=True)
|
| 81 |
-
# else:
|
| 82 |
-
# merged = df
|
| 83 |
-
|
| 84 |
-
# # 3) Save locally
|
| 85 |
-
# merged.to_excel(local_file, index=False, engine="openpyxl")
|
| 86 |
-
|
| 87 |
-
# # 4) Upload to Hugging Face
|
| 88 |
-
# if hf_token and dataset_repo_id:
|
| 89 |
-
# api = HfApi()
|
| 90 |
-
# path_in_repo = f"Cond3/{local_file}"
|
| 91 |
-
# future_upload = api.upload_file(
|
| 92 |
-
# path_or_fileobj=local_file,
|
| 93 |
-
# path_in_repo=path_in_repo, # Keep the same name in the repo
|
| 94 |
-
# repo_id=dataset_repo_id,
|
| 95 |
-
# repo_type="dataset",
|
| 96 |
-
# token=hf_token,
|
| 97 |
-
# run_as_future=True
|
| 98 |
-
# )
|
| 99 |
-
# future_upload
|
| 100 |
-
# future_upload.result()
|
| 101 |
-
# print(f"Uploaded {local_file} to HF dataset.")
|
| 102 |
-
# else:
|
| 103 |
-
# print("No HF token or dataset repo provided, skipping HF upload.")
|
| 104 |
-
|
| 105 |
-
# except Exception as e:
|
| 106 |
-
# print(f"Error in save_cond3_data_in_unique_file: {e}")
|
| 107 |
-
# raise
|
| 108 |
-
|
| 109 |
def save_to_excel(df, user_state):
|
| 110 |
"""
|
| 111 |
Saves the user's data into a unique Excel file named:
|
|
@@ -116,6 +66,7 @@ def save_to_excel(df, user_state):
|
|
| 116 |
Afterwards, uploads the file to an S3-compatible bucket
|
| 117 |
(e.g., Hetzner) under "Cond3/<filename>".
|
| 118 |
"""
|
|
|
|
| 119 |
# 1) Grab the unique local filename
|
| 120 |
local_file = user_state.get("excel_file")
|
| 121 |
if not local_file:
|
|
@@ -126,7 +77,6 @@ def save_to_excel(df, user_state):
|
|
| 126 |
s3_access_key = os.getenv("HETZNER_ACCESS_KEY_ID")
|
| 127 |
s3_secret_key = os.getenv("HETZNER_SECRET_ACCESS_KEY")
|
| 128 |
s3_endpoint = os.getenv("HETZNER_S3_ENDPOINT")
|
| 129 |
-
# s3_region = os.getenv("HETZNER_S3_REGION")
|
| 130 |
bucket_name = os.getenv("HETZNER_BUCKET_NAME")
|
| 131 |
|
| 132 |
try:
|
|
@@ -145,26 +95,23 @@ def save_to_excel(df, user_state):
|
|
| 145 |
|
| 146 |
# 5) Upload to S3 only if credentials/bucket are present
|
| 147 |
if s3_access_key and s3_secret_key and s3_endpoint and bucket_name:
|
| 148 |
-
my_config = Config(
|
| 149 |
-
|
|
|
|
|
|
|
| 150 |
s3_client = boto3.client(
|
| 151 |
"s3",
|
| 152 |
aws_access_key_id=s3_access_key,
|
| 153 |
aws_secret_access_key=s3_secret_key,
|
| 154 |
-
# region_name=s3_region,
|
| 155 |
endpoint_url=s3_endpoint,
|
| 156 |
config=my_config
|
| 157 |
)
|
| 158 |
|
| 159 |
-
# We'll store it in a 'Cond3' subfolder
|
| 160 |
-
s3_key = f"Cond3/{local_file}"
|
| 161 |
|
| 162 |
# 6) Upload the file
|
| 163 |
-
s3_client.upload_file(
|
| 164 |
-
local_file,
|
| 165 |
-
bucket_name,
|
| 166 |
-
s3_key
|
| 167 |
-
)
|
| 168 |
print(f"Uploaded {local_file} to S3 bucket '{bucket_name}' at key: {s3_key}")
|
| 169 |
else:
|
| 170 |
print("No S3 credentials or bucket provided, skipping S3 upload.")
|
|
|
|
| 56 |
print(f"Created empty Excel file: {filename}")
|
| 57 |
|
| 58 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
def save_to_excel(df, user_state):
|
| 60 |
"""
|
| 61 |
Saves the user's data into a unique Excel file named:
|
|
|
|
| 66 |
Afterwards, uploads the file to an S3-compatible bucket
|
| 67 |
(e.g., Hetzner) under "Cond3/<filename>".
|
| 68 |
"""
|
| 69 |
+
|
| 70 |
# 1) Grab the unique local filename
|
| 71 |
local_file = user_state.get("excel_file")
|
| 72 |
if not local_file:
|
|
|
|
| 77 |
s3_access_key = os.getenv("HETZNER_ACCESS_KEY_ID")
|
| 78 |
s3_secret_key = os.getenv("HETZNER_SECRET_ACCESS_KEY")
|
| 79 |
s3_endpoint = os.getenv("HETZNER_S3_ENDPOINT")
|
|
|
|
| 80 |
bucket_name = os.getenv("HETZNER_BUCKET_NAME")
|
| 81 |
|
| 82 |
try:
|
|
|
|
| 95 |
|
| 96 |
# 5) Upload to S3 only if credentials/bucket are present
|
| 97 |
if s3_access_key and s3_secret_key and s3_endpoint and bucket_name:
|
| 98 |
+
my_config = Config(
|
| 99 |
+
signature_version="s3v4",
|
| 100 |
+
s3={"addressing_style": "path"}
|
| 101 |
+
)
|
| 102 |
s3_client = boto3.client(
|
| 103 |
"s3",
|
| 104 |
aws_access_key_id=s3_access_key,
|
| 105 |
aws_secret_access_key=s3_secret_key,
|
|
|
|
| 106 |
endpoint_url=s3_endpoint,
|
| 107 |
config=my_config
|
| 108 |
)
|
| 109 |
|
| 110 |
+
# We'll store it in a 'Cond3' subfolder, using the basename to avoid unwanted path details
|
| 111 |
+
s3_key = f"Cond3/{os.path.basename(local_file)}"
|
| 112 |
|
| 113 |
# 6) Upload the file
|
| 114 |
+
s3_client.upload_file(local_file, bucket_name, s3_key)
|
|
|
|
|
|
|
|
|
|
|
|
|
| 115 |
print(f"Uploaded {local_file} to S3 bucket '{bucket_name}' at key: {s3_key}")
|
| 116 |
else:
|
| 117 |
print("No S3 credentials or bucket provided, skipping S3 upload.")
|