Spaces:
No application file
No application file
Create utils.py
Browse files
utils.py
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import time, urllib.parse, logging, random
|
| 2 |
+
from dateutil import parser as dateparser
|
| 3 |
+
|
| 4 |
+
logger = logging.getLogger(__name__)
|
| 5 |
+
logging.basicConfig(level=logging.INFO)
|
| 6 |
+
|
| 7 |
+
def canonicalize_url(url, base=None):
|
| 8 |
+
if base:
|
| 9 |
+
return urllib.parse.urljoin(base, url)
|
| 10 |
+
return url
|
| 11 |
+
|
| 12 |
+
def parse_date(s):
|
| 13 |
+
try:
|
| 14 |
+
return dateparser.parse(s)
|
| 15 |
+
except Exception:
|
| 16 |
+
return None
|
| 17 |
+
|
| 18 |
+
def backoff_sleep(attempt):
|
| 19 |
+
time.sleep(min(2**attempt + random.random(), 10))
|