mirror of
https://github.com/explosion/spaCy.git
synced 2025-07-11 08:42:28 +03:00
Check for email address first
This commit is contained in:
parent
afd9fc9d36
commit
6455b574fc
|
@ -74,6 +74,8 @@ def like_email(text):
|
||||||
|
|
||||||
|
|
||||||
def like_url(text):
|
def like_url(text):
|
||||||
|
if '@' in text: # prevent matches on e-mail addresses
|
||||||
|
return False
|
||||||
# We're looking for things that function in text like URLs. So, valid URL
|
# We're looking for things that function in text like URLs. So, valid URL
|
||||||
# or not, anything they say http:// is going to be good.
|
# or not, anything they say http:// is going to be good.
|
||||||
if text.startswith('http://') or text.startswith('https://'):
|
if text.startswith('http://') or text.startswith('https://'):
|
||||||
|
@ -82,8 +84,6 @@ def like_url(text):
|
||||||
return True
|
return True
|
||||||
if text[0] == '.' or text[-1] == '.':
|
if text[0] == '.' or text[-1] == '.':
|
||||||
return False
|
return False
|
||||||
if '@' in text: #prevent matches on e-mail addresses
|
|
||||||
return False
|
|
||||||
for i in range(len(text)):
|
for i in range(len(text)):
|
||||||
if text[i] == '.':
|
if text[i] == '.':
|
||||||
break
|
break
|
||||||
|
|
Loading…
Reference in New Issue
Block a user