33 lines
1.1 KiB
Plaintext
33 lines
1.1 KiB
Plaintext
From: dalke at bioreason.com (Andrew Dalke)
|
|
Date: Thu, 22 Apr 1999 19:15:19 -0600
|
|
Subject: try vs. has_key()
|
|
References: <aahzFAM4oJ.M7M@netcom.com>
|
|
Message-ID: <371FC9A7.D3C77413@bioreason.com>
|
|
X-UID: 1366
|
|
|
|
Aahz Maruch <aahz at netcom.com> asked:
|
|
> Can people explain their preferences [for try/except over has_key]?
|
|
|
|
As I recall, has_key is a relatively new addition to Python
|
|
so older code (and people who learned from older code) will
|
|
use try/except.
|
|
|
|
I usually use has_key because coming from a shop where most
|
|
people have C experience, it's easier to explain that than using
|
|
what it in essense the side effect of exceptions.
|
|
|
|
Also, I believe Barry Warsaw did a benchmark on the tradeoffs of
|
|
when you do one over the other. The exception case of a try/except
|
|
has some pretty high overhead, while the has_key call is slightly
|
|
worse than the non-exception case of the try/except. Barry's
|
|
results showed that if 5% of the cases or less were "exceptional"
|
|
then try/except is faster, otherwise, use has_key. Alas, I cannot
|
|
find the URL for that paper.
|
|
|
|
Andrew
|
|
dalke at acm.org
|
|
|
|
|
|
|
|
|