News for hackers. Stuff that matters.
Re:Python Idioms [pdf] by csense on Thursday January 30, 2014 @12:06PM (#7151735)
I disagree with promoting try / catch. Exceptions like ValueError can really happen almost anywhere, so it is usually better to sanitize your inputs.

E.g. something like:

    try: 
        something = myfunc(d['x'])
    except ValueError:
        something = None
The programmer's intent is probably to only catch errors in the key lookup d['x'], but if there is some bug in the implementation of myfunc() or any of the functions called by myfunc() which causes a ValueError to unintentionally be raised, it will be caught by the except.

For dictionary lookups specifically, get() is usually preferable:

    something = d.get('x')
    if something is not None:
        something = myfunc(something)
Or if dictionary may potentially contain None values:

    if 'x' in d:
        something = myfunc(d['x'])
Re:Python Idioms [pdf] by smnrchrds on Thursday January 30, 2014 @01:04PM (#7152181)
Not to counter your point, but I would like to quote from PEP8 here:

Additionally, for all try/except clauses, limit the try clause to the absolute minimum amount of code necessary. Again, this avoids masking bugs.

Yes:

    try:
        value = collection[key]
    except KeyError:
        return key_not_found(key)
    else:
        return handle_value(value)
No:

    try:
        # Too broad!
        return handle_value(collection[key])
    except KeyError:
        # Will also catch KeyError raised by handle_value()
        return key_not_found(key)
Re:Python Idioms [pdf] by yetanotherphd on Thursday January 30, 2014 @10:29PM (#7155064)
These two comments together make a lot of sense.

When your try block is a single lookup, you might as well use an if statement or get. However, when the 'absolute minimum' is nontrivial try/except is still a good option, e.g.

  try:
      name = employee['name']
      first_name = name['first_name']
      last_name = name['last_name']
  except KeyError:
      print "Bad employee data"
      return
Re:Python Idioms [pdf] by eliteraspberrie on Thursday January 30, 2014 @12:20PM (#7151852)
Also note the second argument to the get method specifies the default value to return in case of a KeyError:

    something = d.get('x', default)
Re:Python Idioms [pdf] by middleclick on Thursday January 30, 2014 @12:12PM (#7151778)
While not disagreeing with your case, there are some cases where I try to follow this rule because in most of those cases, I would expect that there is no exception and if there is, then catch it.

  try:
      open(FILE)
  except IOError:
      it failed
Re:Python Idioms [pdf] by drakeandrews on Thursday January 30, 2014 @12:22PM (#7151863)
That example would surely be better as:

    something = myfunc(d['x']) if 'x' in d else None
Re:Python Idioms [pdf] by calroc on Thursday January 30, 2014 @02:31PM (#7152821)
In this form you perform the lookup twice: once to test 'x' in d and then again to actually get the value d['x']. Try clauses in Python are very inexpensive if they pass (don't raise an exception), so often the try..except version would be preferable.

In any event don't optimize prematurely and use a profiler rather than guessing if performance is an issue. ;-)

Re:Python Idioms [pdf] by prawks on Thursday January 30, 2014 @12:56PM (#7152106)
Doesn't this go against the Pythonic notion of handling exceptions rather than "looking before you leap"?
Re:Python Idioms [pdf] by wiredfool on Thursday January 30, 2014 @01:04PM (#7152184)
That looks like perl.
Re:Python Idioms [pdf] by GFK_of_xmaspast on Friday January 31, 2014 @01:08AM (#7155562)
what's going to happen in your second example if somebody gives you a defaultdict instead of a regular dict for d.