None vs Empty String in Python

Question

I have an existing application which connects to a database. It is running under Python 2.7.

The application is inconsistent in the way it uses None and "" to populate variables which do not have a value. I want to make this consistent and try to update the code to one way or the other.

Thinking as a database person I think of None as the same as Null and would assume that would be the correct choice for empty variables but then this causes problems when the application does things like

if variable.upper() == "X":
    #Do something

As this raises an error if the variable is None type.

I can do

if variable is not None and variable.upper() == "X":
    #Do something

But this seems unnecessarily verbose.

Is there a best practice for how this should be handled?

To quote from The Zen of Python: Explicit is better than implicit. — Some programmer dude
– Some programmer dude, Commented Dec 15, 2011 at 11:53

obmarg · Accepted Answer · 2011-12-15 11:52:28Z

17

You could cut down on code slightly by just writing

if variable and variable.upper() == "X":
    #Do something

If the variable is none or empty, then it's equivalent to False.

answered Dec 15, 2011 at 11:52

obmarg

9,60539 silver badges60 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

jjwchoy · Accepted Answer · 2011-12-15 11:53:09Z

4

if variable and variable.upper() == 'X'

is a little less wordy. It will also treat None and the empty string in the same way if that is something you want

Edit: Note that this does have different semantics to the expression you posted in the way it handles empty strings... i.e. in your expression the rhs of the and would get evaluated if variable is the empty string, but in this expression it would not as the empty string evaluates to False

answered Dec 15, 2011 at 11:53

jjwchoy

1,92816 silver badges20 bronze badges

Comments

Alexander Gessler · Accepted Answer · 2011-12-15 11:54:32Z

4

if (variable or '').upper() == 'X' or variable and variable.upper() == 'X' would both be a shorthand, but it's certainly not exactly as compact as the original code. I fear you won't get much better, though.

You could do

def norm(s):
   return s or ''

...
if norm(s) == 'X':

but ultimatively it won't change to much also. Perhaps the original distinction was not so inconsistent at all, but rather very pragmatic in the context?

answered Dec 15, 2011 at 11:54

Alexander Gessler

46.9k7 gold badges86 silver badges124 bronze badges

Comments

Eric Vook · Accepted Answer · 2020-06-11 15:36:05Z

4

For those who want to distinguish None from the empty string (''), use str

def is_empty_string(value):
...     return str(value) == ''
...
is_empty_string(None)
False

is_empty_string(0)
False

is_empty_string(' ')
False

is_empty_string('')
True

answered Jun 11, 2020 at 15:36

Eric Vook

412 bronze badges

Collectives™ on Stack Overflow

None vs Empty String in Python

4 Answers 4

Comments

Comments

Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related