Is comparing two characters (that is one character str) in Python (3.x if that matters) well defined?
or do I have to make an explicit conversion?
In other words, is:
'a' > 'b'
the same as:
ord('a') > ord('b')
1 Answer
When not sure, check the docs:
Strings (instances of
str) compare lexicographically using the numerical Unicode code points (the result of the built-in functionord()) of their characters.
So yes, the behavior is well defined.
2 Comments
Błotosmętek
BTW, that might not be what one expects using collating rules of a given natural language - for example, in Polish letter
ń precedes letter ó, however in Python 'ń'<'ó' is False, because ord('ń') is 324 and ord('ó') is 243. Similarly 'å' < 'ä' is False even though in Swedish å precedes ä.
Jongware
@Bloto: that's because of two unrelated reasons. First, Unicode order is "historical" rather than logical. Existing Unicode points cannot be redefined once they are assigned. But, probably more important, such sorting rules differ per language. That's why a proper comparison needs a locale.
memcmp. That is pretty well-defined. It indeed uses the literal character codes. (And while typing this, a question arises. What else would you expect?)'a' > 'b'is a lexicographic comparison, andord('a') > ord('b')is a numerical comparison. I'm unsure if this makes a difference when only 1 character is used.