This code properly flags an error:
print "Value b and tab: ", b,"\t-"
b=1
print "Value b and tab: ", b,"\t-"
.
NameError: name 'b' is not defined
But using this code, where the comma was forgotten before the tab, does not do the same:
print "Value b and tab: ", b"\t-"
b=1
print "Value b and tab: ", b"\t-"
What is python thinking, when it sees b"\t-"? And why doesn't it print out the value of b even when it is assigned?
b'...' is a byte string literal in Python 3.
In Python 2 that is an alias for a regular string, for forwards compatibility with Python 3. See the Python 2.6 What's New document, specifically the PEP 3112: Byte Literals section:
Python 3.0 adopts Unicode as the language’s fundamental string type and denotes 8-bit literals differently, either as
b'string'or using a bytes constructor. For future compatibility, Python 2.6 addsbytesas a synonym for thestrtype, and it also supports theb''notation.
and the String literals section in the lexical analysis chapter of the Python reference documentation:
A prefix of
'b'or'B'is ignored in Python 2; it indicates that the literal should become a bytes literal in Python 3 (e.g. when code is automatically converted with 2to3).
Because in Python 2 the syntax creates a regular Python string, echoing the value afterwards shows the normal string syntax, without the b prefix.
The reason this happens is because b before a string literal is interpreted as a command to interpret the string as a byte string. So b'...' is very different from a raw string literal '...'.
This is very common, and is not restricted to just 'b':
r'...' is interpreted as a raw string. u'...' is interpreted as a Unicode string.and so on.
Basically, try to avoid prepending characters to Python characters as they may change your data types completely.
