Welcome, guest | Sign In | My Account | Store | Cart

Unicode repr (Python recipe) by Oren Tirosh
ActiveState Code (http://code.activestate.com/recipes/541082/)

Alternative to the builtin repr() uses the \N{FULL NAME} format instead of \uXXXX for unicode strings.

Python, 26 lines
Download
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
def urepr(x):
    import re, unicodedata

    def toname(m):
        try:
            return r"\N{%s}" % unicodedata.name(unichr(int(m.group(1), 16)))
        except ValueError:
            return m.group(0)

    return re.sub(
        r"\\[xu]((?<=x)[0-9a-f]{2}|(?<=u)[0-9a-f]{4})",
        toname,
        repr(x)
    )

def displayhook(x):
    if x is not None:
        print urepr(x)

def install():
    import sys
    sys.displayhook = displayhook

def uninstall():
    import sys
    sys.displayhook = sys.__displayhook__

When working in the interactive prompt the terminal or IDE often cannot display unicode characters. This module helps make the representation of unicode strings more readable.

>>> s = unichr(0xA0) + unichr(0x20AC)
>>> s
u'\xa0\u20ac'
>>> import urepr
>>> urepr.install()
>>> s
u'\N{NO-BREAK SPACE}\N{EURO SIGN}'
>>> eval(urepr.urepr(s)) == s
True

Know issues: this is significantly longer than the default representation.