Re: urldecode function?

[Fredrik Lundh]

rk*****@gmail.com wrote:

is there a function that does the opposite of urllib.urlencode?

for example
urldecode('Cat=1&by=down&start=1827')

returns a dictionary with {'Cat':1, 'by':'down','start':1827)

the cgi modules contains assorted stuff for parsing serialized HTTP
forms and query strings. here's a quick way to get a dictionary:

>>import cgi
dict(cgi.parse_qsl("Cat=1&by=down&start=1827") )

{'start': '1827', 'by': 'down', 'Cat': '1'}

there's also:

>>cgi.parse_qs("Cat=1&by=down&start=1827")

{'start': ['1827'], 'by': ['down'], 'Cat': ['1']}

>>cgi.parse_qsl("Cat=1&by=down&start=1827")

[('Cat', '1'), ('by', 'down'), ('start', '1827')]

(both these forms handle multiple instances of the same key)

to autoconvert things that look like integers, you can do e.g.

>>def safeint(x):

.... try:
.... return int(x)
.... except ValueError:
.... return x # leave as is
....

>>dict((k, safeint(v)) for k, v in

cgi.parse_qsl("Cat=1&by=down&start=1827"))
{'start': 1827, 'by': 'down', 'Cat': 1}

</F>