Question

I have (anonymized) a hash constructed from values in Python 2.6.8:

sys.stderr.write('#' + str(dictionary['Field 4']) + '#\n')
kpis_found.append(float(int(dictionary['Field 1']), 1) *
    max(float(dictionary['Field 2']), 1) *
    max(float(dictionary['Field 3']), 1) *
    max(float(dictionary['Field 4']), 1) *
    max(float(dictionary['Field 5']), 1))

The output I get is:

[Fri Jul 13 09:04:44 2012] [error] [client ::1] #3#
[Fri Jul 13 09:04:44 2012] [error] [client ::1] Traceback (most recent call last):
[Fri Jul 13 09:04:44 2012] [error] [client ::1]   File "/Users/jonathan/mirror/civic/google_maps/index.cgi", line 357, in <module>
[Fri Jul 13 09:04:44 2012] [error] [client ::1]     max(float(dictionary['Field 4']), 1) *
[Fri Jul 13 09:04:44 2012] [error] [client ::1] TypeError: float() takes at most 1 argument (2 given)

To the best of my knowledge, the CSV files yield (usually) strings convertible to ints, or (occasionally) strings convertible to floats. If I run into a NULL, that should be easier to diagnose. The debugging output appears to confirm that the field in question is ‘3’.

How is this getting a TypeError? I’ve run through parentheses to ensure that I’m not calculating

max(float(foo, bar))

but instead calculating

max(float(foo), bar)

Any insight would be welcome.

Answer 1

float(int(dictionary['Field 1']), 1)

Pretty sure that’s your problem. You end up with float(<int_value>, 1).

This was tested with Python 3.1.2, but…

>>> float(1, 1)
Traceback (most recent call last):
  File "<pyshell#1>", line 1, in <module>
    float(1, 1)
TypeError: float() takes at most 1 argument (2 given)