[PYTHONMAC-SIG] MacPython v1.5a4 / NumPy bugs?
Rob Managan
managan@llnl.gov
Tue, 4 Nov 1997 09:35:47 -0800
Tom Bridgman wrote:
>Since Jack is out of town most of this month, let me send this to the list...
>....
>><class ValueError at 39f2e20>: total size of new array must be unchanged
>><type 'string'>
>>72 53 <class ValueError at 39f2e20>: total size of new array must be
>>unchanged ValueError: total size of new array must be unchanged
>>Traceback (innermost last):
>> File "Macintosh HD:Development:Python v1.5:Python
>>1.5a4:Extensions:NumPy:test:test_all.py", line 53, in ?
>> test_lines(fp.readlines())
>> File "Macintosh HD:Development:Python v1.5:Python
>>1.5a4:Extensions:NumPy:test:test_all.py", line 26, in test_lines
>> raise ValueError, "%s doesn't match expected %s" % (last_out, line[1:-1])
>>ValueError: <class ValueError at 39f2e20>: total size of new array must be
>>unchanged doesn't match expected ValueError: total size of new array must
>>be unchanged
I have confirmed this error. It also occurs in 1.5a3 with which Jack
included a compiled Numpy. This looks to me to be simply a change in how
exceptions are returned. The test code expected a certain error and the
format of the returned error string changed so it bombs out. It appears
that the change is in the string "<class ValueError at 39f2e20>".
I also confirmed the error with RandomArray. This comes from converting a
float that is larger than a signed integer can hold. Must be that 1.4 used
an unsigned int in this case?? This definition of seed() in RandomArray
seems to work.
def seed(x=0,y=0):
if x == y == 0:
import time
t = time.time()
x = long(t)
y = long((t-long(t))*1000000)
ranlib.set_seeds(x,y)
*-*-*-*-*-*-*-*-*-*-**-*-*-*-*-*-*-*-*-*-*-
Rob Managan mailto://managan@llnl.gov
LLNL ph: 510-423-0903
P.O. Box 808, L-183 FAX: 510-423-5804
Livermore, CA 94551-0808
_______________
PYTHONMAC-SIG - SIG on Python for the Apple Macintosh
send messages to: pythonmac-sig@python.org
administrivia to: pythonmac-sig-request@python.org
_______________