[Numpy-discussion] Why are ufunc docstrings useless?

Charles R Harris charlesr.harris@gmail....
Thu May 8 22:17:42 CDT 2008


On Thu, May 8, 2008 at 9:12 PM, Charles R Harris <charlesr.harris@gmail.com>
wrote:

>
>
> On Thu, May 8, 2008 at 9:07 PM, Robert Kern <robert.kern@gmail.com> wrote:
>
>> On Thu, May 8, 2008 at 9:52 PM, Anne Archibald
>> <peridot.faceted@gmail.com> wrote:
>>
>> > Thanks! Done add, reduce, outer, and reduceat. What about __call__?
>>
>> If anyone knows enough to explicitly request a docstring from
>> __call__, they already know what it does.
>>
>> --
>>
>
> It's easier/better to do this in numpy/add_newdocs.py. For example:
>
> In [14]: from  numpy.lib import add_newdoc as add
>
> In [15]: add('numpy.core','ufunc',('reduce','hello world'))
>
> In [16]: ufunc.reduce.__doc__
> Out[16]: 'hello world'
>
> You don't have to clutter up the c sources and you can use """ """, instead
> of putting all those newlines in.
>

Also, not all c compilers will join strings on separate lines. You need to
add explicit continuation backslashes.

Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://projects.scipy.org/pipermail/numpy-discussion/attachments/20080508/05f3cb0f/attachment.html 


More information about the Numpy-discussion mailing list