Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #61313 > unrolled thread

Re: Eliminate "extra" variable

Back to article view | Back to comp.lang.python

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: Eliminate "extra" variable Tim Chase <python.list@tim.thechases.com> - 2013-12-08 12:58 -0600

#61313 — Re: Eliminate "extra" variable

On 2013-12-08 15:04, Peter Otten wrote:
> >     data = dict(
> >       (str(date), {source: freq})
> >       for date, freq, source in original_data
> >       )  
> 
> or even just
> 
> data = {str(date): {source: freq}
>         for date, freq, source in original_data}

I maintain enough pre-2.7 code that I tend to eschew
dict-comprehensions for the time being.  I like them as a language
addition, but can't use them yet in most of the code-bases with which
I work.  To the OP, if you don't have to support pre-2.7 code, then
this is a tidier way to go.

> But do you really need a dict with a single key? And is it even
> correct? If a date occurs twice only the last source:freq pair is
> kept. Without knowing the context the humble
> 
> data = {}
> for date, freq, source in original_data:
>     source_to_freq = data.setdefault(date, {})
>     if source in source_to_freq:
>         raise ValueError(
>             "Multiple frequencies for one source not supported")
>     source_to_freq[source] = freq
> 
> appears so much more plausible...

The OP's code didn't do any such sanity-checking so I made the
assumption that the data came in clean.  I know better than to trust
incoming data (even my own) in my own code, but...

(followup to the OP's reply elsewhere in the thread coming shortly)

-tkc

[toc] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web