proposed language change to int/int==float (was: PEP0238 lament)

[Bengt Richter]

On Tue, 31 Jul 2001 17:04:20 GMT, "Terry Reedy" <tjreedy at home.com> wrote:

>
>"Bengt Richter" <bokr at accessone.com> wrote in message
>news:3b666579.880046189 at wa.news.verio.net...
>> BTW, I couldn't find where it officially says a sequence
>expression's elements
>> are evaluated guaranteed left to right. E.g., if I write
>>     [foo(), bar(), baz()]
>> can I depend on global side effects of foo() to be available to
>bar(), and of bar()
>> to baz() in strict sequence? The sequences are at the top of the
>precedence table,
>> but what of the elements? Just want to make sure ;-)
>
>I remember reading that function arguments are (guaranteed) evaluated
>l to r.  I believe dict entries are also, so {'n':1, 'n':2} ==
>{'n':2}.  I presume list items are also.  This should be in ref manual
>if not already.
>
I tried to find it in the distribution HTML docs, and in the PER and
didn't find it.

It just occurred to me to think about allowing lists' elements to be
created in parallel if there's not an algorithm forcing sequentialness.

Strictly sequential tuple and dictionary building, but possibly parallel,
even distributed, list creation? Just popped into my head.