Re: [Python-ideas] Python-ideas Digest, Vol 131, Issue 106

[Steven D'Aprano]

On Tue, Oct 31, 2017 at 10:31:50AM +, אלעזר wrote:

> Off topic: why can't we simply allow something like this:
> 
> (the_bob) = (name for name in ('bob','fred') if name=='bob')

Parens don't make a tuple. They are just for grouping. If you want a 
tuple, you need a comma:

the_bob, = ...

with or without the parens. It would be terribly surprising if (x) was a 
sequence on the left hand side but not on the right hand side of an 
assignment.


> Why does Python treat the parenthesis at the LHS as grouping parens?
> operators are not allowed anyway; (a + (b + c)) = [1] is syntax error.

a, (b, c), d = [1, "xy", 2]



> Currently
> 
> (x) = 1
> 
> works, but I can't see why should it.

Why shouldn't it?

Its just a trivial case of the fact that the left hand side can be 
certain kinds of expressions, some of which require parens:

(spam or ham)[x] = value

There are lots of possible expressions allowed on the LHS, and no good 
reason to prohibit (x) even though its pointless.



-- 
Steve
___
Python-ideas mailing list
Python-ideas@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/

Re: [Python-ideas] Python-ideas Digest, Vol 131, Issue 106

[אלעזר]

On Tue, Oct 31, 2017 at 12:18 PM Koos Zevenhoven  wrote:

> On Tue, Oct 31, 2017 at 11:24 AM, Petr Viktorin  wrote:
>
>> On 10/31/2017 09:54 AM, Koos Zevenhoven wrote:
>>>
>>>
>>> ​I wonder if that's more easily understood if you write it along these
>>> line(s):
>>>
>>>(the_bob,) = ​(name for name in ('bob','fred') if name=='bob')
>>>
>>
>> There are (unfortunately) several ways to do it. I prefer one that avoids
>> a trailing comma:
>>
>> [the_bob] = ​(name for name in ('bob','fred') if name=='bob')
>>
>>
> ​Maybe it's just me, but somehow that list-like syntax as an assignment
> target feels wrong in somewhat the same way that (1, 2).append(3) does.​
>
>
Off topic: why can't we simply allow something like this:

(the_bob) = (name for name in ('bob','fred') if name=='bob')

Why does Python treat the parenthesis at the LHS as grouping parens?
operators are not allowed anyway; (a + (b + c)) = [1] is syntax error.

Currently

(x) = 1

works, but I can't see why should it.

Elazar
___
Python-ideas mailing list
Python-ideas@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/

Re: [Python-ideas] Python-ideas Digest, Vol 131, Issue 106

[Koos Zevenhoven]

On Tue, Oct 31, 2017 at 11:24 AM, Petr Viktorin  wrote:

> On 10/31/2017 09:54 AM, Koos Zevenhoven wrote:
>>
>>
>> ​I wonder if that's more easily understood if you write it along these
>> line(s):
>>
>>(the_bob,) = ​(name for name in ('bob','fred') if name=='bob')
>>
>
> There are (unfortunately) several ways to do it. I prefer one that avoids
> a trailing comma:
>
> [the_bob] = ​(name for name in ('bob','fred') if name=='bob')
>
>
​Maybe it's just me, but somehow that list-like syntax as an assignment
target feels wrong in somewhat the same way that (1, 2).append(3) does.​

––Koos
​
PS. In your previous email, something (your email client?) removed the
vertical line from the quoted Chris's email, so it looks like just an
indented block. I wonder if a setting could fix that.​


-- 
+ Koos Zevenhoven + http://twitter.com/k7hoven +
___
Python-ideas mailing list
Python-ideas@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/

Re: [Python-ideas] Python-ideas Digest, Vol 131, Issue 106

[Petr Viktorin]

On 10/31/2017 09:54 AM, Koos Zevenhoven wrote:
On Tue, Oct 31, 2017 at 10:01 AM, Chris Angelico >wrote:


On Tue, Oct 31, 2017 at 6:46 PM, Steven D'Aprano
> wrote:
> On Tue, Oct 31, 2017 at 06:02:34PM +1100, Chris Angelico wrote:
>> One small change: If you use next(i) instead of i.next(), your code
>> should work on both Py2 and Py3. But other than that, I think it's
>> exactly the same as most people would expect of this function.
>
> Not me. As far as I can tell, that's semantically equivalent to:
>
> def single(i):
>     result, = i
>     return result
>
> apart from slightly different error messages.

I saw the original code as being like the itertools explanatory
functions - you wouldn't actually USE those functions, but they tell
you what's going on when you use the simpler, faster, more compact
form.


​I wonder if that's more easily understood if you write it along these 
line(s):


   (the_bob,) = ​(name for name in ('bob','fred') if name=='bob')


There are (unfortunately) several ways to do it. I prefer one that 
avoids a trailing comma:


[the_bob] = ​(name for name in ('bob','fred') if name=='bob')

___
Python-ideas mailing list
Python-ideas@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/

Re: [Python-ideas] Python-ideas Digest, Vol 131, Issue 106

[Koos Zevenhoven]

On Tue, Oct 31, 2017 at 10:01 AM, Chris Angelico  wrote:

> On Tue, Oct 31, 2017 at 6:46 PM, Steven D'Aprano 
> wrote:
> > On Tue, Oct 31, 2017 at 06:02:34PM +1100, Chris Angelico wrote:
> >> One small change: If you use next(i) instead of i.next(), your code
> >> should work on both Py2 and Py3. But other than that, I think it's
> >> exactly the same as most people would expect of this function.
> >
> > Not me. As far as I can tell, that's semantically equivalent to:
> >
> > def single(i):
> > result, = i
> > return result
> >
> > apart from slightly different error messages.
>
> I saw the original code as being like the itertools explanatory
> functions - you wouldn't actually USE those functions, but they tell
> you what's going on when you use the simpler, faster, more compact
> form.
>

​I wonder if that's more easily understood if you write it along these
line(s):

  (the_bob,) = ​(name for name in ('bob','fred') if name=='bob')

People need to learn about how to make a 1-tuple quite early on anyway, and
omitting the parentheses doesn't really help there, AFAICT. Then again, the
idiom looks even better when doing

  a, b = find_complex_roots(polynomial_of_second_order)

Except of course that I couldn't really come up with a good example of
something that is expected to find exactly two values from a larger
collection, and the students are already coming into the lecture hall.

Or should it be

  (a, b,) = find_complex_roots(polynomial_of_second_order)

?

––Koos


-- 
+ Koos Zevenhoven + http://twitter.com/k7hoven +
___
Python-ideas mailing list
Python-ideas@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/

Re: [Python-ideas] Python-ideas Digest, Vol 131, Issue 106

[Chris Angelico]

On Tue, Oct 31, 2017 at 6:46 PM, Steven D'Aprano  wrote:
> On Tue, Oct 31, 2017 at 06:02:34PM +1100, Chris Angelico wrote:
>> One small change: If you use next(i) instead of i.next(), your code
>> should work on both Py2 and Py3. But other than that, I think it's
>> exactly the same as most people would expect of this function.
>
> Not me. As far as I can tell, that's semantically equivalent to:
>
> def single(i):
> result, = i
> return result
>
> apart from slightly different error messages.

I saw the original code as being like the itertools explanatory
functions - you wouldn't actually USE those functions, but they tell
you what's going on when you use the simpler, faster, more compact
form.

ChrisA
___
Python-ideas mailing list
Python-ideas@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/

Re: [Python-ideas] Python-ideas Digest, Vol 131, Issue 106

[Chris Angelico]

On Tue, Oct 31, 2017 at 5:18 PM, Ivan Pozdeev via Python-ideas
 wrote:
>> raise WhitespaceDamagedException from None
>
>
> Thunderbird jerked on me big time. It never did anything like this before!
> Switched off Digest mode, individual messages aren't so complicated.
>
> def single(i):
> try:
> v =i.next()
> except StopIteration:
> raise ValueError('No items')
> try:
> i.next()
> except StopIteration:
> return v
> else:
> raise ValueError('More than one item')
>
> print single(name for name in('bob','fred') if name=='bob')
>

Thanks :)

One small change: If you use next(i) instead of i.next(), your code
should work on both Py2 and Py3. But other than that, I think it's
exactly the same as most people would expect of this function.

ChrisA
___
Python-ideas mailing list
Python-ideas@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/

Re: [Python-ideas] Python-ideas Digest, Vol 131, Issue 106

[Ivan Pozdeev via Python-ideas]

On 31.10.2017 8:37, python-ideas-requ...@python.org wrote:
On Tue, Oct 31, 2017 at 3:50 PM, Ivan Pozdeev via Python-ideas 
 wrote:

On 30.10.2017 17:32, Guido van Rossum wrote:

This is a key example of a case where code speaks. Can you write an
implementation of how you would want single() to work in Python code?

On Mon, Oct 30, 2017 at 2:49 AM, Ivan Pozdeev via Python-ideas
> wrote:

 The initial post on the above link summarizes the suggested
 implementation pretty well.


|defsingle(i): try: ||v =i.next()
|||exceptStopIteration:raiseException('No values')|||try: ||i.next()
||exceptStopIteration: ||returnv||else: ||raiseException('Too many values')|
||printsingle(name forname in('bob','fred')ifname=='bob')||| |

||

raise WhitespaceDamagedException from None

ChrisA


Thunderbird jerked on me big time. It never did anything like this 
before! Switched off Digest mode, individual messages aren't so complicated.


def single(i):
    try:
    v =i.next()
    except StopIteration:
    raise ValueError('No items')
    try:
    i.next()
    except StopIteration:
    return v
    else:
    raise ValueError('More than one item')

print single(name for name in('bob','fred') if name=='bob')

--

Regards,
Ivan

___
Python-ideas mailing list
Python-ideas@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/