Re: List replication operator

On Mon, May 28, 2018 at 1:46 PM, Richard Damon wrote: > On 5/25/18 6:28 PM, Steven D'Aprano wrote: >> On Sat, 26 May 2018 02:58:06 +1000, Chris Angelico wrote: >> >>> Your mail/news client might choose to represent configure.ac as a link, >>> since ".ac" is a valid TLD.

Re: List replication operator

On 5/25/18 6:28 PM, Steven D'Aprano wrote: > On Sat, 26 May 2018 02:58:06 +1000, Chris Angelico wrote: > >> Your mail/news client might choose to represent configure.ac as a link, >> since ".ac" is a valid TLD. > Isn't *everything* a valid TLD now? > > For the right price, at least. > Not quite,

Re: List replication operator

If the behaviour of * on a sequence is confusing, then isn't the following behaviour also confusing for exactly the same reason? >>> a = [[]] >>> b = a + a >>> b [[], []] >>> b[0].append(5) >>> b [[5], [5]] And if so, shouldn't there also be a concatenation operator that

Re: List replication operator

On Fri, 25 May 2018 09:58:19 -0700, Rob Gaddi wrote: > I agree that it's non-obvious, but I disagree with your diagnosis. A further data point: "I was not properly appreciating that that these repeated objects were the *same identical* objects that were in the pre-replicated

Re: List replication operator

Steven D'Aprano wrote: > On Fri, 25 May 2018 09:28:01 +0200, Peter Otten wrote: > >> Yet another arcanum to learn for beginners with little return. If you >> cannot refrain from tinkering with the language at least concentrate on >> the features with broad application. Thank you. > > Broader

Re: List replication operator

bartc wrote: /The/ matrix multiplication operator? In which language? And what was wrong with "*"? It seems you're unaware that Python *already* has an '@' operator. It was added specifically so that numpy could use it for matrix multiplication. A new operator was needed because numpy already

Re: List replication operator

Dennis Lee Bieber wrote: @ requires use of the weaker/shorter "ring finger" on (for me) the weaker left hand. But... choosing an operator on that basis would be discriminating against left-handed people! -- Greg -- https://mail.python.org/mailman/listinfo/python-list

Re: List replication operator

On Fri, 25 May 2018 09:58:19 -0700, Rob Gaddi wrote: [...] >> This is a frequent, recurring pain point. Experienced programmers >> forget how confusing the behaviour of * is because they're so used to >> the execution model. They forget that writing a list comp is not even >> close to obvious,

Re: List replication operator

On Fri, 25 May 2018 13:05:01 -0400, Dennis Lee Bieber wrote: > Let me play devil's advocate... and propose a simple change, with no > new operators... > > sl = [] * n #current behavior > dl = n * [] #deep copy behavior n*[] is already supported. You will suddenly

Re: List replication operator

On Sat, 26 May 2018 02:58:06 +1000, Chris Angelico wrote: > Your mail/news client might choose to represent configure.ac as a link, > since ".ac" is a valid TLD. Isn't *everything* a valid TLD now? For the right price, at least. -- Steve --

Re: List replication operator

On 2018-05-25 18:05, Dennis Lee Bieber wrote: On Fri, 25 May 2018 15:40:41 + (UTC), Steven D'Aprano declaimed the following: On Fri, 25 May 2018 22:46:37 +1000, Chris Angelico wrote: We've already had a suggestion for [[]]@5 and that should deal

Re: List replication operator

On 05/25/2018 10:13 AM, bartc wrote: On 25/05/2018 17:58, Rob Gaddi wrote: So, in the spirit of explicit being better than implicit, please assume that for actual implementation replicate would be a static method of actual list, rather than the conveniently executable hackjob below. _list

Re: List replication operator

On 25/05/2018 17:58, Rob Gaddi wrote: So, in the spirit of explicit being better than implicit, please assume that for actual implementation replicate would be a static method of actual list, rather than the conveniently executable hackjob below. _list = list _nodefault = object() class

Re: List replication operator

On 05/25/2018 04:53 AM, Steven D'Aprano wrote: On Fri, 25 May 2018 09:28:01 +0200, Peter Otten wrote: Yet another arcanum to learn for beginners with little return. If you cannot refrain from tinkering with the language at least concentrate on the features with broad application. Thank you.

Re: List replication operator

On Sat, May 26, 2018 at 2:43 AM, bartc wrote: > On 25/05/2018 17:11, Alexandre Brault wrote: >> >> >> >> On 2018-05-25 11:40 AM, bartc wrote: >>> >>> On 25/05/2018 16:27, Chris Angelico wrote: You're way WAY too late to debate the matrix multiplication operator. >>> >>>

Re: List replication operator

On Fri, 25 May 2018 17:12:12 +0100, bartc wrote: > On 25/05/2018 16:40, Steven D'Aprano wrote: [...] >> On the other hand, it is arguable that what we really need is a >> standard function to return an N-dimensional array/list: >> >> # return a 3x4x5x6 4-D list initialised to all zeroes arr = >>

Re: List replication operator

On 25/05/2018 17:11, Alexandre Brault wrote: On 2018-05-25 11:40 AM, bartc wrote: On 25/05/2018 16:27, Chris Angelico wrote: You're way WAY too late to debate the matrix multiplication operator. /The/ matrix multiplication operator? In which language? And what was wrong with "*"? In

Re: List replication operator

On Fri, 25 May 2018 11:59:38 -0400, Dennis Lee Bieber wrote: > What is your definition of a multi-dimensional array -- I tend to think > of them as pre-declared sizes; not the variable (row|column) lengths > possible when using nested lists. Any fixed size array can be implemented using

Re: List replication operator

On 2018-05-25 11:40 AM, bartc wrote: > On 25/05/2018 16:27, Chris Angelico wrote: >> You're way WAY too late to debate the matrix multiplication operator. > > /The/ matrix multiplication operator? > > In which language? And what was wrong with "*"? > In Python, the language we're discussing

Re: List replication operator

On 25/05/2018 16:40, Steven D'Aprano wrote: On Fri, 25 May 2018 22:46:37 +1000, Chris Angelico wrote: We've already had a suggestion for [[]]@5 and that should deal with that issue. Steven is proposing "multiply by copying" as an alternative to "multiply by referencing", so an alternative

Re: List replication operator

On Sat, May 26, 2018 at 1:40 AM, bartc wrote: > On 25/05/2018 16:27, Chris Angelico wrote: >> >> On Fri, May 25, 2018 at 10:58 PM, bartc wrote: >>> >>> I'm in general not in favour of piling in special symbols into a language >>> just to solve some obscure or

Re: List replication operator

On Sat, May 26, 2018 at 1:46 AM, Steven D'Aprano wrote: > On Fri, 25 May 2018 18:06:00 +1000, Chris Angelico wrote: > >> Downside: while it's all very well to say that this is equivalent to >> copy.deepcopy(), that would imply replicating copy.deepcopy's

Re: List replication operator

On Fri, 25 May 2018 18:06:00 +1000, Chris Angelico wrote: > Downside: while it's all very well to say that this is equivalent to > copy.deepcopy(), that would imply replicating copy.deepcopy's semantics > in the core list type (unless it's actually literally defined as > importing a module and

Re: List replication operator

On Fri, 25 May 2018 18:06:00 +1000, Chris Angelico wrote: > Downside: while it's all very well to say that this is equivalent to > copy.deepcopy(), that would imply replicating copy.deepcopy's semantics > in the core list type (unless it's actually literally defined as > importing a module and

Re: List replication operator

On 25/05/2018 16:27, Chris Angelico wrote: On Fri, May 25, 2018 at 10:58 PM, bartc wrote: I'm in general not in favour of piling in special symbols into a language just to solve some obscure or rare problem. As I went on to demonstrate, function-like syntax (or even actual

Re: List replication operator

On Fri, 25 May 2018 22:46:37 +1000, Chris Angelico wrote: > We've already had a suggestion for [[]]@5 and that should deal with that > issue. Steven is proposing "multiply by copying" as an alternative to > "multiply by referencing", so an alternative multiplication operator > should fit that

Re: List replication operator

On Fri, 25 May 2018 13:58:03 +0100, bartc wrote: > As for '@', if a variable name can come before it /and/ after it, and > either or both can be dotted, wouldn't that cause it to be highlighted > as an email address in many circumstances? Such as in code posted here. Sure. And x/y might be

Re: List replication operator

On Fri, May 25, 2018 at 10:58 PM, bartc wrote: > I'm in general not in favour of piling in special symbols into a language > just to solve some obscure or rare problem. > > As I went on to demonstrate, function-like syntax (or even actual functions) > could do that job better, by

Re: List replication operator

On 25/05/2018 13:46, Chris Angelico wrote: On Fri, May 25, 2018 at 10:36 PM, bartc wrote: On 24/05/2018 19:17, Steven D'Aprano wrote: But what do people think about proposing a new list replication with copy operator? [[]]**5 would return a new list consisting of

Re: List replication operator

On Fri, May 25, 2018 at 10:36 PM, bartc wrote: > On 24/05/2018 19:17, Steven D'Aprano wrote: > >> But what do people think about proposing a new list replication with copy >> operator? >> >> [[]]**5 >> >> would return a new list consisting of five shallow copies of the inner

Re: List replication operator

On 25/05/2018 13:36, bartc wrote: Of course you have to implement dupllist(), but you'd have to implement ** too, and that is harder. For this specific example, it can just be: def dupllist(x,n):     return [x[0].copy() for _ in range(n)] On 25/05/2018 03:25, Steven D'Aprano wrote: > You

Re: List replication operator

On 24/05/2018 19:17, Steven D'Aprano wrote: But what do people think about proposing a new list replication with copy operator? [[]]**5 would return a new list consisting of five shallow copies of the inner list. Thoughts? Choice of ** doesn't seem right for a start, as it suggests it

Re: List replication operator

On Fri, 25 May 2018 09:28:01 +0200, Peter Otten wrote: > Yet another arcanum to learn for beginners with little return. If you > cannot refrain from tinkering with the language at least concentrate on > the features with broad application. Thank you. Broader than multi-dimensional arrays? There

Re: List replication operator

On Fri, 25 May 2018 11:48:49 +0100, Ben Bacarisse wrote: > Another way of looking at it would be in terms of evaluation rather than > copying. [] evaluates to a new list object, so if there were an > alternate version of L * n (for the sake of argument L ** n) that > evaluated the list

Re: List replication operator

Steven D'Aprano writes: > On Thu, 24 May 2018 16:05:32 -0700, Paul wrote: > >> How would one make a multi-dimensional list now, with truly-separate sub >> lists? Is there just no way to do it with the replication operator? > > Correct. Let's say you want to

Re: List replication operator

25.05.18 10:28, Peter Otten пише: Steven D'Aprano wrote: But what do people think about proposing a new list replication with copy operator? [[]]**5 would return a new list consisting of five shallow copies of the inner list. Yet another arcanum to learn for beginners with little

Re: List replication operator

On Fri, May 25, 2018 at 5:50 PM, Stefan Behnel wrote: > Peter Otten schrieb am 25.05.2018 um 09:28: >> Steven D'Aprano wrote: >> >>> But what do people think about proposing a new list replication with copy >>> operator? >>> >>> [[]]**5 >>> >>> would return a new list

Re: List replication operator

Peter Otten schrieb am 25.05.2018 um 09:28: > Steven D'Aprano wrote: > >> But what do people think about proposing a new list replication with copy >> operator? >> >> [[]]**5 >> >> would return a new list consisting of five shallow copies of the inner >> list. > > Yet another arcanum to

Re: List replication operator

Steven D'Aprano wrote: > But what do people think about proposing a new list replication with copy > operator? > > [[]]**5 > > would return a new list consisting of five shallow copies of the inner > list. Yet another arcanum to learn for beginners with little return. If you cannot refrain

Re: List replication operator

On Fri, 25 May 2018 02:45:38 +, Paul wrote: > how often would people here have needed this new operator, if it had > existed? It is common enough that it is a FAQ on the Python website. The list * operator isn't the most heavily used operator, but it does get used a bit. It comes up quite

Re: List replication operator

On Fri, 25 May 2018 15:30:36 +1000, Cameron Simpson wrote: > So, how to various forms of multidimensional lists play out as code? With my suggestion, we get: x = [0]**3 # one-dimensional y = [[0]**3]**3 # two-dimensional z = [[[0]**3]**3]**3 # three-dimensional Or there's MRAB's suggestion

Re: List replication operator

On Fri, 25 May 2018 08:11:52 +0200, Stefan Behnel wrote: > Steven D'Aprano schrieb am 25.05.2018 um 04:25: [...] >> You might be right: on further thought, I think I want deep copies, not >> shallow. > > But how would that protocol work then? What would happen with a data > structure like this:

Re: List replication operator

Steven D'Aprano schrieb am 25.05.2018 um 04:25: > On Thu, 24 May 2018 15:12:09 -0400, Ned Batchelder wrote: > >> On 5/24/18 2:17 PM, Steven D'Aprano wrote: > [...] >>> But what do people think about proposing a new list replication with >>> copy operator? >>> >>> [[]]**5 >>> >>> would return

Re: List replication operator

On 25May2018 02:32, Steven D'Aprano wrote: On Fri, 25 May 2018 08:02:38 +1000, Cameron Simpson wrote: I'm also against the "**" spelling I find, for much the same reasons that people oppose allowing "=" and "==" in the same syntactic location: they're

Re: List replication operator

On 25May2018 02:25, Steven D'Aprano wrote: On Thu, 24 May 2018 15:12:09 -0400, Ned Batchelder wrote: On 5/24/18 2:17 PM, Steven D'Aprano wrote: [...] But what do people think about proposing a new list replication with copy operator? [[]]**5

Re: List replication operator

how often would people here have needed this new operator, if it had existed? -- https://mail.python.org/mailman/listinfo/python-list

Re: List replication operator

On Thu, 24 May 2018 16:05:32 -0700, Paul wrote: > How would one make a multi-dimensional list now, with truly-separate sub > lists? Is there just no way to do it with the replication operator? Correct. Let's say you want to make a 1-D list with three items initialised to zero. This works

Re: List replication operator

On Fri, 25 May 2018 08:02:38 +1000, Cameron Simpson wrote: > I'm also against the "**" spelling I find, for much the same reasons > that people oppose allowing "=" and "==" in the same syntactic location: > they're easy to get wrong through typing inaccuracy. Do you often write y = 2**x

Re: List replication operator

On Thu, 24 May 2018 15:12:09 -0400, Ned Batchelder wrote: > On 5/24/18 2:17 PM, Steven D'Aprano wrote: [...] >> But what do people think about proposing a new list replication with >> copy operator? >> >> [[]]**5 >> >> would return a new list consisting of five shallow copies of the inner >>

Re: List replication operator

On Thu, 24 May 2018 11:24:41 -0700, Rob Gaddi wrote: > On 05/24/2018 11:17 AM, Steven D'Aprano wrote: [...] >> But what do people think about proposing a new list replication with >> copy operator? >> >> [[]]**5 >> >> would return a new list consisting of five shallow copies of the inner

Re: List replication operator

On 2018-05-25 00:05, Paul wrote: How would one make a multi-dimensional list now, with truly-separate sub lists? Is there just no way to do it with the replication operator? IE, would I just have to do X = [[], [], [], [], []] or perhaps write a function to insert new sub lists into a list,

Re: List replication operator

Paul writes: > How would one make a multi-dimensional list now, with truly-separate sub > lists? Is there just no way to do it with the replication operator? IE, > would I just have to do > X = [[], [], [], [], []] The expressions in a comprehension are evaluated each

Re: List replication operator

How would one make a multi-dimensional list now, with truly-separate sub lists? Is there just no way to do it with the replication operator? IE, would I just have to do X = [[], [], [], [], []] or perhaps write a function to insert new sub lists into a list, or...? Thanks > Paul C. > > --

Re: List replication operator

On 24May2018 18:17, Steven D'Aprano wrote: Python has a sequence replication operator: py> [1, 2]*3 [1, 2, 1, 2, 1, 2] Unfortunately, it is prone to a common "gotcha": py> x = [[]]*5 # make a multi-dimensional list py> x [[], [], [], [], []] py>

Re: List replication operator

On 2018-05-24 20:12, Ned Batchelder wrote: On 5/24/18 2:17 PM, Steven D'Aprano wrote: Python has a sequence replication operator: py> [1, 2]*3 [1, 2, 1, 2, 1, 2] Unfortunately, it is prone to a common "gotcha": py> x = [[]]*5 # make a multi-dimensional list py> x [[], [], [], [], []] py>

Re: List replication operator

On 5/24/18 2:17 PM, Steven D'Aprano wrote: Python has a sequence replication operator: py> [1, 2]*3 [1, 2, 1, 2, 1, 2] Unfortunately, it is prone to a common "gotcha": py> x = [[]]*5 # make a multi-dimensional list py> x [[], [], [], [], []] py> x[0].append(1) py> x [[1], [1], [1], [1],

Re: List replication operator

On 05/24/2018 11:17 AM, Steven D'Aprano wrote: Python has a sequence replication operator: py> [1, 2]*3 [1, 2, 1, 2, 1, 2] Unfortunately, it is prone to a common "gotcha": py> x = [[]]*5 # make a multi-dimensional list py> x [[], [], [], [], []] py> x[0].append(1) py> x [[1], [1], [1], [1],

List replication operator

Python has a sequence replication operator: py> [1, 2]*3 [1, 2, 1, 2, 1, 2] Unfortunately, it is prone to a common "gotcha": py> x = [[]]*5 # make a multi-dimensional list py> x [[], [], [], [], []] py> x[0].append(1) py> x [[1], [1], [1], [1], [1]] The reason for this behaviour is that *