Smalltalk › Squeak › Squeak - Dev

name puzzle: how to name a message

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

14 messages Options

Tobias Pape

name puzzle: how to name a message

Hi all

I've got a small naming puzzle:

So we have
x to: y
that makes
Intervall (x, x+1, x+2, ... , y)

And we can do
x to: x + z
to get
Intervall (x, x+1, x+2, ..., x+z)

But we have x twice in there.

What would you name the message
x ???? z
that gives
Intervall (x, x+1, x+2, ..., x+z)

I'm curious for your answers :)

Best regards
-Tobias

PS: #by: is already taken for #to:by:
PPS: A workaround is
(0 to: z) + x

Stéphane Rollandin

Re: name puzzle: how to name a message

x extra: z

?

Stef

Jakob Reschke-2

Re: name puzzle: how to name a message

In reply to this post by Tobias Pape

x withSuccessors: z

Jakob

2015-07-07 15:14 GMT+02:00 Stéphane Rollandin <[hidden email]>:
> x extra: z
>
> ?
>
> Stef
>

Tobias Pape

Re: name puzzle: how to name a message

but what if z is negative?

On 07.07.2015, at 15:39, Jakob Reschke <[hidden email]> wrote:

> x withSuccessors: z
>
> Jakob
>
> 2015-07-07 15:14 GMT+02:00 Stéphane Rollandin <[hidden email]>:
>> x extra: z
>>
>> ?
>>
>> Stef

Jakob Reschke-2

Re: name puzzle: how to name a message

In reply to this post by Jakob Reschke-2

Then maybe x toOffset: z?

2015-07-07 15:40 GMT+02:00 Tobias Pape <[hidden email]>:

> but what if z is negative?
>
> On 07.07.2015, at 15:39, Jakob Reschke <[hidden email]> wrote:
>
>> x withSuccessors: z
>>
>> Jakob
>>
>> 2015-07-07 15:14 GMT+02:00 Stéphane Rollandin <[hidden email]>:
>>> x extra: z
>>>
>>> ?
>>>
>>> Stef
>
>
>

Stéphane Rollandin

Re: name puzzle: how to name a message

x walking: z

?

Nicolai Hess

Re: name puzzle: how to name a message

In reply to this post by Tobias Pape

2015-07-07 15:09 GMT+02:00 Tobias Pape <[hidden email]>:

Hi all

I've got a small naming puzzle:

So we have
x to: y
that makes
Intervall (x, x+1, x+2, ... , y)

And we can do
x to: x + z
to get
Intervall (x, x+1, x+2, ..., x+z)

But we have x twice in there.

What would you name the message
x ???? z
that gives
Intervall (x, x+1, x+2, ..., x+z)

I'm curious for your answers :)

Best regards
-Tobias

PS: #by: is already taken for #to:by:
PPS: A workaround is
(0 to: z) + x

How about doing it the otherway around, and find a good name for (0 to: z)

for example #asRange

z asRange + x

z asRangeOffsetBy:x

maybe, asRange is not a good name.

nicolai

cbc

Re: name puzzle: how to name a message

x andNext: z

x length: z

On Tue, Jul 7, 2015 at 7:06 AM, Nicolai Hess <[hidden email]> wrote:

2015-07-07 15:09 GMT+02:00 Tobias Pape <[hidden email]>:
Hi all

I've got a small naming puzzle:

So we have
x to: y
that makes
Intervall (x, x+1, x+2, ... , y)

And we can do
x to: x + z
to get
Intervall (x, x+1, x+2, ..., x+z)

But we have x twice in there.

What would you name the message
x ???? z
that gives
Intervall (x, x+1, x+2, ..., x+z)

I'm curious for your answers :)

Best regards
-Tobias

PS: #by: is already taken for #to:by:
PPS: A workaround is
(0 to: z) + x

How about doing it the otherway around, and find a good name for (0 to: z)
for example #asRange

z asRange + x
or
z asRangeOffsetBy:x

maybe, asRange is not a good name.

nicolai

Stéphane Rollandin

Re: name puzzle: how to name a message

> x andNext: z

+1

Stef

Eliot Miranda-2

Re: name puzzle: how to name a message

In reply to this post by Tobias Pape

Hi Tobias,

On Jul 7, 2015, at 6:09 AM, Tobias Pape <[hidden email]> wrote:

> Hi all
>
> I've got a small naming puzzle:
>
> So we have
> x to: y
> that makes
> Intervall (x, x+1, x+2, ... , y)
>
> And we can do
> x to: x + z
> to get
> Intervall (x, x+1, x+2, ..., x+z)
>
>
> But we have x twice in there.
>
> What would you name the message
> x ???? z
> that gives
> Intervall (x, x+1, x+2, ..., x+z)
>
> I'm curious for your answers :)

I'm in two mins about this. I really think that considering the reader is important. Is a variable to hold x so bad? Adding yet another API call puts cognitive load in us and forces the new reader to go to the browser to learn what this tiny piece of syntactic sugar adds. Is it really worth it?

But since extreme brevity and lots of brief syntactic sugar is the de jure style I guess I'll have to deal with this so let me say

andNext:

doesn't have any correlates that I know of but

throughNext:
throughSum:

have correlates in stream methods, and

toSum:

correlates with to: and implies toProduct: if ever someone really wanted it ;-)

>
> Best regards
> -Tobias
>
> PS: #by: is already taken for #to:by:
> PPS: A workaround is
> (0 to: z) + x

Eliot (phone)

Bert Freudenberg

Re: name puzzle: how to name a message

In reply to this post by Tobias Pape

On 07.07.2015, at 06:09, Tobias Pape <[hidden email]> wrote:

>
> Hi all
>
> I've got a small naming puzzle:
>
> So we have
> x to: y
> that makes
> Intervall (x, x+1, x+2, ... , y)
>
> And we can do
> x to: x + z
> to get
> Intervall (x, x+1, x+2, ..., x+z)
>
>
> But we have x twice in there.
>
> What would you name the message
> x ???? z
> that gives
> Intervall (x, x+1, x+2, ..., x+z)
>
> I'm curious for your answers :)
>
> Best regards
> -Tobias

x steps: z
x steps: z by: b

x withNext: z
x withNext: z by: b

- Bert -

smime.p7s (5K) Download Attachment

cbc

Re: name puzzle: how to name a message

#steps: reminds me of mophic - step interval, right?

#withNext: reminds me of #withIndexDo:. I'd expect it to provide two arguments to the block (until I learned otherwise).

#next: is stream, and definitely not what we want.

I like Eliot's questioning of adding a new method at all, although I'm guilty of doing that a lot.

What about a new constructor on Interval?

Interval startAt: x next: z

Interval startAt: x next: z by: b

On Tue, Jul 7, 2015 at 1:06 PM, Bert Freudenberg <[hidden email]> wrote:

On 07.07.2015, at 06:09, Tobias Pape <[hidden email]> wrote:
>
> Hi all
>
> I've got a small naming puzzle:
>
> So we have
> x to: y
> that makes
> Intervall (x, x+1, x+2, ... , y)
>
> And we can do
> x to: x + z
> to get
> Intervall (x, x+1, x+2, ..., x+z)
>
>
> But we have x twice in there.
>
> What would you name the message
> x ???? z
> that gives
> Intervall (x, x+1, x+2, ..., x+z)
>
> I'm curious for your answers :)
>
> Best regards
> -Tobias

x steps: z
x steps: z by: b

x withNext: z
x withNext: z by: b

- Bert -

Tobias Pape

Re: name puzzle: how to name a message

Hi all

thanks for your feedback, and to calm Chris C. and Eliot, I don't
intend to add yet another convenience message™ just because.
If there had been immediate consensus, probably. I just found it
stunning that quite a few people at my lab had to think hard to
a) explain the problem and b) suggest a name :).

Best regards
-Tobias

On 07.07.2015, at 23:27, Chris Cunningham <[hidden email]> wrote:

> #steps: reminds me of mophic - step interval, right?
>
> #withNext: reminds me of #withIndexDo:. I'd expect it to provide two arguments to the block (until I learned otherwise).
>
> #next: is stream, and definitely not what we want.
>
> I like Eliot's questioning of adding a new method at all, although I'm guilty of doing that a lot.
>
> What about a new constructor on Interval?
>
> Interval startAt: x next: z
> Interval startAt: x next: z by: b
> ?
>
> On Tue, Jul 7, 2015 at 1:06 PM, Bert Freudenberg <[hidden email]> wrote:
> On 07.07.2015, at 06:09, Tobias Pape <[hidden email]> wrote:
> >
> > Hi all
> >
> > I've got a small naming puzzle:
> >
> > So we have
> > x to: y
> > that makes
> > Intervall (x, x+1, x+2, ... , y)
> >
> > And we can do
> > x to: x + z
> > to get
> > Intervall (x, x+1, x+2, ..., x+z)
> >
> >
> > But we have x twice in there.
> >
> > What would you name the message
> > x ???? z
> > that gives
> > Intervall (x, x+1, x+2, ..., x+z)
> >
> > I'm curious for your answers :)
> >
> > Best regards
> > -Tobias
>
>
> x steps: z
> x steps: z by: b
>
> x withNext: z
> x withNext: z by: b
>
> - Bert -

Tobias Pape

Re: name puzzle: how to name a message

On 08.07.2015, at 15:19, Tobias Pape <[hidden email]> wrote:

> Hi all
>
>
> thanks for your feedback, and to calm Chris C. and Eliot, I don't
> intend to add yet another convenience message™ just because.
> If there had been immediate consensus, probably. I just found it
> stunning that quite a few people at my lab had to think hard to
> a) explain the problem and b) suggest a name :).

This came just to my mind:
5 toBias: 4

… I show myself out
-T