From: cbcurl on
On Feb 16, 10:21 pm, mike <m....(a)irl.cri.replacethiswithnz> wrote:
> In article <5225f641-a135-4c5f-a091-11eeb2731264
> @z11g2000yqz.googlegroups.com>, cbc...(a)gmail.com says...> On Feb 16, 2:33 pm, "bartc" <ba...(a)freeuk.com> wrote:
>
> > > What if you had an array of those?
>
> > I can't imagine why one would have such a thing, but the plural would
> > be numbers-of-persons,
> > which I suppose one could abbreviate to ns-persons, although I
> > wouldn't. I would probably just
> > call it n-persons-array. See there is no hard and fast rule other than
> > to make the code readable
> > by humans.
>
> And then the number of entries in _that_ array would be n-n-persons-
> array or n-ns-persons or ...

First explain why you would even have such an array. I often find that
data structures that are hard to name indicate the code needs to be
refactored.



From: mike on
In article <0c7fc58c-cc7d-4292-880f-aefab63e7622
@t21g2000vbo.googlegroups.com>, cbcurl(a)gmail.com says...
> On Feb 16, 10:21 pm, mike <m....(a)irl.cri.replacethiswithnz> wrote:
> > In article <5225f641-a135-4c5f-a091-11eeb2731264
> > @z11g2000yqz.googlegroups.com>, cbc...(a)gmail.com says...> On Feb 16, 2:33 pm, "bartc" <ba...(a)freeuk.com> wrote:
> >
> > > > What if you had an array of those?
> >
> > > I can't imagine why one would have such a thing, but the plural would
> > > be numbers-of-persons,
> > > which I suppose one could abbreviate to ns-persons, although I
> > > wouldn't. I would probably just
> > > call it n-persons-array. See there is no hard and fast rule other than
> > > to make the code readable
> > > by humans.
> >
> > And then the number of entries in _that_ array would be n-n-persons-
> > array or n-ns-persons or ...
>
> First explain why you would even have such an array. I often find that
> data structures that are hard to name indicate the code needs to be
> refactored.
>
That is my point - an array is a useful data structure, but is not
generally a good or useful in the high-level description of the data.
Knowing, for example, that variable 'x' is an array of arrays of arrays
is (with a few exceptions) not particularly useful and shouldn't be part
of the name of the variable.

If you have a Soldier object, then you could group them in Squad
objects, which are themselves grouped into Platoons, Companies,...etc.
Grouping them into Soldier_Array and Soldier_Array_Array etc would be
unhelpful.

So if you have a need for a list of pointers to arrays of strings, isn't
it better to give it a name like Catalogue to both describe and justify
the grouping of the data, rather than some horror like str_ary_ptr_lst?

Cheers,
Mike
From: bartc on
mike wrote:
> In article <0c7fc58c-cc7d-4292-880f-aefab63e7622
> @t21g2000vbo.googlegroups.com>, cbcurl(a)gmail.com says...
>> On Feb 16, 10:21 pm, mike <m....(a)irl.cri.replacethiswithnz> wrote:
>>> In article <5225f641-a135-4c5f-a091-11eeb2731264
>>> @z11g2000yqz.googlegroups.com>, cbc...(a)gmail.com says...> On Feb
>>> 16, 2:33 pm, "bartc" <ba...(a)freeuk.com> wrote:
>>>
>>>>> What if you had an array of those?
>>>
>>>> I can't imagine why one would have such a thing, but the plural
>>>> would be numbers-of-persons,
>>>> which I suppose one could abbreviate to ns-persons, although I
>>>> wouldn't. I would probably just
>>>> call it n-persons-array. See there is no hard and fast rule other
>>>> than to make the code readable
>>>> by humans.
>>>
>>> And then the number of entries in _that_ array would be n-n-persons-
>>> array or n-ns-persons or ...
>>
>> First explain why you would even have such an array. I often find
>> that data structures that are hard to name indicate the code needs
>> to be refactored.
>>
> That is my point - an array is a useful data structure, but is not
> generally a good or useful in the high-level description of the data.
> Knowing, for example, that variable 'x' is an array of arrays of
> arrays
> is (with a few exceptions) not particularly useful and shouldn't be
> part
> of the name of the variable.

Yes, I don't think I've ever used 'array' or any of it's abbreviations as
part of a variable name.

But I have used 'list' or 'table' or whatever, or a plural, or sometimes a
name that implies a list or collection or table. In short, I've used
everything except 'array' (although I do make use of that for type names).

--
Bartc

From: Curtis Dyer on
bartc wrote:

> mike wrote:
>> In article <0c7fc58c-cc7d-4292-880f-aefab63e7622
>> @t21g2000vbo.googlegroups.com>, cbcurl(a)gmail.com says...
>>> On Feb 16, 10:21 pm, mike <m....(a)irl.cri.replacethiswithnz>
>>> wrote:
>>>> In article <5225f641-a135-4c5f-a091-11eeb2731264
>>>> @z11g2000yqz.googlegroups.com>, cbc...(a)gmail.com says...> On
>>>> Feb 16, 2:33 pm, "bartc" <ba...(a)freeuk.com> wrote:

<snip methods for indicating arrays in identifiers>

>>>> And then the number of entries in _that_ array would be
>>>> n-n-persons- array or n-ns-persons or ...
>>>
>>> First explain why you would even have such an array. I often
>>> find that data structures that are hard to name indicate the
>>> code needs to be refactored.
>>>
>> That is my point - an array is a useful data structure, but is
>> not generally a good or useful in the high-level description of
>> the data. Knowing, for example, that variable 'x' is an array of
>> arrays of arrays is (with a few exceptions) not particularly
>> useful and shouldn't be part of the name of the variable.
>
> Yes, I don't think I've ever used 'array' or any of it's
> abbreviations as part of a variable name.

IIRC, the only time I've used "array" or an abbreviation of "array"
was when writing general array manipulation functions in PHP.

> But I have used 'list' or 'table' or whatever, or a plural, or
> sometimes a name that implies a list or collection or table. In
> short, I've used everything except 'array' (although I do make use
> of that for type names).

I think you make a good point here. When it comes to discussing
identifiers in general, it can be tough, because choosing a good
identifier greatly depends on the context in which a variable is
declared. So "list" or "table" will certainly be sufficient in the
proper contexts.

I find Rob Pike's "Notes on Programming in C"* an insightful guide
concerning programming style in general.

* <http://www.lysator.liu.se/c/pikestyle.html>

--
Curtis Dyer
"Don't worry about efficiency until you've attained correctness."
-- Eric Sosman, CLC