Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.c > #77300 > unrolled thread

Re: "subexpression"

Back to article view | Back to comp.lang.c

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: "subexpression" James Kuyper <jameskuyper@verizon.net> - 2015-11-27 20:16 -0500
    Re: "subexpression" Keith Thompson <kst-u@mib.org> - 2015-11-27 19:42 -0800
      Re: "subexpression" Tim Rentsch <txr@alumni.caltech.edu> - 2015-11-28 06:33 -0800
      Re: "subexpression" James Kuyper <jameskuyper@verizon.net> - 2015-11-28 10:21 -0500

#77300 — Re: "subexpression"

On 11/27/2015 08:02 PM, Stefan Ram wrote:
>   C uses the term »subexpression«, but does not define it.
> 
>   It is not easy to define »subexpression« on an elementary level.
> 
>   For example, in the web, one can find:
> 
>       »A subexpression is a part of an expression that is by
>       itself a correct expression.«
> 
>   By this, »2/3« would be a subexpression of »1/2/3«, but I
>   don't want this to be true.
> 
>   In the web, one can also find:
> 
>       »a subexpression is a part of an expression that
>       corresponds to a subtree in the parse tree«.
> 
>   That's better! So, »2/3« is /no/ subexpression in »1/2/3«,
>   because »1/2/3« is parsed as »(1/2)/3«.
> 
>   But can we explain »subexpression« to beginners who do
>   not know what a »parse tree« is?
> 
>   My first attempt was: »A subexpression is a part of a larger
>   expression that is an expression and can be replaced by its
>   value in the larger expression without changing the value of
>   the larger expression.«
> 
>   But by this definition, in »1/1/1«, the right »1/1« would be
>   a subexpression, because replacing it by its value does not
>   change the value of the full expression. So my first attempt
>   does not have all the desired properties.

A sub-expression of an expression is either one of it's own direct
operands, or recursively, one of it's operand's sub-expressions.

I don't claim that definition to be authoritative, but I think it
expresses correctly my understanding of the term, and I don't think it
conflicts with anything the standard says about sub-expressions.

[toc] | [next] | [standalone]

#77305

James Kuyper <jameskuyper@verizon.net> writes:
[...]
> A sub-expression of an expression is either one of it's own direct
> operands, or recursively, one of it's operand's sub-expressions.

Is an expression a subexpression of itself?  If so, we could refer to a
"proper subexpression" much as we talk about "proper subsets".

The standard's usage of the term is consistent with an expression *not*
being its own subexpression.

> I don't claim that definition to be authoritative, but I think it
> expresses correctly my understanding of the term, and I don't think it
> conflicts with anything the standard says about sub-expressions.

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"

[toc] | [prev] | [next] | [standalone]

#77317

Keith Thompson <kst-u@mib.org> writes:

> James Kuyper <jameskuyper@verizon.net> writes:
> [...]
>> A sub-expression of an expression is either one of it's own direct
>> operands, or recursively, one of it's operand's sub-expressions.
>
> Is an expression a subexpression of itself?  If so, we could refer
> to a "proper subexpression" much as we talk about "proper subsets".

Usually the term subexpression admits the case of the entire
expression (or what the Standard calls a full expression).
When it's important to distinguish the two possibilities,
I would expect the term "proper subexpression" rather than
just subexpression.

> The standard's usage of the term is consistent with an expression
> *not* being its own subexpression.

I believe it is (also?) consistent with whole expressions being
subexpressions of themselves (sometimes vacuously so).  In one
case that I remember ("common subexpression elimination", IIRC)
the term loses some of its force if only proper subexpressions
(ie, less than full expressions) are admitted.

[toc] | [prev] | [next] | [standalone]

#77322

On 11/27/2015 10:42 PM, Keith Thompson wrote:
> James Kuyper <jameskuyper@verizon.net> writes:
> [...]
>> A sub-expression of an expression is either one of it's own direct
>> operands, or recursively, one of it's operand's sub-expressions.
> 
> Is an expression a subexpression of itself?  If so, we could refer to a
> "proper subexpression" much as we talk about "proper subsets".
> 
> The standard's usage of the term is consistent with an expression *not*
> being its own subexpression.

I agree, which is why the definition I suggested above was intended to
be designed to not support that interpretation. It's not that I thought
about that issue directly - merely that I never considered the
possibility, and my definition reflects the fact that I didn't consider
it a possibility.

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.c

csiph-web