Topic: Compile Time


Author: allan_w@my-dejanews.com (Allan W)
Date: Sun, 5 Jan 2003 10:33:28 +0000 (UTC)
Raw View
The term "compile time" is used several places in the standard:
  * 18.2.1.2/40
  * 23.1 (tables 65 and 66)
  * 23.1.2 (table 69)
  * 24.3.3
As far as I can tell, this term isn't defined anywhere.

Is this a case where English Language is precise enough? Or is this
a (minor) defect?

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: allan_w@my-dejanews.com (Allan W)
Date: Sun, 5 Jan 2003 18:32:20 +0000 (UTC)
Raw View
The term "compile time" is used several places in the standard:
  * 18.2.1.2/40
  * 23.1 (tables 65 and 66)
  * 23.1.2 (table 69)
  * 24.3.3
As far as I can tell, this term isn't defined anywhere.

Is this a case where English Language is precise enough? Or is this
a (minor) defect?

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: gdr@integrable-solutions.net (Gabriel Dos Reis)
Date: Sun, 5 Jan 2003 19:01:07 +0000 (UTC)
Raw View
allan_w@my-dejanews.com (Allan W) writes:

| The term "compile time" is used several places in the standard:
|   * 18.2.1.2/40
|   * 23.1 (tables 65 and 66)
|   * 23.1.2 (table 69)
|   * 24.3.3
| As far as I can tell, this term isn't defined anywhere.

Right.  You may also notice that yours references are from the libary
part; it is seems that the loiibrary working group is very influenced
by the mostly tools used.

| Is this a case where English Language is precise enough? Or is this
| a (minor) defect?

I believe it is a minor defect since I think what is meant is that
something happens during translation phases (see =A72.1).

--=20
Gabriel Dos Reis, gdr@integrable-solutions.net

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: gdr@integrable-solutions.net (Gabriel Dos Reis)
Date: 05 Jan 2003 20:24:55 +0100
Raw View
allan_w@my-dejanews.com (Allan W) writes:

| The term "compile time" is used several places in the standard:
|   * 18.2.1.2/40
|   * 23.1 (tables 65 and 66)
|   * 23.1.2 (table 69)
|   * 24.3.3
| As far as I can tell, this term isn't defined anywhere.

Right.  You may also notice that your references are from the libary
part; it seems that the library working group is very influenced
by the mostly used tools.

| Is this a case where English Language is precise enough? Or is this
| a (minor) defect?

I believe it is a minor defect since I think what is meant is that
a computation happens at one of the translation phases (see    2.1).

--
Gabriel Dos Reis, gdr@integrable-solutions.net

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: thp@cs.ucr.edu
Date: Mon, 6 Jan 2003 00:39:39 +0000 (UTC)
Raw View
Allan W <allan_w@my-dejanews.com> wrote:
+ The term "compile time" is used several places in the standard:
+  * 18.2.1.2/40
+  * 23.1 (tables 65 and 66)
+  * 23.1.2 (table 69)
+  * 24.3.3
+ As far as I can tell, this term isn't defined anywhere.
+
+ Is this a case where English Language is precise enough? Or is this
+ a (minor) defect?

IMHO, "compile time" should be understood to mean "translation phases 1
through 7":

   The resulting tokens are syntactically and semantically analyzed
   and translated.  [2.1 Phases of Translation, item 7]

It would be an improvement if that were stated explicitly.

Tom Payne

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: "Victor Bazarov" <v.Abazarov@attAbi.com>
Date: Sun, 5 Jan 2003 18:40:29 CST
Raw View
"Allan W" <allan_w@my-dejanews.com> wrote...
> The term "compile time" is used several places in the standard:
>   * 18.2.1.2/40
>   * 23.1 (tables 65 and 66)
>   * 23.1.2 (table 69)
>   * 24.3.3
> As far as I can tell, this term isn't defined anywhere.
>
> Is this a case where English Language is precise enough? Or is this
> a (minor) defect?


Also in 21.1.1, table 37, the term "compile-time" is used.  What
would that one mean and is it different from "compile time"?
Do you feel that "compile time" is used in the Standard somehow
in several different meanings?...

What did you find imprecise about the uses of the term?  Do we
really need to define every word used in the Standard beyond the
most simple ones?  And who's going to decide which ones are simple
enough?  Wouldn't that require growing the subclause 1.3, to the
size of a Programmer's Dictionary?  The Standard is indeed created
for literate people, not as a learning aid, IMHO.

One thing I find rather strange, though.  The subclause 2.1 gives
the description of the phases of _translation_ and later the term
_compile_, _compiler_, etc. is used.  Shouldn't then it be "phases
of _compilation_" or _translate_ and _translator_, etc. ?  Or am
I stepping over the same boundary, and the reader is supposed to
know that a _compiler_ and _translator_ is the same thing?

Victor
--
Please remove capital A's from my address when replying by mail

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: gdr@integrable-solutions.net (Gabriel Dos Reis)
Date: Mon, 6 Jan 2003 04:30:34 +0000 (UTC)
Raw View
"Victor Bazarov" <v.Abazarov@attAbi.com> writes:

| What did you find imprecise about the uses of the term?

The standard does not define what that means and *does not require
compilation.*  So the question is legitimate.

| Do we
| really need to define every word used in the Standard beyond the
| most simple ones?

Certainly not, but those words that describe key properties are better
off not lead to confusion.

| And who's going to decide which ones are simple
| enough?  Wouldn't that require growing the subclause 1.3, to the
| size of a Programmer's Dictionary?

I don't know.  But if it should take that to make the specification not
ambiguous nor imprecise then so be it.  But I sincerly doubt that
would be the case.

| The Standard is indeed created
| for literate people, not as a learning aid, IMHO.

So?

--
Gabriel Dos Reis, gdr@integrable-solutions.net

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: noone@nowhere.com ("Buster")
Date: Mon, 6 Jan 2003 06:36:53 +0000 (UTC)
Raw View
> Also in 21.1.1, table 37, the term "compile-time" is used.  What
> would that one mean and is it different from "compile time"?

Might it be that "compile-time" is being used attributively (like an
adjective before a noun, as in "the red dog" or "the compile-time
error"), and "compile time" as an ordinary noun phrase? I won't check
the references, they sound scary.

Buster


---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: kanze@gabi-soft.de (James Kanze)
Date: Mon, 6 Jan 2003 14:38:56 +0000 (UTC)
Raw View
thp@cs.ucr.edu wrote in message news:<ava298$c4k$2@glue.ucr.edu>...
> Allan W <allan_w@my-dejanews.com> wrote:
> + The term "compile time" is used several places in the standard:
> +  * 18.2.1.2/40
> +  * 23.1 (tables 65 and 66)
> +  * 23.1.2 (table 69)
> +  * 24.3.3
> + As far as I can tell, this term isn't defined anywhere.

> + Is this a case where English Language is precise enough? Or is this
> + a (minor) defect?

> IMHO, "compile time" should be understood to mean "translation phases
> 1 through 7":

>    The resulting tokens are syntactically and semantically analyzed
>    and translated.  [2.1 Phases of Translation, item 7]

> It would be an improvement if that were stated explicitly.

In general, the standard makes no distinction between compiler and
linker.  Everything that is necessary to build an executable program is
"translation" to it -- including phases that we would normally consider
linking.

In all of the cases, I'm pretty sure that this is the intent as well.

In the case of 18.2.1.2/40, it could be rephrased as "if it is not
possible to determine during translation whether the type allows
denormalized values."

In the case of the tables chapter 23, "N.A." would doubtlessly be more
appropriate: what can you say about the run-time complexity of a typedef
or a class definition?

In the case of 24.3.3, replacing "at compile time" with "during
translation" should be sufficient.

In all cases, I think that anyone reading the standard would understand
it in this way, so it certainly isn't a major defect.  But if you do try
and define "compile time", within the standard, it means ALL of the
phases of translation.

--
James Kanze                           mailto:jkanze@caicheuvreux.com
Conseils en informatique orient   e objet/
                    Beratung in objektorientierter Datenverarbeitung

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]





Author: rodolfofrino@dodo.com.au ("Rodolfo Frino")
Date: Mon, 6 Jan 2003 19:52:13 +0000 (UTC)
Raw View
Ummm...by experience I can say that no language is, *in general*, accurate
enough to
define technical words. Take the word bug (as program defect), for example,
is the english language good enough to define a defect? clearly not.

and what about "Compile Time? "Do we define "compile time" as the time
required to compile a computer program by a program called "compiler" ?

When the preprocessor is called automatically by the compiler, do we include
the processing time due to the "preprocessor" as well ?

And what is our definition of compile time if the preprocessor is not called
automatically by the compiler, or if there is no preprocessor at all?

Rodolfo

"Allan W" <allan_w@my-dejanews.com> wrote in message
news:7f2735a5.0301050230.70481f4f@posting.google.com...
> The term "compile time" is used several places in the standard:
>   * 18.2.1.2/40
>   * 23.1 (tables 65 and 66)
>   * 23.1.2 (table 69)
>   * 24.3.3
> As far as I can tell, this term isn't defined anywhere.
>
> Is this a case where English Language is precise enough? Or is this
> a (minor) defect?
>
> ---
> [ comp.std.c++ is moderated.  To submit articles, try just posting with ]
> [ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
> [              --- Please see the FAQ before posting. ---               ]
> [ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]
>

---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.jamesd.demon.co.uk/csc/faq.html                       ]