Formatting decimal places

[Jeff Schwab]

Re: Formatting decimal places

Julie wrote:[color=blue]
> Jeff Schwab wrote:
>[color=green]
>>Julie wrote:
>>[color=darkred]
>>>Jeff Schwab wrote:
>>>
>>>
>>>>Claudio Puviani wrote:
>>>>
>>>>
>>>>>"Jeff Schwab" <jeffplus@comca st.net> wrote
>>>>>
>>>>>
>>>>>
>>>>>>Claudio Puviani wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>These libraries are NOT the core standard C++
>>>>>>>langua ge/libraries which is the topic of this newsgroup.
>>>>>>
>>>>>>Where is that topic stated?
>>>>>
>>>>>
>>>>>http://www.parashift.com/c++-faq-lit...t.html#faq-5.9
>>>>>
>>>>>"Only post to comp.lang.c++ if your question is about the C++ language itself.
>>>>>For example, C++ code design, syntax, style, rules, bugs, etc. Ultimately this
>>>>>means your question must be answerable by looking into the C++ language
>>>>>definiti on as determined by the ISO/ANSI C++ Standard document, and by planned
>>>>>extensio ns and adjustments."
>>>>>
>>>>>Clearly, this excludes third party libraries, unless they somehow made it into
>>>>>the ISO/ANSI C++ Standard document.
>>>>
>>>>This excludes questions about thirdy-party libraries. IMO, it is in no
>>>>way OT to direct someone to an appropriate resource.
>>>
>>>
>>>That last statement is true only if qualified:
>>>
>>>...if there is no suitable implementation or construct in C++[/color]
>>
>>Boost *is* a suitable implementation in C++.[/color]
>
>
> I can't believe you still don't understand what I'm saying. I'll try to
> clarify:
>
> ...if there is no suitable implementation in the standard C++ library or no
> suitable construct in the native C++ language[/color]

And putting that construct in a namespace called "boost" makes it OT?

[Julie]

Re: Formatting decimal places

Jeff Schwab wrote:[color=blue]
>
> Julie wrote:[color=green]
> > Jeff Schwab wrote:
> >[color=darkred]
> >>Julie wrote:
> >>
> >>>Jeff Schwab wrote:
> >>>
> >>>
> >>>>Claudio Puviani wrote:
> >>>>
> >>>>
> >>>>>"Jeff Schwab" <jeffplus@comca st.net> wrote
> >>>>>
> >>>>>
> >>>>>
> >>>>>>Claudio Puviani wrote:
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>>These libraries are NOT the core standard C++
> >>>>>>>langua ge/libraries which is the topic of this newsgroup.
> >>>>>>
> >>>>>>Where is that topic stated?
> >>>>>
> >>>>>
> >>>>>http://www.parashift.com/c++-faq-lit...t.html#faq-5.9
> >>>>>
> >>>>>"Only post to comp.lang.c++ if your question is about the C++ language itself.
> >>>>>For example, C++ code design, syntax, style, rules, bugs, etc. Ultimately this
> >>>>>means your question must be answerable by looking into the C++ language
> >>>>>definiti on as determined by the ISO/ANSI C++ Standard document, and by planned
> >>>>>extensio ns and adjustments."
> >>>>>
> >>>>>Clearly, this excludes third party libraries, unless they somehow made it into
> >>>>>the ISO/ANSI C++ Standard document.
> >>>>
> >>>>This excludes questions about thirdy-party libraries. IMO, it is in no
> >>>>way OT to direct someone to an appropriate resource.
> >>>
> >>>
> >>>That last statement is true only if qualified:
> >>>
> >>>...if there is no suitable implementation or construct in C++
> >>
> >>Boost *is* a suitable implementation in C++.[/color]
> >
> >
> > I can't believe you still don't understand what I'm saying. I'll try to
> > clarify:
> >
> > ...if there is no suitable implementation in the standard C++ library or no
> > suitable construct in the native C++ language[/color]
>
> And putting that construct in a namespace called "boost" makes it OT?[/color]

I've already answered that numerous times. Just forget it.

[Jeff Schwab]

Re: Formatting decimal places

Julie wrote:
[color=blue][color=green][color=darkred]
>>>>>>>http://www.parashift.com/c++-faq-lit...t.html#faq-5.9
>>>>>>>
>>>>>>>"Only post to comp.lang.c++ if your question is about the C++ language itself.
>>>>>>>For example, C++ code design, syntax, style, rules, bugs, etc. Ultimately this
>>>>>>>means your question must be answerable by looking into the C++ language
>>>>>>>definiti on as determined by the ISO/ANSI C++ Standard document, and by planned
>>>>>>>extensio ns and adjustments."
>>>>>>>
>>>>>>>Clearl y, this excludes third party libraries, unless they somehow made it into
>>>>>>>the ISO/ANSI C++ Standard document.
>>>>>>
>>>>>>This excludes questions about thirdy-party libraries. IMO, it is in no
>>>>>>way OT to direct someone to an appropriate resource.
>>>>>
>>>>>
>>>>>That last statement is true only if qualified:
>>>>>
>>>>>...if there is no suitable implementation or construct in C++
>>>>
>>>>Boost *is* a suitable implementation in C++.
>>>
>>>
>>>I can't believe you still don't understand what I'm saying. I'll try to
>>>clarify:
>>>
>>>...if there is no suitable implementation in the standard C++ library or no
>>>suitable construct in the native C++ language[/color]
>>
>>And putting that construct in a namespace called "boost" makes it OT?[/color]
>
>
> I've already answered that numerous times. Just forget it.[/color]

You certainly have not answered it. Sorry, you're not getting off that
easily. :)

Posting C++ code to answer a question is clearly on-topic in
comp.lang.c++. Sometimes, a solution involves more code than may be
posted comfortably, so it must be discussed in a more abstract way.
Mentioning an existing repository of C++ code is one way of having such
a discussion. In this case, neither the problem posed by the OP nor the
solution provided by Boost are platform-specific; both involve only
"pure," standard C++. The complaint you've made seems to be that the
actual code is not part of the C++ standard; I can't believe this really
is what you meant, since most of the code posted here is the original
work of individuals, not excerpted from the standard. So, it seems that
the exception (no pun) you've taken to Boost is simply that it has a
name. Had I posted the code directly here, instead of redirecting the
OP to a named body of code, would you still have claimed it off-topic?
I really would like to understand the distinction you've made, and
whether I've correctly assessed your opinion.

[Julie]

Re: Formatting decimal places

Jeff Schwab wrote:[color=blue][color=green]
> > I've already answered that numerous times. Just forget it.[/color]
>
> You certainly have not answered it. Sorry, you're not getting off that
> easily. :)[/color]

Getting off easy??? Look at the length of this inane thread, nothing easy
about it...
[color=blue]
> Posting C++ code to answer a question is clearly on-topic in
> comp.lang.c++.[/color]

Agree.
[color=blue]
> Sometimes, a solution involves more code than may be
> posted comfortably, so it must be discussed in a more abstract way.[/color]

Agree. However, if there is a less-involved and equally valid solution, then
that is what should be posted. If the more involved solution uses external
libraries, then that should be considered off-topic.
[color=blue]
> Mentioning an existing repository of C++ code is one way of having such
> a discussion.[/color]

Absolutely, but it should be in the context of C++.

If someone asks about a string class, and a respondent simply states:

Look at MFC CString

that is off-topic because: there is a suitable implementation that is part of
C++.

However, that doesn't strictly mean that CString is off limits. An equally
legitimate response to the hypothetical string class question could be:

Look at the std::string class.

However, if you have some Windows-specific needs,
consider the MFC CString class because it
encapsulates a lot of XXX functionality that may
be more suitable over the std::string interface
and/or implementation. If you go that route, refer
to numerous windows-specific newsgroups for follow-up
questions.
[color=blue]
> In this case, neither the problem posed by the OP nor the
> solution provided by Boost are platform-specific; both involve only
> "pure," standard C++.[/color]

In and of itself, being platform-neutral does _not_ qualify something as being
on-topic in this forum. Consider OpenGL, platform-neutral, and off-topic.
[color=blue]
> The complaint you've made seems to be that the
> actual code is not part of the C++ standard; I can't believe this really
> is what you meant, since most of the code posted here is the original
> work of individuals, not excerpted from the standard. So, it seems that
> the exception (no pun) you've taken to Boost is simply that it has a
> name. Had I posted the code directly here, instead of redirecting the
> OP to a named body of code, would you still have claimed it off-topic?
> I really would like to understand the distinction you've made, and
> whether I've correctly assessed your opinion.[/color]

My 'complaint' was that the original respondent should have either posted about
std::fixed or not responded, as the OP question was about "trying to use the
<iomanip> setprecision()" , of which Boost doesn't immediately apply, and is
therefore off-topic.

I realize that my original response is splitting hairs, and has decomposed into
splitting atoms -- I don't have much more to add.

[Jeff Schwab]

Re: Formatting decimal places

(warning: various paragraphs snipped)

Julie wrote:
[color=blue][color=green]
>>Sometimes, a solution involves more code than may be
>>posted comfortably, so it must be discussed in a more abstract way.[/color]
>
>
> Agree. However, if there is a less-involved and equally valid solution, then
> that is what should be posted. If the more involved solution uses external
> libraries, then that should be considered off-topic.[/color]

I disagree.
[color=blue]
> If someone asks about a string class, and a respondent simply states:
>
> Look at MFC CString
>
> that is off-topic because: there is a suitable implementation that is part of
> C++.[/color]

No, that is OT because it is platform-specific.
[color=blue]
> However, that doesn't strictly mean that CString is off limits. An equally
> legitimate response to the hypothetical string class question could be:
>
> Look at the std::string class.
>
> However, if you have some Windows-specific needs,
> consider the MFC CString class because it
> encapsulates a lot of XXX functionality that may
> be more suitable over the std::string interface
> and/or implementation. If you go that route, refer
> to numerous windows-specific newsgroups for follow-up
> questions.[/color]

To me, *that* discussion would seem borderline OT. I wouldn't actually
complain about it, because it's brief, tasteful, and explicit about its
platform-specific nature.
[color=blue][color=green]
>>In this case, neither the problem posed by the OP nor the
>>solution provided by Boost are platform-specific; both involve only
>>"pure," standard C++.[/color]
>
>
> In and of itself, being platform-neutral does _not_ qualify something as being
> on-topic in this forum. Consider OpenGL, platform-neutral, and off-topic.[/color]

You think pointing someone to OpenGL would be off-topic? I again
disagree. An in-depth discussion of OpenGL would be OT, but simply
mentioning OpenGL would not be.
[color=blue]
> My 'complaint' was that the original respondent should have either posted about
> std::fixed or not responded, as the OP question was about "trying to use the
> <iomanip> setprecision()" , of which Boost doesn't immediately apply, and is
> therefore off-topic.
>
> I realize that my original response is splitting hairs, and has decomposed into
> splitting atoms -- I don't have much more to add.[/color]

OK. Thanks for taking the time to explain your opinion. I guess we'll
agree to disagree on this one.

[Dave Moore]

Re: Formatting decimal places

Jeff Schwab <jeffplus@comca st.net> wrote in message news:<wtSdnaXis vOVLBTd4p2dnA@c omcast.com>...

You guys do realize that you are having this discussion in a thread
entitled, "Formatting decimal places", right?

<mischievous grin>

[Jeff Schwab]

Re: Formatting decimal places

Dave Moore wrote:[color=blue]
> Jeff Schwab <jeffplus@comca st.net> wrote in message news:<wtSdnaXis vOVLBTd4p2dnA@c omcast.com>...
>
> You guys do realize that you are having this discussion in a thread
> entitled, "Formatting decimal places", right?
>
> <mischievous grin>[/color]

The titles of threads about topicality rarely say "topicality ."

Anyway, I think the discussion has pretty much been concluded.