Announcing C++ posting guidelines

[paulkp]

Something I have wanted to do for a long time, because
I get so inspired by reading this newsgroup. I'm going to
extend this article when I see new inspiring questions.

http://koti.mbnet.fi/paulkp/articles/post_this.htm

 

[Juha Nieminen]

http://koti.mbnet.fi/paulkp/articles/post_this.htm

"Should I set the pointer NULL after delete?"
"No."

I find that answer a bit questionable. What do you think eg. auto_ptr
does with its pointer?

 

[red floyd]

Something I have wanted to do for a long time, because
I get so inspired by reading this newsgroup. I'm going to
extend this article when I see new inspiring questions.

http://koti.mbnet.fi/paulkp/articles/post_this.htm

Or, since we *already* have "posting guidelines", which are
called "the FAQ", you could just point to the FAQ.

http://www.parashift.com/c++-faq-lite

And, specifically for Netiquette issues,

http://www.parashift.com/c++-faq-lite/how-to-post.html

 

[SG]

http://koti.mbnet.fi/paulkp/articles/post_this.htm

"1. Is language X better than C++?"

Your answer implies that it's only a matter of taste. That's not the
whole truth. The programming language choice is also driven by the
application domain. Different languages have different strengths in
different domains.


"2. I have a problem with this template..."

You are implying that templates are only useful for "container of T"
things. That's not true. You even suggest that templates are a sign
of bad design. I'm sure not many would agree with you. In Java I
used runtime polymorphism because it was the only abstraction
mechanism to make code reusable. In C++ you can use both features.
If you don't need runtime polymorhism but still want to keep your code
reusable templates are a nice option. Templates and runtime
polymorphism can even be combined via "type erasure".


"5. Why function pointer in member function doesn't work?"

Your answer to that one advocates replacing function pointers by
classes with virtual functions because function pointers are
supposedly bad. It's actually not all that black and white. See
templates, functors, Boost.Function, ...


In C++ you have an abundance of design choices. The use of one
particular language feature is not necessarily a bad thing. It
depends on the context which makes it difficult to write general rules
about what feature to use and what not to use.


Cheers!
SG

 

[James Kanze]

"Should I set the pointer NULL after delete?"
"No."

I find that answer a bit questionable. What do you think eg.
auto_ptr does with its pointer?

I find most of the answers a bit questionable.

 

[Michael DOUBEZ]

Juha Nieminen a écrit :

"Should I set the pointer NULL after delete?"
"No."

I find that answer a bit questionable. What do you think eg. auto_ptr
does with its pointer?

Your argument is not pertinent: auto_ptr<> doesn't set it to NULL after
deleting it. Neither in its destructor nor in release() or reset() -
although the parameter's default value is NULL.

 

[Triple-DES]

Something I have wanted to do for a long time, because
I get so inspired by reading this newsgroup. I'm going to
extend this article when I see new inspiring questions.

http://koti.mbnet.fi/paulkp/articles/post_this.htm

Your page has a clean, yet stylish look. The font you are using looks
really good, at least on my browser.