comp.lang.c++.moderated Google Group

TkGen - A Lexical Analyzer Generator (online)

2009-11-27T13:36:21Z

Hi folks,
I have implemented an online tokenizer generator for C++ and I would
like to share with those interested.
The web site is:
[link]
It is an initial project and it isn’t the most complete tokenizer
generator of the world, but I hope you find it useful.
Comments, critics, questions and suggestions are very welcome.

Shall we always includes definition into head file when define template class?

2009-11-27T13:34:36Z

{ The basics are covered in FAQ item 35.12 "Why can't I separate the definition
of my templates class from it's declaration and put it inside a .cpp file?"
<url: [link]>. -mod }
When I define a template class,I found that i can't separate the

Re: using declaration hides local declaration?

2009-11-27T13:34:57Z

"Chris Uzdavinis" <cuz...@gmail.com>
As I read the standard [7.3.4], Cameau is right.
using declaration most of the time works exactly like a normal declaration,
with the similar effect. The one exception I found is with member functions
where collision is processed as 'hide' instead of 'conflict'. What makes

Re: about inline function

2009-11-27T13:37:16Z

Including times when the programmer has not asked for it. Modern
compilers are becoming increasingly sophisticated at doing this using
whole program analysis.
The common cases for inlining are where a function is small and has been
declared as static. The compiler knows that such a function cannot be

for memory leaks

2009-11-27T13:36:34Z

Dear all,
In MSVC, there is some function called _CrtDumpMemoryLeaks which helps
us to find the memory leak point in the source code. But may I ask if
I were programming on Linux platform or Mac OS X, is there any similar
way to detect the memory leak in the C++ source code? Is there any
such like functions in the standard library?

Re: Seeking recommendation on free or cheap C++ compiler

2009-11-27T13:28:29Z

We might note that IDE is a common abbreviation of Integrated
Development Environment. In theory this only implies that all the
components needed to develop software in C++ are well and truly
integrated. (Editor Compiler Linker Debugger ...)
Integration - for example - means that I change a source line and then

Re: conventions for #include delimiter (quotes versus angle brackets)

2009-11-27T13:27:26Z

I use
and use an environment variable to point the compiler at
directory myLibVersionNumber that contains direcory myLib.
This seems to force the compiler (vc8) to look only in that
directory (do other compilers do the same?).
The reason for doing that is that using just this
vc8 goes on a grand tour, should it not find myheader.h

Re: inlining of functions returning an unwindable object

2009-11-27T13:29:25Z

On Wed, 25 Nov 2009 08:19:39 CST, "Martin B." <0xCDCDC...@gmx.at>
wrote:
My bad. I blanked on buffer check because I've been doing a lot of
managed code recently and got used to seeing checks in disassemblies.
Miscommunication I think. I generally associate "unwind(able)" with
use of exceptions, not just having a dtor. Since string and stream (as

Re: about inline function

2009-11-26T16:31:37Z

You can read your compiler's documentation to see if it has an option
that enables such informational output. For example, GCC has the
option -Winline that causes it to emit a warning whenever it decides
not to inline a function that is actually declared inline.
Last resort, you can set your compiler to emit assembly and inspect

Re: Seeking recommendation on free or cheap C++ compiler

2009-11-26T16:31:23Z

{ This has moved off-topic for clc++m. Accepted as follow-up. -mod }
That's patently untrue. I really don't think you realize just how
flexible the command-line is, no matter how long or intensively you
claim to have used it. For example, when I've performed a sequence of
actions in an IDE, there is no backscroll buffer for me to copy and

Re: about inline function

2009-11-26T16:31:55Z

Once your program is compiled, you open the binary executable/library
and inspect the assembler instructions. How you do it is highly
dependent on your platform.
Regards,
&rzej

Re: Suffix Return Type Syntax

2009-11-26T16:29:07Z

{ edits: quoted sig and banner removed. don't quote sigs or the banner. -mod }
Yes, it should be optional, like the lambda syntax.
According to Mr. Stroustrup, the "->decltype(...)" section is
redundant, casue the compiler already knows the return type. The
commitee is still coming to a conclusion on this, but so far, the

Re: about inline function

2009-11-26T16:29:28Z

The best shot by far: look at the disassembly. ;-)
In general, compiler has a final say at "inline". Inline is a mere
hint you can give to it because you think a particular function should
be inlined. But in my experience, compiler knows better in some cases,
and in others, it doesn't know how to inline (see a recent post on

Re: Perfect forwarding in C++03

2009-11-26T16:25:55Z

{ edits: quoted sig and banner removed. don't quote sigs or the banner. -mod }
Yes, teh belwo does run:
void func(int& i){
++i;
template <class Prmr1>
void forward(const Prmr1& p1)
{
func(p1);
int main(int argc, char** argv){
int x = 444;
std::cout << x << std::endl;
forward(boost::ref(x));
std::cout << x << std::endl;

Re: Deleting from vector while iterating allowed?

2009-11-26T16:29:08Z

Thanks for the solution! I like this better than having yet another
condition-function laying around.