Is this a STLport bug

[Aman JIANG]

vc8 + STLport 5.1.3:

cout << fixed << numeric_limits< double>::max();

It result a runtime error.

[Victor Bazarov]

Aman JIANG wrote:

vc8 + STLport 5.1.3:

cout << fixed << numeric_limits< double>::max();

It result a runtime error.

Have you tried debugging it? I don't have STLport (nor do I
intend to use it) but it should be fairly easily to step into
the operator<<, I would think... VC++ has a very good debugger
it would be a shame not to use it...

V
--
Please remove capital 'A's when replying by e-mail
I do not respond to top-posted replies, please don't ask

[BobR]

Victor Bazarov wrote in message...

Aman JIANG wrote:

vc8 + STLport 5.1.3:

cout << fixed << numeric_limits< double>::max();

It result a runtime error.

Have you tried debugging it? I don't have STLport (nor do I
intend to use it) but it should be fairly easily to step into
the operator<<, I would think... VC++ has a very good debugger
it would be a shame not to use it...

It has nothing to do with STL/Standard C++ Lib. It's a bug in MSVCRT.DLL.

The biggest number you can print out (directly) on window$ in 'fixed' format
is:

double tmp( std::numeric_li mits<double>::m ax() / 1.0e+276 );
std::cout << std::fixed << tmp;

Change that magic-number to '275' and BANG.

The whole '*.max()' will print on GNU/Linux (and probably others).

[ those were my findings on an Pentium4/win98se/GCC(MinGW)3.3.1 . ]
--
Bob R
POVrookie

[Aman JIANG]

On Sep 23, 12:10 am, "BobR" <removeBadB...@ worldnet.att.ne twrote:

Victor Bazarov wrote in message...

Aman JIANG wrote:

vc8 + STLport 5.1.3:

cout << fixed << numeric_limits< double>::max();

It result a runtime error.

Have you tried debugging it? I don't have STLport (nor do I
intend to use it) but it should be fairly easily to step into
the operator<<, I would think... VC++ has a very good debugger
it would be a shame not to use it...

It has nothing to do with STL/Standard C++ Lib. It's a bug in MSVCRT.DLL.

The biggest number you can print out (directly) on window$ in 'fixed' format
is:

double tmp( std::numeric_li mits<double>::m ax() / 1.0e+276 );
std::cout << std::fixed << tmp;

Change that magic-number to '275' and BANG.

The whole '*.max()' will print on GNU/Linux (and probably others).

[ those were my findings on an Pentium4/win98se/GCC(MinGW)3.3.1 . ]
--
Bob R
POVrookie

Thank you.
But if I use vc8's STL it will works fine :(

[Aman JIANG]

On Sep 22, 10:27 pm, "Victor Bazarov" <v.Abaza...@com Acast.netwrote:

Aman JIANG wrote:

vc8 + STLport 5.1.3:

cout << fixed << numeric_limits< double>::max();

It result a runtime error.

Have you tried debugging it? I don't have STLport (nor do I
intend to use it) but it should be fairly easily to step into
the operator<<, I would think... VC++ has a very good debugger
it would be a shame not to use it...

V
--
Please remove capital 'A's when replying by e-mail
I do not respond to top-posted replies, please don't ask

Thank you.
My question was just about the bug, and was there a bug,
but not about debug.

[=?UTF-8?B?RXJpayBXaWtzdHLDtm0=?=]

On 2007-09-22 10:46, Aman JIANG wrote:

vc8 + STLport 5.1.3:

cout << fixed << numeric_limits< double>::max();

It result a runtime error.

It works just fine if you use the STL that comes with VS, so the bug
must be with STLPort, for more information ask in a group discussing
STLPort.

--
Erik Wikström