www.digitalmars.com         C & C++   DMDScript  

c++.announce - DM C/C++ 8.45

reply "Walter Bright" <newshound digitalmars.com> writes:
DMC++ now passes 100% of the tests in Dr. Brian Malloy's test suite at
Conformance Comparison of C++ compilers: Oct 2003 published in Dr. Dobb's
http://www.cs.clemson.edu/~malloy/projects/ddj2/ddj2.html

http://www.digitalmars.com/compiler.htm

For comparison, here's the table from the article on how the other compilers
did:

Compiler  Clause                                                 Results
                 3 4  5  6  7  8  9 10 11 12 13  14 15   Fails  % Passed
edg 3.2          1 0  0  0  0  0  0  0  0  0  0   1  0   2      99.70
Comeau 4.3.2     1 0  0  0  0  0  0  0  0  0  1   1  0   3      99.55
Intel 7.1        0 0  0  0  0  0  0  0  0  0  2   1  0   3      99.55
PGCC 4.1.2       0 0  0  0  0  0  1  0  0  0  2   3  0   6      99.11
VC++ 7.1         1 0  0  0  2  0  1  0  1  1  2   5  0  12      98.22
gcc 3.3          1 0  0  2  3  5  1  0  6  0  1   6  1  26      96.15
Borland 6.0      0 0  1  0  9  2  1  0  8  3  3  21  1  49      92.73
Watcom 1.0      11 1  4  2 15  4  4  1  9 10  6  76  2 145      78.49
Total Cases     65 2 18 13 76 81 37 33 45 50 54 188 12 674

For DMC++:

DMC++ 8.45     0 0  0  0  0  0  0  0  0  0  0   0  0   0  100.00

Dr. Malloy created the test suite by adapting the 'book' examples from the
C++ Standard. Not reflected in these tests are the fact that DMC++ also
passes 100% of the 'book' preprocessor examples.

No, DMC++ still doesn't do export <g>.

Many thanks to all who helped with this release. It's an important milestone
for DMC++.
Sep 08 2005
next sibling parent Jan Knepper <jan smartsoft.us> writes:
Dude!

You ***ROCK***

Jan



Walter Bright wrote:
 DMC++ now passes 100% of the tests in Dr. Brian Malloy's test suite at
 Conformance Comparison of C++ compilers: Oct 2003 published in Dr. Dobb's
 http://www.cs.clemson.edu/~malloy/projects/ddj2/ddj2.html
 
 http://www.digitalmars.com/compiler.htm
 
 For comparison, here's the table from the article on how the other compilers
 did:
 
 Compiler  Clause                                                 Results
                  3 4  5  6  7  8  9 10 11 12 13  14 15   Fails  % Passed
 edg 3.2          1 0  0  0  0  0  0  0  0  0  0   1  0   2      99.70
 Comeau 4.3.2     1 0  0  0  0  0  0  0  0  0  1   1  0   3      99.55
 Intel 7.1        0 0  0  0  0  0  0  0  0  0  2   1  0   3      99.55
 PGCC 4.1.2       0 0  0  0  0  0  1  0  0  0  2   3  0   6      99.11
 VC++ 7.1         1 0  0  0  2  0  1  0  1  1  2   5  0  12      98.22
 gcc 3.3          1 0  0  2  3  5  1  0  6  0  1   6  1  26      96.15
 Borland 6.0      0 0  1  0  9  2  1  0  8  3  3  21  1  49      92.73
 Watcom 1.0      11 1  4  2 15  4  4  1  9 10  6  76  2 145      78.49
 Total Cases     65 2 18 13 76 81 37 33 45 50 54 188 12 674
 
 For DMC++:
 
 DMC++ 8.45     0 0  0  0  0  0  0  0  0  0  0   0  0   0  100.00
 
 Dr. Malloy created the test suite by adapting the 'book' examples from the
 C++ Standard. Not reflected in these tests are the fact that DMC++ also
 passes 100% of the 'book' preprocessor examples.
 
 No, DMC++ still doesn't do export <g>.
 
 Many thanks to all who helped with this release. It's an important milestone
 for DMC++.
 
 

-- ManiaC++ Jan Knepper But as for me and my household, we shall use Mozilla... www.mozilla.org
Sep 08 2005
prev sibling next sibling parent jay.a.carlson gmail.com writes:
Outstanding Walter!!!  Truly Outstanding!!!

/r
Jay


In article <dfqbfj$gm1$1 digitaldaemon.com>, Walter Bright says...
DMC++ now passes 100% of the tests in Dr. Brian Malloy's test suite at
Conformance Comparison of C++ compilers: Oct 2003 published in Dr. Dobb's
http://www.cs.clemson.edu/~malloy/projects/ddj2/ddj2.html

http://www.digitalmars.com/compiler.htm

For comparison, here's the table from the article on how the other compilers
did:

Compiler  Clause                                                 Results
                 3 4  5  6  7  8  9 10 11 12 13  14 15   Fails  % Passed
edg 3.2          1 0  0  0  0  0  0  0  0  0  0   1  0   2      99.70
Comeau 4.3.2     1 0  0  0  0  0  0  0  0  0  1   1  0   3      99.55
Intel 7.1        0 0  0  0  0  0  0  0  0  0  2   1  0   3      99.55
PGCC 4.1.2       0 0  0  0  0  0  1  0  0  0  2   3  0   6      99.11
VC++ 7.1         1 0  0  0  2  0  1  0  1  1  2   5  0  12      98.22
gcc 3.3          1 0  0  2  3  5  1  0  6  0  1   6  1  26      96.15
Borland 6.0      0 0  1  0  9  2  1  0  8  3  3  21  1  49      92.73
Watcom 1.0      11 1  4  2 15  4  4  1  9 10  6  76  2 145      78.49
Total Cases     65 2 18 13 76 81 37 33 45 50 54 188 12 674

For DMC++:

DMC++ 8.45     0 0  0  0  0  0  0  0  0  0  0   0  0   0  100.00

Dr. Malloy created the test suite by adapting the 'book' examples from the
C++ Standard. Not reflected in these tests are the fact that DMC++ also
passes 100% of the 'book' preprocessor examples.

No, DMC++ still doesn't do export <g>.

Many thanks to all who helped with this release. It's an important milestone
for DMC++.

Sep 08 2005
prev sibling parent reply "Matthew" <matthew hat.stlsoft.dot.org> writes:
Does that mean 8.45 now throws bad_alloc on op new?

"Walter Bright" <newshound digitalmars.com> wrote in message
news:dfqbfj$gm1$1 digitaldaemon.com...
 DMC++ now passes 100% of the tests in Dr. Brian Malloy's test suite at
 Conformance Comparison of C++ compilers: Oct 2003 published in Dr. Dobb's
 http://www.cs.clemson.edu/~malloy/projects/ddj2/ddj2.html

 http://www.digitalmars.com/compiler.htm

 For comparison, here's the table from the article on how the other

 did:

 Compiler  Clause                                                 Results
                  3 4  5  6  7  8  9 10 11 12 13  14 15   Fails  % Passed
 edg 3.2          1 0  0  0  0  0  0  0  0  0  0   1  0   2      99.70
 Comeau 4.3.2     1 0  0  0  0  0  0  0  0  0  1   1  0   3      99.55
 Intel 7.1        0 0  0  0  0  0  0  0  0  0  2   1  0   3      99.55
 PGCC 4.1.2       0 0  0  0  0  0  1  0  0  0  2   3  0   6      99.11
 VC++ 7.1         1 0  0  0  2  0  1  0  1  1  2   5  0  12      98.22
 gcc 3.3          1 0  0  2  3  5  1  0  6  0  1   6  1  26      96.15
 Borland 6.0      0 0  1  0  9  2  1  0  8  3  3  21  1  49      92.73
 Watcom 1.0      11 1  4  2 15  4  4  1  9 10  6  76  2 145      78.49
 Total Cases     65 2 18 13 76 81 37 33 45 50 54 188 12 674

 For DMC++:

 DMC++ 8.45     0 0  0  0  0  0  0  0  0  0  0   0  0   0  100.00

 Dr. Malloy created the test suite by adapting the 'book' examples from the
 C++ Standard. Not reflected in these tests are the fact that DMC++ also
 passes 100% of the 'book' preprocessor examples.

 No, DMC++ still doesn't do export <g>.

 Many thanks to all who helped with this release. It's an important

 for DMC++.

Sep 10 2005
parent reply "Walter Bright" <newshound digitalmars.com> writes:
"Matthew" <matthew hat.stlsoft.dot.org> wrote in message
news:dfvjnv$psg$1 digitaldaemon.com...
 Does that mean 8.45 now throws bad_alloc on op new?

No. I'm working on a way to do it that would be backwards compatible with the NULL return behavior. I'll not claim that 8.45 has reached 100% conformance; it hasn't. In the meantime, global operator new can be overridden to throw.
Sep 10 2005
parent "Matthew" <admin stlsoft.dot.dot.dot.dot.org> writes:
"Walter Bright" <newshound digitalmars.com> wrote in message 
news:dg06lh$1mk8$3 digitaldaemon.com...
 "Matthew" <matthew hat.stlsoft.dot.org> wrote in message
 news:dfvjnv$psg$1 digitaldaemon.com...
 Does that mean 8.45 now throws bad_alloc on op new?

No. I'm working on a way to do it that would be backwards compatible with the NULL return behavior. I'll not claim that 8.45 has reached 100% conformance; it hasn't. In the meantime, global operator new can be overridden to throw.

Ok btw, I'm upgrading the STLSoft libs wrt 0x0845, and am very pleased to find a lot of things that DMC++ has caught up with several aspects where is has previously lagged behind the crowd. There're still plenty of areas for improvement, though. ;-)
Sep 11 2005