www.digitalmars.com         C & C++   DMDScript  

c++.stlsoft - DTL, STLSoft, etc. coming soon

reply "Matthew" <matthew.hat stlsoft.dot.org> writes:
I've *finally* just sent off the last round of revisions of "Imperfect C++",
which have occupied me for the last week or so. After two years of effort, it's
a
pretty weird feeling to know that it's over the line at last.

Anyway, apart from sharing an electronic sigh of relief, this means that the
releases of STLSoft v1.7.1 and DTL 0.1 will now be pretty imminent. Expect news
on both fronts within the next two weeks. I will also now put more effort into
getting Walter to release std.loader.d and accept my updates to
std.windows.registry.d and std.recls.d.

Thanks for patience to all who've been kind enough to wait for the libraries,
and
to those who've wished me well with the book writing. What an idiot I must have
been to have thought that I should write a book. And what an idiot for embarking
on two more this year - "Extended STL", and another that's not about C++. ;)

Cheers


-- 
Matthew Wilson

Contributing editor, C/C++ Users Journal
    (www.synesis.com.au/articles.html#columns)
STLSoft moderator
    (http://www.stlsoft.org)

"You can tell a Yorkshireman, but you can't tell him much!" -- Uncle Michael

-------------------------------------------------------------------------------
May 04 2004
parent reply "fred" <info fleet-manage.com> writes:
Congratulations Matthew

When will we be able to pick up a copy of your new book ?

fred


"Matthew" <matthew.hat stlsoft.dot.org> wrote in message
news:c785tn$267$1 digitaldaemon.com...

 I've *finally* just sent off the last round of revisions of "Imperfect C++",
 which have occupied me for the last week or so. After two years of effort,
it's a
 pretty weird feeling to know that it's over the line at last.

 Anyway, apart from sharing an electronic sigh of relief, this means that the
 releases of STLSoft v1.7.1 and DTL 0.1 will now be pretty imminent. Expect news
 on both fronts within the next two weeks. I will also now put more effort into
 getting Walter to release std.loader.d and accept my updates to
 std.windows.registry.d and std.recls.d.

 Thanks for patience to all who've been kind enough to wait for the libraries,
and
 to those who've wished me well with the book writing. What an idiot I must have
 been to have thought that I should write a book. And what an idiot for
embarking
 on two more this year - "Extended STL", and another that's not about C++. ;)

 Cheers


 -- 
 Matthew Wilson

 Contributing editor, C/C++ Users Journal
     (www.synesis.com.au/articles.html#columns)
 STLSoft moderator
     (http://www.stlsoft.org)

 "You can tell a Yorkshireman, but you can't tell him much!" -- Uncle Michael

 -------------------------------------------------------------------------------
May 04 2004
parent reply "Matthew" <matthew.hat stlsoft.dot.org> writes:
Should be Sept/Oct.

It could be later if I can't find a decent subtitle for it. ;/




"fred" <info fleet-manage.com> wrote in message
news:c79ahr$1rra$1 digitaldaemon.com...

 Congratulations Matthew

 When will we be able to pick up a copy of your new book ?

 fred


 "Matthew" <matthew.hat stlsoft.dot.org> wrote in message


news:c785tn$267$1 digitaldaemon.com...

 I've *finally* just sent off the last round of revisions of "Imperfect C++",
 which have occupied me for the last week or so. After two years of effort,




it's a

 pretty weird feeling to know that it's over the line at last.

 Anyway, apart from sharing an electronic sigh of relief, this means that the
 releases of STLSoft v1.7.1 and DTL 0.1 will now be pretty imminent. Expect




news

 on both fronts within the next two weeks. I will also now put more effort




into

 getting Walter to release std.loader.d and accept my updates to
 std.windows.registry.d and std.recls.d.

 Thanks for patience to all who've been kind enough to wait for the libraries,




and

 to those who've wished me well with the book writing. What an idiot I must




have

 been to have thought that I should write a book. And what an idiot for




embarking

 on two more this year - "Extended STL", and another that's not about C++. ;)

 Cheers


 -- 
 Matthew Wilson

 Contributing editor, C/C++ Users Journal
     (www.synesis.com.au/articles.html#columns)
 STLSoft moderator
     (http://www.stlsoft.org)

 "You can tell a Yorkshireman, but you can't tell him much!" -- Uncle Michael


 -------------------------------------------------------------------------------







 May 04 2004



  parent reply C <dont respond.com>  writes:


Whats the other book about ?  Do you have something for D planned ( post 
1.0 maybe  :) ) ?

Also slightly OT , is The D Journal still in the works ?

C
Matthew wrote:

 Should be Sept/Oct.
 
 It could be later if I can't find a decent subtitle for it. ;/
 
 
 
 
 "fred" <info fleet-manage.com> wrote in message
 news:c79ahr$1rra$1 digitaldaemon.com...
 

Congratulations Matthew

When will we be able to pick up a copy of your new book ?

fred


"Matthew" <matthew.hat stlsoft.dot.org> wrote in message


 
 news:c785tn$267$1 digitaldaemon.com...
 

I've *finally* just sent off the last round of revisions of "Imperfect C++",
which have occupied me for the last week or so. After two years of effort,




 
 it's a
 

pretty weird feeling to know that it's over the line at last.

Anyway, apart from sharing an electronic sigh of relief, this means that the
releases of STLSoft v1.7.1 and DTL 0.1 will now be pretty imminent. Expect




 
 news
 

on both fronts within the next two weeks. I will also now put more effort




 
 into
 

getting Walter to release std.loader.d and accept my updates to
std.windows.registry.d and std.recls.d.

Thanks for patience to all who've been kind enough to wait for the libraries,




 
 and
 

to those who've wished me well with the book writing. What an idiot I must




 
 have
 

been to have thought that I should write a book. And what an idiot for




 
 embarking
 

on two more this year - "Extended STL", and another that's not about C++. ;)

Cheers


-- 
Matthew Wilson

Contributing editor, C/C++ Users Journal
    (www.synesis.com.au/articles.html#columns)
STLSoft moderator
    (http://www.stlsoft.org)

"You can tell a Yorkshireman, but you can't tell him much!" -- Uncle Michael


-------------------------------------------------------------------------------





 
 





 May 04 2004



  parent reply "Matthew" <matthew.hat stlsoft.dot.org>  writes:


 Whats the other book about ?  Do you have something for D planned ( post
 1.0 maybe  :) ) ?


It's a wait and see thing, as the proposal's not yet in to the publisher. But it
won't be long


 Also slightly OT , is The D Journal still in the works ?


I don't know. We'd certainly like to do it, but it may still (even after two
years) be premature. Everybody says they'd like to read it, but there are very
few who want to, or, more likely, have the time to, write for it.




 Matthew wrote:

 Should be Sept/Oct.

 It could be later if I can't find a decent subtitle for it. ;/




 "fred" <info fleet-manage.com> wrote in message
 news:c79ahr$1rra$1 digitaldaemon.com...


Congratulations Matthew

When will we be able to pick up a copy of your new book ?

fred


"Matthew" <matthew.hat stlsoft.dot.org> wrote in message


 news:c785tn$267$1 digitaldaemon.com...


I've *finally* just sent off the last round of revisions of "Imperfect C++",
which have occupied me for the last week or so. After two years of effort,




 it's a


pretty weird feeling to know that it's over the line at last.

Anyway, apart from sharing an electronic sigh of relief, this means that the
releases of STLSoft v1.7.1 and DTL 0.1 will now be pretty imminent. Expect




 news


on both fronts within the next two weeks. I will also now put more effort




 into


getting Walter to release std.loader.d and accept my updates to
std.windows.registry.d and std.recls.d.

Thanks for patience to all who've been kind enough to wait for the








libraries,

 and


to those who've wished me well with the book writing. What an idiot I must




 have


been to have thought that I should write a book. And what an idiot for




 embarking


on two more this year - "Extended STL", and another that's not about C++. ;)

Cheers


-- 
Matthew Wilson

Contributing editor, C/C++ Users Journal
    (www.synesis.com.au/articles.html#columns)
STLSoft moderator
    (http://www.stlsoft.org)

"You can tell a Yorkshireman, but you can't tell him much!" -- Uncle Michael







-------------------------------------------------------------------------------











 May 04 2004



  parent reply C <qbert atari-soldiers.com>  writes:


 I don't know. We'd certainly like to do it, but it may still (even after 
 two
 years) be premature. Everybody says they'd like to read it, but there 
 are very
 few who want to, or, more likely, have the time to, write for it.



Hmm, it might be a 'build it they will come' scenario.  I know Id like to 
contribute ( id like to do a keyword correction based on string distances 
, also want to try FSM framework and when I get more familiar with it an 
ANN [ artificial neural network ] framework ) but I never sent in the 
proposals.

Maybe we can throw them up on dsource.org for the mean time, brad ?

Another OT, are you planning on porting STLSoft's performance modules to D 
?  I was in need of this just today.

Charlie

On Wed, 5 May 2004 11:26:45 +1000, Matthew <matthew.hat stlsoft.dot.org> 
wrote:


 Whats the other book about ?  Do you have something for D planned ( post
 1.0 maybe  :) ) ?


 It's a wait and see thing, as the proposal's not yet in to the 
 publisher. But it
 won't be long


 Also slightly OT , is The D Journal still in the works ?


 I don't know. We'd certainly like to do it, but it may still (even after 
 two
 years) be premature. Everybody says they'd like to read it, but there 
 are very
 few who want to, or, more likely, have the time to, write for it.




 Matthew wrote:

 Should be Sept/Oct.

 It could be later if I can't find a decent subtitle for it. ;/




 "fred" <info fleet-manage.com> wrote in message
 news:c79ahr$1rra$1 digitaldaemon.com...


Congratulations Matthew

When will we be able to pick up a copy of your new book ?

fred


"Matthew" <matthew.hat stlsoft.dot.org> wrote in message


 news:c785tn$267$1 digitaldaemon.com...


I've *finally* just sent off the last round of revisions of 






 "Imperfect C++",

which have occupied me for the last week or so. After two years of 






 effort,

 it's a


pretty weird feeling to know that it's over the line at last.

Anyway, apart from sharing an electronic sigh of relief, this means 






 that the

releases of STLSoft v1.7.1 and DTL 0.1 will now be pretty imminent. 






 Expect

 news


on both fronts within the next two weeks. I will also now put more 






 effort

 into


getting Walter to release std.loader.d and accept my updates to
std.windows.registry.d and std.recls.d.

Thanks for patience to all who've been kind enough to wait for the








 libraries,

 and


to those who've wished me well with the book writing. What an idiot 






 I must

 have


been to have thought that I should write a book. And what an idiot 






 for

 embarking


on two more this year - "Extended STL", and another that's not about 






 C++. ;)

Cheers


--
Matthew Wilson

Contributing editor, C/C++ Users Journal
    (www.synesis.com.au/articles.html#columns)
STLSoft moderator
    (http://www.stlsoft.org)

"You can tell a Yorkshireman, but you can't tell him much!" -- Uncle 






 Michael







 -------------------------------------------------------------------------------












-- 
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/



 May 04 2004



 next sibling parent reply "Matthew" <matthew.hat stlsoft.dot.org>  writes:


"C" <qbert atari-soldiers.com> wrote in message
news:opr7ieyxr3aoygh6 news.digitalmars.com...

 I don't know. We'd certainly like to do it, but it may still (even after
 two
 years) be premature. Everybody says they'd like to read it, but there
 are very
 few who want to, or, more likely, have the time to, write for it.



 Hmm, it might be a 'build it they will come' scenario.  I know Id like to
 contribute ( id like to do a keyword correction based on string distances
 , also want to try FSM framework and when I get more familiar with it an
 ANN [ artificial neural network ] framework ) but I never sent in the
 proposals.


Very good point. The only problem there is that neither myself nor big-W have
any
spare bandwidth for TDJ at the moment. Maybe in a couple of months, but then how
many times have I said that since June 2002? :/


 Maybe we can throw them up on dsource.org for the mean time, brad ?

 Another OT, are you planning on porting STLSoft's performance modules to D
 ?  I was in need of this just today.


Yes. I've asked Walter about this a couple of times, but his plate's too small,
and it's got too much seafood on it.

What I'd like to do is have the following modules:

    std.perf
    std.linux.perf
    std.windows.perf

The latter two would be straightforward port of the UNIXSTL
(unixstl_performance_counter.h) and WinSTL classes
(winstl_performance_counter.h,
winstl_highperformance_counter.h, winstl_tick_counter.h,
winstl_processtimes_counter.h, winstl_threadtimes_counter.h). WinSTL has a
number
of flavours, the relative merits of which are described in
http://www.windevnet.com/documents/win0305a/. UNIXSTL has a comparatively
embarassing paucity of variety, with only one performance counter based on
gettimeofday.

For a port to D I'd like to see the following:
    - someone who's expert on Linux to come up with equivalents to the WinSTL
processtimes_counter, threadtimes_counter and highperformance_counter. The first
two depend on whether Linux maintains process/thread time stats; clearly it does
at least per-process timing, since the time utility does that. The latter should
probably be able to be effected using x86 commands, but I've not had chance to
look at this yet.
    - me to define std/perf.d, something along the lines of (forgiving
syntactical mistakes please):

module std.perf;

version(Windows)
{
    private import std.windows.perf;
    public alias std.windows.perf.PerformanceCounter    PerformanceCounter;
}
else version(Linux)
{
    private import std.linux.perf;
    public alias std.linux.perf.PerformanceCounter    PerformanceCounter;
}
else
{
    // usual static assertion stuff
}

Maybe I can do this now, then, eh? :) The other classes -
HighPerformanceCounter,
ThreadTimesCounter and ProcessTimesCounter can be brought into std.perf when
there are implementations for Linux.

I'm off to have a play with this now. You inspirer, you!



 Charlie

 On Wed, 5 May 2004 11:26:45 +1000, Matthew <matthew.hat stlsoft.dot.org>
 wrote:


 Whats the other book about ?  Do you have something for D planned ( post
 1.0 maybe  :) ) ?


 It's a wait and see thing, as the proposal's not yet in to the
 publisher. But it
 won't be long


 Also slightly OT , is The D Journal still in the works ?


 I don't know. We'd certainly like to do it, but it may still (even after
 two
 years) be premature. Everybody says they'd like to read it, but there
 are very
 few who want to, or, more likely, have the time to, write for it.




 Matthew wrote:

 Should be Sept/Oct.

 It could be later if I can't find a decent subtitle for it. ;/




 "fred" <info fleet-manage.com> wrote in message
 news:c79ahr$1rra$1 digitaldaemon.com...


Congratulations Matthew

When will we be able to pick up a copy of your new book ?

fred


"Matthew" <matthew.hat stlsoft.dot.org> wrote in message


 news:c785tn$267$1 digitaldaemon.com...


I've *finally* just sent off the last round of revisions of






 "Imperfect C++",

which have occupied me for the last week or so. After two years of






 effort,

 it's a


pretty weird feeling to know that it's over the line at last.

Anyway, apart from sharing an electronic sigh of relief, this means






 that the

releases of STLSoft v1.7.1 and DTL 0.1 will now be pretty imminent.






 Expect

 news


on both fronts within the next two weeks. I will also now put more






 effort

 into


getting Walter to release std.loader.d and accept my updates to
std.windows.registry.d and std.recls.d.

Thanks for patience to all who've been kind enough to wait for the








 libraries,

 and


to those who've wished me well with the book writing. What an idiot






 I must

 have


been to have thought that I should write a book. And what an idiot






 for

 embarking


on two more this year - "Extended STL", and another that's not about






 C++. ;)

Cheers


--
Matthew Wilson

Contributing editor, C/C++ Users Journal
    (www.synesis.com.au/articles.html#columns)
STLSoft moderator
    (http://www.stlsoft.org)

"You can tell a Yorkshireman, but you can't tell him much!" -- Uncle






 Michael











 -----------------------------------------------------------------------------






--












 -- 
 Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/





 May 04 2004



  parent reply "Matthew" <matthew.hat stlsoft.dot.org>  writes:


 Another OT, are you planning on porting STLSoft's performance modules to D
 ?  I was in need of this just today.


 Yes. I've asked Walter about this a couple of times, but his plate's too small,
 and it's got too much seafood on it.

 What I'd like to do is have the following modules:

     std.perf
     std.linux.perf
     std.windows.perf

 The latter two would be straightforward port of the UNIXSTL
 (unixstl_performance_counter.h) and WinSTL classes


(winstl_performance_counter.h,

 winstl_highperformance_counter.h, winstl_tick_counter.h,
 winstl_processtimes_counter.h, winstl_threadtimes_counter.h). WinSTL has a


number

 of flavours, the relative merits of which are described in
 http://www.windevnet.com/documents/win0305a/. UNIXSTL has a comparatively
 embarassing paucity of variety, with only one performance counter based on
 gettimeofday.

 For a port to D I'd like to see the following:
     - someone who's expert on Linux to come up with equivalents to the WinSTL
 processtimes_counter, threadtimes_counter and highperformance_counter. The


first

 two depend on whether Linux maintains process/thread time stats; clearly it


does

 at least per-process timing, since the time utility does that. The latter


should

 probably be able to be effected using x86 commands, but I've not had chance to
 look at this yet.
     - me to define std/perf.d, something along the lines of (forgiving
 syntactical mistakes please):

 module std.perf;

 version(Windows)
 {
     private import std.windows.perf;
     public alias std.windows.perf.PerformanceCounter    PerformanceCounter;
 }
 else version(Linux)
 {
     private import std.linux.perf;
     public alias std.linux.perf.PerformanceCounter    PerformanceCounter;
 }
 else
 {
     // usual static assertion stuff
 }

 Maybe I can do this now, then, eh? :) The other classes -


HighPerformanceCounter,

 ThreadTimesCounter and ProcessTimesCounter can be brought into std.perf when
 there are implementations for Linux.

 I'm off to have a play with this now. You inspirer, you!


Apparently, it helps not to be an idiot. I'll try and try it sometime.

It transpires that I did most of what I've just said in mid-March. Doh!

I'll patch it up and pass it onto big-W, with hope in my heart and fingers
entwined.



 May 04 2004



  parent  C <qbert atari-soldiers.com>  writes:


sweet

On Wed, 5 May 2004 14:39:34 +1000, Matthew <matthew.hat stlsoft.dot.org> 
wrote:


 Another OT, are you planning on porting STLSoft's performance modules 


 to D

 ?  I was in need of this just today.


 Yes. I've asked Walter about this a couple of times, but his plate's 
 too small,
 and it's got too much seafood on it.

 What I'd like to do is have the following modules:

     std.perf
     std.linux.perf
     std.windows.perf

 The latter two would be straightforward port of the UNIXSTL
 (unixstl_performance_counter.h) and WinSTL classes


 (winstl_performance_counter.h,

 winstl_highperformance_counter.h, winstl_tick_counter.h,
 winstl_processtimes_counter.h, winstl_threadtimes_counter.h). WinSTL 
 has a


 number

 of flavours, the relative merits of which are described in
 http://www.windevnet.com/documents/win0305a/. UNIXSTL has a 
 comparatively
 embarassing paucity of variety, with only one performance counter based 
 on
 gettimeofday.

 For a port to D I'd like to see the following:
     - someone who's expert on Linux to come up with equivalents to the 
 WinSTL
 processtimes_counter, threadtimes_counter and highperformance_counter. 
 The


 first

 two depend on whether Linux maintains process/thread time stats; 
 clearly it


 does

 at least per-process timing, since the time utility does that. The 
 latter


 should

 probably be able to be effected using x86 commands, but I've not had 
 chance to
 look at this yet.
     - me to define std/perf.d, something along the lines of (forgiving
 syntactical mistakes please):

 module std.perf;

 version(Windows)
 {
     private import std.windows.perf;
     public alias std.windows.perf.PerformanceCounter    
 PerformanceCounter;
 }
 else version(Linux)
 {
     private import std.linux.perf;
     public alias std.linux.perf.PerformanceCounter    
 PerformanceCounter;
 }
 else
 {
     // usual static assertion stuff
 }

 Maybe I can do this now, then, eh? :) The other classes -


 HighPerformanceCounter,

 ThreadTimesCounter and ProcessTimesCounter can be brought into std.perf 
 when
 there are implementations for Linux.

 I'm off to have a play with this now. You inspirer, you!


 Apparently, it helps not to be an idiot. I'll try and try it sometime.

 It transpires that I did most of what I've just said in mid-March. Doh!

 I'll patch it up and pass it onto big-W, with hope in my heart and 
 fingers
 entwined.




-- 
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/



 May 05 2004



prev sibling  parent reply Brad Anderson <brad dsource.dot.org>  writes:


C wrote:

 Maybe we can throw them up on dsource.org for the mean time, brad ?
 


You're talking about posting drafts of articles for TDJ?  I have no 
problem with that.

Do you want a forum?  That would be the best, because you could post 
your article as a sticky new topic, and people could quote, comment.

Let me know if this is what you want, or if you want another tab across 
the top, and post html files of the articles.

BA



 May 04 2004



  parent reply C <qbert atari-soldiers.com>  writes:


 Do you want a forum?  That would be the best, because you could post 
 your article as a sticky new topic, and people could quote, comment.


That sounds , only drawback is no way to search for articles.  But I dont 
imagine that being an issue for a while, forum sounds good thanks BA.  
( thats right i did it! )

C

On Wed, 05 May 2004 01:27:18 -0500, Brad Anderson <brad dsource.dot.org> 
wrote:


 C wrote:

 Maybe we can throw them up on dsource.org for the mean time, brad ?


 You're talking about posting drafts of articles for TDJ?  I have no 
 problem with that.

 Do you want a forum?  That would be the best, because you could post 
 your article as a sticky new topic, and people could quote, comment.

 Let me know if this is what you want, or if you want another tab across 
 the top, and post html files of the articles.

 BA




-- 
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/



 May 05 2004



 next sibling parent  "Matthew" <matthew.hat stlsoft.dot.org>  writes:


Let's see how this goes for a while. It could be the best way to start TDJ. Once
there are several there, we can apply a bit of an editorial thing to it - weed
it
and reap - and maybe get the body of the first issue from there. Then we'd need
an editorial and a Word-From-Walter, and we're there!

I have a warm feeling about this ...

"C" <qbert atari-soldiers.com> wrote in message
news:opr7jj2vx6aoygh6 news.digitalmars.com...

 Do you want a forum?  That would be the best, because you could post
 your article as a sticky new topic, and people could quote, comment.


 That sounds , only drawback is no way to search for articles.  But I dont
 imagine that being an issue for a while, forum sounds good thanks BA.
 ( thats right i did it! )

 C

 On Wed, 05 May 2004 01:27:18 -0500, Brad Anderson <brad dsource.dot.org>
 wrote:


 C wrote:

 Maybe we can throw them up on dsource.org for the mean time, brad ?


 You're talking about posting drafts of articles for TDJ?  I have no
 problem with that.

 Do you want a forum?  That would be the best, because you could post
 your article as a sticky new topic, and people could quote, comment.

 Let me know if this is what you want, or if you want another tab across
 the top, and post html files of the articles.

 BA




 -- 
 Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/





 May 05 2004



prev sibling  parent  Brad Anderson <brad dsource.dot.org>  writes:


http://www.dsource.org/forums/viewforum.php?f=18

And you can search the entire forum section of the site.  May not help 
you search *only* articles.

BA

C wrote:

 Do you want a forum?  That would be the best, because you could post 
 your article as a sticky new topic, and people could quote, comment.


 
 
 That sounds , only drawback is no way to search for articles.  But I 
 dont imagine that being an issue for a while, forum sounds good thanks 
 BA.  ( thats right i did it! )
 
 C
 
 On Wed, 05 May 2004 01:27:18 -0500, Brad Anderson <brad dsource.dot.org> 
 wrote:
 

 C wrote:


 Maybe we can throw them up on dsource.org for the mean time, brad ?


 You're talking about posting drafts of articles for TDJ?  I have no 
 problem with that.

 Do you want a forum?  That would be the best, because you could post 
 your article as a sticky new topic, and people could quote, comment.

 Let me know if this is what you want, or if you want another tab 
 across the top, and post html files of the articles.

 BA


 
 
 
 





 May 06 2004