corelinux-public — General discussions of CoreLinux++

[Corelinux-public] State checked in

[Frank V. C. <fr...@co...>]

The last pattern requirement (State) has been implemented and checked
in.

It will be in the 0.4.22 release which should be a few days. I am having
trouble figuring out how LaTeX can find the latex2html styles.
-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package

Re: [Corelinux-public] packages

[Frank V. C. <fr...@co...>]

Christophe Prud'homme wrote:
> 
> Now it is packaging time
> redhat and debian packages should be expected next week
> both the doc will be packaged
> 
> libcorelinux     (.deb|.rpm): shared libs only (small packages)
> libcorelinux-dev (.deb|.rpm): for development (bigger package with
>                                headers and static libs)

Why the static libs?

> corelinux-doc (.deb|.rpm):complete web site and documentation!!!
>                           but I need to fix some license issues:
>                           put GPL or Open Publication License here
>                           and there

We may also want to segregate the docs up to the dvi, html, pdf, tex. As
a matter of fact, I was planning on doing this with the upcoming release
(0.4.22).

> 
> does anyone have comments on that?
> any debian/redhat official packager in the list?
> 
> regards
> christophe
> 

-- 
Frank V. Castellucci

[Corelinux-public] packages

[Christophe Prud'h. <pru...@MI...>]

Now it is packaging time
redhat and debian packages should be expected next week
both the doc will be packaged

libcorelinux     (.deb|.rpm): shared libs only (small packages)
libcorelinux-dev (.deb|.rpm): for development (bigger package with
                               headers and static libs)
corelinux-doc (.deb|.rpm):complete web site and documentation!!!
                          but I need to fix some license issues:
                          put GPL or Open Publication License here 
                          and there

does anyone have comments on that?
any debian/redhat official packager in the list?

regards
christophe

-- 
 Christophe Prud'homme                |
 MIT, 77, Mass Ave, Rm 3-243          |  C'est de la buche?
 Cambridge MA 02139                   |  Non c'est kloug!
 Tel (Office) : (00 1) (617) 253 0229 |  C'est colmatté avec du schpountz...
 Fax (Office) : (00 1) (617) 258 8559 |  -- Le Pere Noel est une ordure
 http://augustine.mit.edu/~prudhomm   |

Re: [Corelinux-public] documents draft

[Christophe Prud'h. <pru...@MI...>]

Ok the web is updated !
the document dafts are available here

<http://corelinux.sourceforge.net/draft.php>

enjoy
chris

Re: [Corelinux-public] documents draft

[Frank V. C. <fr...@co...>]

Christophe Prud'homme wrote:

>     Frank> I take it that:
>
>     Frank> 1. The link goes into headers.inc to ./drafts.php
>     Frank> 2. A new subdirectory is created /htdocs/drafts
>     Frank> 3. A drafts.php enumerates the current drafts (versions, etc.) and links
>     Frank> to /htdocs/drafts/xxxxxx.txt
>     Frank> 4. No Makefile needed, no conversions, etc.
> Yes
> I already did that on my own cvs version
> I can do an update if you want
>

Yes, if you have the time, I am tied up at work with some problems.

Re: [Corelinux-public] documents draft

[Christophe Prud'h. <pru...@MI...>]

    Frank> I take it that:

    Frank> 1. The link goes into headers.inc to ./drafts.php
    Frank> 2. A new subdirectory is created /htdocs/drafts
    Frank> 3. A drafts.php enumerates the current drafts (versions, etc.) and links
    Frank> to /htdocs/drafts/xxxxxx.txt
    Frank> 4. No Makefile needed, no conversions, etc.
Yes
I already did that on my own cvs version
I can do an update if you want

Re: [Corelinux-public] documents draft

[Frank V. C. <fr...@co...>]

Christophe Prud'homme wrote:
> 
> I think it should be a good idea to put
> the document draft on the web site
> 
> for example add a submenu drafts in developer's corner
> and added any draft in there
> 
> Frank I added this to my own copy of the website
> so you can have a look
> 
> regards
> 
> --
>  Christophe Prud'homme                |
>  MIT, 77, Mass Ave, Rm 3-243          |  C'est de la buche?
>  Cambridge MA 02139                   |  Non c'est kloug!
>  Tel (Office) : (00 1) (617) 253 0229 |  C'est colmatté avec du schpountz...
>  Fax (Office) : (00 1) (617) 258 8559 |  -- Le Pere Noel est une ordure
>  http://augustine.mit.edu/~prudhomm   |
> 

Chris, 

I take it that:

1. The link goes into headers.inc to ./drafts.php
2. A new subdirectory is created /htdocs/drafts
3. A drafts.php enumerates the current drafts (versions, etc.) and links
to /htdocs/drafts/xxxxxx.txt
4. No Makefile needed, no conversions, etc.

-- 
Frank V. Castellucci

Re: [Corelinux-public] documents draft

[Frank V. C. <fr...@co...>]

Christophe Prud'homme wrote:
> 
> I think it should be a good idea to put
> the document draft on the web site

I agree
 
> for example add a submenu drafts in developer's corner
> and added any draft in there
> 
> Frank I added this to my own copy of the website
> so you can have a look

Will do.

> 
> regards
> 
> --
>  Christophe Prud'homme                |
>  MIT, 77, Mass Ave, Rm 3-243          |  C'est de la buche?
>  Cambridge MA 02139                   |  Non c'est kloug!
>  Tel (Office) : (00 1) (617) 253 0229 |  C'est colmatté avec du schpountz...
>  Fax (Office) : (00 1) (617) 258 8559 |  -- Le Pere Noel est une ordure
>  http://augustine.mit.edu/~prudhomm   |
> 
> _______________________________________________
> Corelinux-public mailing list
> Cor...@li...
> http://lists.sourceforge.net/mailman/listinfo/corelinux-public

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package

[Corelinux-public] documents draft

[Christophe Prud'h. <pru...@MI...>]

I think it should be a good idea to put 
the document draft on the web site

for example add a submenu drafts in developer's corner
and added any draft in there

Frank I added this to my own copy of the website
so you can have a look

regards

-- 
 Christophe Prud'homme                |
 MIT, 77, Mass Ave, Rm 3-243          |  C'est de la buche?
 Cambridge MA 02139                   |  Non c'est kloug!
 Tel (Office) : (00 1) (617) 253 0229 |  C'est colmatté avec du schpountz...
 Fax (Office) : (00 1) (617) 258 8559 |  -- Le Pere Noel est une ordure
 http://augustine.mit.edu/~prudhomm   |

[Corelinux-public] web site and documentation update

[Christophe Prud'h. <pru...@MI...>]

Hi all,

The web site as been updated. New nice look. Now it should be easier to read the
documentation online and to navigate! All documentation is accessible through the same look and
feel. EVEN the analysis and design documentation.

GO  to <http://corelinux.sourceforge.net/index.php>

Moreover the documentation(all of it) has been changed. Now several format are available:
txt,ps,pdf,dvi and html ! We use the LaTeX format to handle the documentation.

So please give this new site a try and tell us what you think, what could be enhanced.

have a look for example at:
Functional Requirements <http://corelinux.sourceforge.net/doc/reqs.php>
and go to the design patterns... 

[Corelinux-public] SmartPointer draft 0.1

[Frank V. C. <fr...@co...>]

Ok, just to keep things organized, I started to put it into a future
specification document:

CoreLinux++ SmartPointer Draft 0.1
==================================

A SmartPointer is a lightweight proxy that substitutes itself for a
object instance until which point access requires said instance to be
instantiated.

A base SmartPointer should:
---------------------------

* Be thread safe. (Synchronized object)
* Have identity.
* Survive instance type queries (casting, testing).
* Be independent of faulting in the underlying instance. (Templated
methods)
* Be independent of it's own instantiation (Factory templated).

Pros
----
* Less memory required.
* Reduced heap fragmentation. 
* More scalable

Cons
----
* Require a less natural way of object referencing:
(*SmartPointerPtr)->instanceOperation(...)
* Require pervasive and strict management of references for example:

	Assuming that the underlying iterator increments the reference count

	for( iterator<SmartPointerPtr> aItr=...; 
           aItr->isValid(); 
           aItr->setNext() )
	{
	   SmartPointerPtr aObject( aItr->getElement() );
           (*aObject)->doSomething(...);
        }

would leak reference, unless it was something like:.

	for( iterator<SmartPointerPtr> aItr=...; 
           aItr->isValid(); 
           aItr->setNext() )
	{
	   SmartPointerPtr aObject( aItr->getElement() );
           (*aObject)->doSomething(...);
           aItr->returnElement( aObject );
        }

Issues
------
* There is a knee of the curve extension by using SmartPointers instead
of resident instances. But, it just moves the knee further out.

* Once the instance is instantiated, the SmartPointer goes away? My
experience has been that the solution space assumes constant use of the
smart pointer as proxy to instance.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package

Re: [Corelinux-public] SmartPointers

[Frank V. C. <fr...@co...>]

Good stuff, here are some thoughts (I haven't read through the
references in detail to make comparissons) about the fundemental
requirements of what I perceive to be flexible SmartPointers:

1. Must share an identity with the underlying object for persistence,
but can't be tied to a specific persist strategy.

2. Must support the hierarchy of the class instance it represents, but
should not have to be hand created for every object.

3. Can't be tied to a specific allocation strategy, this is left to the
strategy implementation that is implemented for the object.

In summary (for these points and there are more) a SmartPointer is
itself a framework with appropriate bridges to resolving run-time
strategy implementations.

Thoughts?

Dan Nuffer wrote:
> 
> There's also some interesting smart pointers from Alan Griffiths at:
> http://www.octopull.demon.co.uk/arglib/index.html
> 
> And boost.org has done a lot of work with smart pointers.  You can find
> some smart pointers at:
> http://www.boost.org/libraries.htm
> 
> and at:
> http://www.egroups.com/files/boost/
> 
> Mark Borgerding has witten up a paper about C++ smart pointer issues
> which is located at:
> http://www.egroups.com/files/boost/smart_pointers/SmartPtrShortList.html
> 
> --Dan Nuffer
> 
> Scott Haug wrote:
> >
> > There's an interesting implementation of a ref-counting smart pointer class in
> > the latest C/C++ Users Journal, by Vladimir Batov.  The source is available
> > here.  I'm not sure what you're looking for in an auto/smart pointer, but this
> > might be an interesting starting point.
> >
> > ftp://ftp.cuj.com/pub/2000/1806/batov.zip
> >
> > -Scott
> >
> > On Fri, May 12, 2000 at 09:31:16AM -0400, Frank V. Castellucci wrote:
> > >
> > > As I have seen SmartPointer implementations ranging from nothing more
> > > than a else named Flyweight to a very intelligent (at a price) Proxy for
> > > objects (including faulting, identification and persistent
> > > awareness/readiness/capabilities) it is neccassary to come to terms with
> > > the CoreLinux++ idea of said class.
> > >
> > > Ok, it's kicked off!
> > >
> > > --

-- 
Frank

Re: [Corelinux-public] SmartPointers

[Dan N. <dn...@ca...>]

There's also some interesting smart pointers from Alan Griffiths at:
http://www.octopull.demon.co.uk/arglib/index.html

And boost.org has done a lot of work with smart pointers.  You can find
some smart pointers at:
http://www.boost.org/libraries.htm

and at:
http://www.egroups.com/files/boost/

Mark Borgerding has witten up a paper about C++ smart pointer issues
which is located at:
http://www.egroups.com/files/boost/smart_pointers/SmartPtrShortList.html

--Dan Nuffer


Scott Haug wrote:
> 
> There's an interesting implementation of a ref-counting smart pointer class in
> the latest C/C++ Users Journal, by Vladimir Batov.  The source is available
> here.  I'm not sure what you're looking for in an auto/smart pointer, but this
> might be an interesting starting point.
> 
> ftp://ftp.cuj.com/pub/2000/1806/batov.zip
> 
> -Scott
> 
> On Fri, May 12, 2000 at 09:31:16AM -0400, Frank V. Castellucci wrote:
> >
> > As I have seen SmartPointer implementations ranging from nothing more
> > than a else named Flyweight to a very intelligent (at a price) Proxy for
> > objects (including faulting, identification and persistent
> > awareness/readiness/capabilities) it is neccassary to come to terms with
> > the CoreLinux++ idea of said class.
> >
> > Ok, it's kicked off!
> >
> > --
> > Frank V. Castellucci
> > http://corelinux.sourceforge.net
> > OOA/OOD/C++ Standards and Guidelines for Linux
> > http://PythPat.sourceforge.net
> > Pythons Pattern Package
> >
> > _______________________________________________
> > Corelinux-public mailing list
> > Cor...@li...
> > http://lists.sourceforge.net/mailman/listinfo/corelinux-public
> 
> --
> 
> _______________________________________________
> Corelinux-public mailing list
> Cor...@li...
> http://lists.sourceforge.net/mailman/listinfo/corelinux-public

Re: [Corelinux-public] SmartPointers

[Scott H. <sc...@id...>]

There's an interesting implementation of a ref-counting smart pointer class in
the latest C/C++ Users Journal, by Vladimir Batov.  The source is available
here.  I'm not sure what you're looking for in an auto/smart pointer, but this
might be an interesting starting point.

ftp://ftp.cuj.com/pub/2000/1806/batov.zip

-Scott

On Fri, May 12, 2000 at 09:31:16AM -0400, Frank V. Castellucci wrote:
> 
> As I have seen SmartPointer implementations ranging from nothing more
> than a else named Flyweight to a very intelligent (at a price) Proxy for
> objects (including faulting, identification and persistent
> awareness/readiness/capabilities) it is neccassary to come to terms with
> the CoreLinux++ idea of said class.
> 
> Ok, it's kicked off!
> 
> -- 
> Frank V. Castellucci
> http://corelinux.sourceforge.net
> OOA/OOD/C++ Standards and Guidelines for Linux
> http://PythPat.sourceforge.net
> Pythons Pattern Package
> 
> _______________________________________________
> Corelinux-public mailing list
> Cor...@li...
> http://lists.sourceforge.net/mailman/listinfo/corelinux-public

-- 

[Corelinux-public] SmartPointers

[Frank V. C. <fr...@co...>]

As I have seen SmartPointer implementations ranging from nothing more
than a else named Flyweight to a very intelligent (at a price) Proxy for
objects (including faulting, identification and persistent
awareness/readiness/capabilities) it is neccassary to come to terms with
the CoreLinux++ idea of said class.

Ok, it's kicked off!

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package

[Corelinux-public] Frameworks

[Frank V. C. <fr...@co...>]

There is enough of the core completed to start considering design and
implementation of frameworks. Many of the requirements in the
CoreLinux++ SourceForge requirement forum have been marked framework
bound, and I believe that there is a few constructs that should comprise
the first framework (it not really being a framework but a framework
utility library).

Some things that come to mind are:

SmartPointers, Meta Class (for objects and fundemental ontology), and
MemoryManagement Prototype.

I have been looking at MMXX (another sourceforge project) as a Meta
class potential but it is a heavy weight, not as heavy as CORBA
(although it may be too soon to tell) but it does reflect some
experience and thoughts on capturing type information for reasoning.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package

[Corelinux-public] Working to hit the milestone

[Frank V. C. <fr...@co...>]

There are a few major areas that need completion until I would consider
the core to be *1.0.0*:

* Completion of patterns : Very close, next week sometime.

* Completion of Semaphore: Timeout wait, event, broadcast and cross
process sharing are the major hurdles.

* Completion of Thread: Signals is the biggest hurdle, the management,
handling, and masking.

* Completion of Memory: Cross process sharing is the biggest hurdle.

* Implementation of Module: Fundemental shared library loading (correct
me if I am wrong).

* Implementation of System: Environment access, etc. Not seen as a major
problem (correct me if I am wrong).

* Documentation to PDF.

Semaphore, Thread, and Memory, being Linux specific, I could use some
help with.

While these are the majors, I don't see any reason not to start in with
some of the other building blocks for frameworks. I will take this up in
another message.

Does anybody have any feedback? I am starting to wonder if I should can
this mailing list.
-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package

[Corelinux-public] CoreLinux++ 0.4.21 Released

[Frank V. C. <fr...@co...>]

Includes:
	Subject/Observer
	Visitor
	Updates to Component (added accept( VisitorPtr )) with nop default.

Fixes:
	Examples 18-20 now compile through distribution check (thanks
Christophe).

This leaves two (2) patterns (State and Strategy) before the
implementable GoF pattern requirements are completed.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://PythPat.sourceforge.net
Pythons Pattern Package

Re: [Corelinux-public] corelinux 0.4.20 and cvs

[Frank V. C. <fr...@co...>]

Christophe Prud'homme wrote:
> 
> did it but How do i assign the bug to myself and how do I close it?
> 
> I think I have to  read the documentation.
> 
> regards
> christophe
> 
> _______________________________________________
> Corelinux-public mailing list
> Cor...@li...
> http://lists.sourceforge.net/mailman/listinfo/corelinux-public

The problem was that I had you as technician (default) for Bug Tracking.
You now have the authority to change, assign, etc.

FYI: Do NOT delete a defect when it is fixed, please close it instead.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux

Re: [Corelinux-public] corelinux 0.4.20 and cvs

[Christophe Prud'h. <pru...@MI...>]

did it but How do i assign the bug to myself and how do I close it?

I think I have to  read the documentation.

regards
christophe

Re: [Corelinux-public] corelinux 0.4.20 and cvs

[Frank V. C. <fr...@co...>]

Christophe Prud'homme wrote:
> 
> yes
> 
> the problem is in the examples
> If I create a distribution the include dirs are missing and the compile
> flags are not correct CPPFLAGS in fact
> 
> I just commited the changes
> 
> to make sure that everything is right do a
> 
>         make distcheck
> 
> it will create the tarball and compile everything

Ok, please open a defect and close it with the fix noted. I want to keep
track of this kind of thing. I will update the developers guide to
reflect this process.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux

Re: [Corelinux-public] corelinux 0.4.20 and cvs

[Christophe Prud'h. <pru...@MI...>]

yes

the problem is in the examples
If I create a distribution the include dirs are missing and the compile
flags are not correct CPPFLAGS in fact

I just commited the changes

to make sure that everything is right do a 

	make distcheck

it will create the tarball and compile everything

Re: [Corelinux-public] corelinux 0.4.20 and cvs

[Frank V. C. <fr...@co...>]

Christophe Prud'homme wrote:
> 
> Hi
> 
> the examples, exmap17 to examp20, in corelinux 0.4.20 do not compile
> 
> you have to wait the 0.4.21(not yet released) or get the CVS source
> or change configure.in: add $top_srcdir in CPPFLAGS
> 
> regards
> christophe
> 

Do you mean that the distribution did not compile? Or that during the
changes I have been making I broke something? I did not see a defect
reporting this in the system.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux

[Corelinux-public] corelinux 0.4.20 and cvs

[Christophe Prud'h. <pru...@MI...>]

Hi

the examples, exmap17 to examp20, in corelinux 0.4.20 do not compile

you have to wait the 0.4.21(not yet released) or get the CVS source
or change configure.in: add $top_srcdir in CPPFLAGS

regards
christophe

-- 
  Christophe Prud'homme                 |
  MIT, 77, Mass Ave, Rm 3-243           |   A wise person makes his own
  Cambridge MA 02139                    |   decisions,
  Tel (Office) : (00 1) (617) 253 0229  |   a weak one obeys public opinion.
  Fax (Office) : (00 1) (617) 258 8559  |   -- Chinese proverb
  http://augustine.mit.edu/~prudhomm    |

[Corelinux-public] Subject Observer checked in

[Frank V. C. <fr...@co...>]

Analysis, design, implementation. Working on test code, which will
appear in 0.4.21 (US-Monday?)

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux