ooc-compiler Mailing List for Optimizing Oberon-2 Compiler (Page 5)

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

do I understand right: OOC is orphaned 

then the question is not: 

Dead or just comatose?

but better

How to reanimate it and who will lead the intensive care unit?
--
  Bernhard 

On 28 July 2010 16:05, Duke Normandin <duk...@ml...> wrote:
> [...]
>> for myself I can say that I stopped working with Oberon several
>> years ago.  By now it's quite hard for me to even read and
>> understand Oberon code.  Go figure...
>
> Well! You're no fun at all, are you! ;) Do you remember how mature ooc
> was when you were last involved with it?

Actually no, I don't remember.  Nowadays I have fun with Clojure :-)

-- mva

On Wed, 28 Jul 2010, Treutwein Bernhard wrote:

> hmm,
>
> > Does anybody know about signs of Oberon lifeforms
> > elsewere in webspace?
>
> ofcourse, there are the semaphores of Zürich, namely
> http://www.ocp.inf.ethz.ch/

Sure! There's one or 2 zealots lurking there! ;)

[snip]

> I do not know what happened to OOC/OO2C project founder Michael van
> Acken ... nor if any of the other admins (Jürgen Zimmermann, guenne, or
> Stewart Greenhill) are still active ...

Michael and Stewart are still around, and probably lurking this list
;) Also, they might be on vacation at the moment.

The comp.lang.oberon News Group is another resources that may help
out. At the moment, the NG is probably the most compiler-independent
Oberon resource. IMHO, OCP should be just that, via the Forum, simply
by adding a few new categories, like "VisualOberon", "ooc/oo2c", Mike
Spivey's "obc", etc. One stop, Oberon shopping. ;)
-- 
On Wed, 28 Jul 2010, Michael van Acken wrote:

> On 28 July 2010 15:08, Treutwein Bernhard
> <Ber...@ve...> wrote:
> > [...]
> > I do not know what happened to OOC/OO2C project founder Michael van
> > Acken ... nor if any of the other admins (Jürgen Zimmermann, guenne, or
> > Stewart Greenhill) are still active ...
>
> Hi all,
>
> for myself I can say that I stopped working with Oberon several
> years ago.  By now it's quite hard for me to even read and
> understand Oberon code.  Go figure...

Well! You're no fun at all, are you! ;) Do you remember how mature ooc
was when you were last involved with it?
-- 
Duke
On 28 July 2010 15:08, Treutwein Bernhard
<Ber...@ve...> wrote:
> [...]
> I do not know what happened to OOC/OO2C project founder Michael van
> Acken ... nor if any of the other admins (Jürgen Zimmermann, guenne, or
> Stewart Greenhill) are still active ...

Hi all,

for myself I can say that I stopped working with Oberon several
years ago.  By now it's quite hard for me to even read and
understand Oberon code.  Go figure...

Jürgen did not continue with this project after our university time.
guenne I just gave access for some other work, unrelated to OOC.

> PS: interestingly this list is configured to private responses by default (at
> least when replied with M$ Outlook) ...

Yes, I have to hit "reply to all" as well.

-- mva

hmm,

> Does anybody know about signs of Oberon lifeforms 
> elsewere in webspace? 

ofcourse, there are the semaphores of Zürich, namely
http://www.ocp.inf.ethz.ch/ 
and the quite active, but difficult to watch and digest sites in Russia
http://oberoncore.ru
and BlackBox http://oberon.ch/blackbox.html (it runs nicely under Linux/Wine)
with its community http://zinnamturm.eu/

I do not know what happened to OOC/OO2C project founder Michael van 
Acken ... nor if any of the other admins (Jürgen Zimmermann, guenne, or 
Stewart Greenhill) are still active ...

regards
--
  Bernhard 

PS: interestingly this list is configured to private responses by default (at
least when replied with M$ Outlook) ...

Upon my last post indicating a possible compiler error, which is a 
strong topic, but I received just 2 private emails and saw not one 
response on the list. Is this list dead or just comatose?

Does anybody know about signs of Oberon lifeforms elsewere in webspace? 
A google seach yields a lot of hits of course, but I am curious about 
activity.
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

Hi there,

I guess I found a compiler error. I have a module L which implements a 
rudimentary list and a module Test which imports L:

MODULE L;
IMPORT Out;

TYPE
   Item* =     POINTER TO ItemD;
   ItemD* =    RECORD
                 next-:      Item;
               END;

   Header* =   RECORD
                 first-, last-:      Item;
               END;

PROCEDURE IsEmpty* (list:Header): BOOLEAN;
BEGIN RETURN list.first=NIL
END IsEmpty;

PROCEDURE Init* (VAR list:Header);
BEGIN
   list.first:= NIL;  list.last:= NIL;
   Out.String("List initialized");  Out.Ln;
END Init;

END L.

MODULE Test;
IMPORT L, Out;

TYPE
   Node  =     POINTER TO NodeD;
   NodeD  =    RECORD (L.ItemD)
                 sons-:  L.Header;
               END;
VAR n1: Node;

PROCEDURE (node:Node) Init;
BEGIN
   L.Init (node.sons);
   IF ~L.IsEmpty(node.sons) THEN
     Out.String("n1.sons not empty!"); Out.Ln;
   END
END Init;

BEGIN
   NEW (n1);
   n1.Init;
END Test.

Executing Test prints the following:

List initialized
n1.sons not empty!

which means that n1.Init didn't change n1. This is an error, isn't it?
If details matter I can tell them (platform, versions etc.).

-- 
Frank Hrebabetzky
Florianopolis, Brazil

Hi Duke,

[...]
>  sudo env "CC=/usr/bin/gcc-4.2" "CFLAGS=-O2" 
> "CPPFLAGS=-no-cpp-precomp -DGC_DARWIN_THREADS 
> -fnested-functions" ./configure --enable-threads=pthreads
> 
> I'm not very savvy with C and its tools like "make" et al. Every time
> I try to build "ooc" now, all I get is the above - the choke
> point. The other long and verbose compilation seems to be
> by-passed. Is there a way to start from scratch, other than to edit
> one of the source files?
> 

neither I'm very familiar with the make process, but I mastered it
successfully several times. As far as I know, it is neither necessary nor
recommended to sudo the configure & compile step, it should only be necessary
to sudo the "make install".

Now back:

You should be able to do a "make distclean" to cleanup everything, which was
configured. Maybe you need to sudo this step to everything cleaned.

Then a

./configure

make

sudo make install

should get you through the plain process.

In your case the configure step appears to be:

env "CC=/usr/bin/gcc-4.2" "CFLAGS=-O2" "CPPFLAGS=-no-cpp-precomp
-DGC_DARWIN_THREADS -fnested-functions" ./configure --enable-threads=pthreads

regards
--
  Bernhard

> -----Original Message-----
> From: Duke Normandin [mailto:duk...@ml...] 
> Sent: Monday, June 21, 2010 3:57 PM
> To: ooc Compiler
> Subject: [ooc-compiler] Problem Building ooc on MacOS X Leopard 10.5.8
> 
> 
> Good Morning!
> 
> Need help solving this please:
> 
> dnormandin@ ~/programming/oberon/oo2c_64-2.1.11
> 07:30 am >> make
> sed -e 's?%libdir%?/usr/local/lib?g' \
> -e 's?%oocdir%?/usr/local/lib/oo2c?g' \
> -e 's?%bindir%?/usr/local/bin?g' \
> -e 's?%INSTALL%?/usr/bin/install -c?g' \
> -e 's?%INSTALL_PROGRAM%?/usr/bin/install-c?g' \
> -e 's?%INSTALL_DATA%?/usr/bin/install	-c -m 644?g' \
> /Users/dnormandin/programming/oberon/oo2c_64-2.1.11/rsrc/OOC/o
> o2crc.xml.mk>/Users/dnormandin/programming/oberon/oo2c_64-2.1.
> 11/rsrc/OOC/oo2crc.xml
> sed	-e's:<file-system>:<!--:g' \
> -e 's:</file-system>:-->:g' \
> -e
> 's:<repositories>:<repositories><file-system>/Users/dnormandin
> /programming/oberon/oo2c_64-2.1.11/lib</file-system><file-syst
em>/Users/dnormandin/programming/oberon/oo2c_64-> 2.1.11</file-system>:' \
> /Users/dnormandin/programming/oberon/oo2c_64-2.1.11/rsrc/OOC/o
> o2crc.xml 
> >/Users/dnormandin/programming/oberon/oo2c_64-2.1.11/oo2crc-in
> stall.xml
> stage0/oo2c
> --config
> oo2crc-install.xml
> -v
> -r
> lib
> -r
> .
> --build-package
> liboo2c
> make:
> ***
> [lib/obj/liboo2c.la]
> Error 1
> 
> I configured with:
> 
>  sudo env "CC=/usr/bin/gcc-4.2" "CFLAGS=-O2" 
> "CPPFLAGS=-no-cpp-precomp -DGC_DARWIN_THREADS 
> -fnested-functions" ./configure --enable-threads=pthreads
> 
> I'm not very savvy with C and its tools like "make" et al. Every time
> I try to build "ooc" now, all I get is the above - the choke
> point. The other long and verbose compilation seems to be
> by-passed. Is there a way to start from scratch, other than to edit
> one of the source files?
> 
> *Sidebar*
> I cannot believe how mauch I'm enjoying programming in Oberon! I also
> have the Oberon system Aos/A2 working on my Mac. That's cool too.
> 
> --
> Duke Normandin
> Alberta, Canada
> -- 
> 
> --------------------------------------------------------------
> ----------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate 
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
> lucky parental unit.  See the prize list and enter to win: 
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> ooc-compiler mailing list
> ooc...@li...
> https://lists.sourceforge.net/lists/listinfo/ooc-compiler
> 

Good Morning!

Need help solving this please:

dnormandin@ ~/programming/oberon/oo2c_64-2.1.11
07:30 am >> make
sed -e 's?%libdir%?/usr/local/lib?g' \
-e 's?%oocdir%?/usr/local/lib/oo2c?g' \
-e 's?%bindir%?/usr/local/bin?g' \
-e 's?%INSTALL%?/usr/bin/install -c?g' \
-e 's?%INSTALL_PROGRAM%?/usr/bin/install-c?g' \
-e 's?%INSTALL_DATA%?/usr/bin/install	-c -m 644?g' \
/Users/dnormandin/programming/oberon/oo2c_64-2.1.11/rsrc/OOC/oo2crc.xml.mk>/Users/dnormandin/programming/oberon/oo2c_64-2.1.11/rsrc/OOC/oo2crc.xml
sed	-e's:<file-system>:<!--:g' \
-e 's:</file-system>:-->:g' \
-e
's:<repositories>:<repositories><file-system>/Users/dnormandin/programming/oberon/oo2c_64-2.1.11/lib</file-system><file-system>/Users/dnormandin/programming/oberon/oo2c_64-2.1.11</file-system>:' \
/Users/dnormandin/programming/oberon/oo2c_64-2.1.11/rsrc/OOC/oo2crc.xml >/Users/dnormandin/programming/oberon/oo2c_64-2.1.11/oo2crc-install.xml
stage0/oo2c
--config
oo2crc-install.xml
-v
-r
lib
-r
.
--build-package
liboo2c
make:
***
[lib/obj/liboo2c.la]
Error 1

I configured with:

 sudo env "CC=/usr/bin/gcc-4.2" "CFLAGS=-O2" "CPPFLAGS=-no-cpp-precomp -DGC_DARWIN_THREADS -fnested-functions" ./configure --enable-threads=pthreads

I'm not very savvy with C and its tools like "make" et al. Every time
I try to build "ooc" now, all I get is the above - the choke
point. The other long and verbose compilation seems to be
by-passed. Is there a way to start from scratch, other than to edit
one of the source files?

*Sidebar*
I cannot believe how mauch I'm enjoying programming in Oberon! I also
have the Oberon system Aos/A2 working on my Mac. That's cool too.

--
Duke Normandin
Alberta, Canada
-- 

http://fumalih.tripod.com/

-- 
Fitz
Cell: (602) 803-7695
http://www.linkedin.com/in/fitzfitzpatrick

Hi,

How can I find out the amount of memory allocated by my program from 
within my program?

My program gets killed, but before that it allocates large amounts of 
memory, and the system starts swapping and I guess it has to do with 
that. So I would like to monitor the allocated memory and terminate in a 
clean way.
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

I used the defaults for installation of oo2c on Windows with MinGW and 
without Cygwin. Searching for 'oo2crc' (Windows Find utility, so search 
corresponds to '*oo2crc*') yields only 'oo2crc.xml'.

The OOC reference manual says oo2c emits an error message if there is no 
initialization file. But it doesn't complain.

On my system, I have:
C:\usr\local\etc
C:\Programs\MinGW\etc
C:\Programs\MSys\etc
C:\Programs\GnuWin32	- without etc

So:
Why doesn't oo2c complain?
Where should I put the global initialization file?
What should be its name?
How can I check that it is actually being read?

-- 
Frank Hrebabetzky	Tel. (48) 3239 2258
Photonita Ltda.		http://www.photonita.com.br
Brazil

Is there a way to import constants and variables from C libraries via
interface modules? I am talking about header file lines like:

extern const sometype someconstant;
extern sometype somevariable;

If not, it has to be done in foreign modules, I guess, right?
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

On 20 Oct 2009 08:49:38 +0200
"Tim Teulings" <ra...@ed...> wrote:

> Hello!
> 
> > After learning a bit more about the C preprocessor, I think the
> > ncurses.h lines I posted yesterday boil down to
> > 
> > ----------------------------------------------------
> > extern unsigned acs_map[]
> > #define NCURSES_ACS(c) (acs_map[(unsigned char)(c)])
> > ----------------------------------------------------
> > 
> > So I need to access the variable 'acs_map'. How do I do that?
> > (I only found hints in the OOC documentation on how to handle types
> > in interface modules, but not variables.)
> 
> I did implement interfaces to Curses years ago as a foreign module
> (since curses does heavy use of macros).
> 
> http://visualoberon.cvs.sourceforge.net/viewvc/visualoberon/Curses/
> 
> Perhaps this is helpful.
> 

Looks very promising. I'll try it out within a few days.
Thanks a lot! 
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

Hello!

> After learning a bit more about the C preprocessor, I think the
> ncurses.h lines I posted yesterday boil down to
> 
> ----------------------------------------------------
> extern unsigned acs_map[]
> #define NCURSES_ACS(c) (acs_map[(unsigned char)(c)])
> ----------------------------------------------------
> 
> So I need to access the variable 'acs_map'. How do I do that?
> (I only found hints in the OOC documentation on how to handle types in
> interface modules, but not variables.)

I did implement interfaces to Curses years ago as a foreign module
(since curses does heavy use of macros).

http://visualoberon.cvs.sourceforge.net/viewvc/visualoberon/Curses/

Perhaps this is helpful.

-- 
Gruß...
       Tim

After learning a bit more about the C preprocessor, I think the
ncurses.h lines I posted yesterday boil down to

----------------------------------------------------
extern unsigned acs_map[]
#define NCURSES_ACS(c) (acs_map[(unsigned char)(c)])
----------------------------------------------------

So I need to access the variable 'acs_map'. How do I do that?
(I only found hints in the OOC documentation on how to handle types in
interface modules, but not variables.)
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

I am struggling against C header files again. Now it is ncurses.h which
exceeds my C preprocessor knowledge. I deduced ('guessed' would be
closer to reality) that

-----------------------------------------------------
CONST
  ACS_ULCORNER* = LONG(ORD("l"));

TYPE
  chtype*       = LONGINT;

PROCEDURE mvaddch* (y,x:C.int; ch:chtype): C.int;
-----------------------------------------------------

should be the correspondence of

-----------------------------------------------------
#if 0 && defined(_LP64)
typedef unsigned chtype;
typedef unsigned mmask_t;
#else
typedef unsigned long chtype;
typedef unsigned long mmask_t;
#endif

#ifdef __cplusplus
extern "C" {
#define NCURSES_CAST(type,value) static_cast<type>(value)
#else
#define NCURSES_CAST(type,value) (type)(value)
#endif

#if 0 || 0
NCURSES_WRAPPED_VAR(chtype*, acs_map);
#define acs_map (_nc_acs_map())
#else
extern NCURSES_EXPORT_VAR(chtype) acs_map[];
#endif

#define NCURSES_ACS(c)	(acs_map[NCURSES_CAST(unsigned char,c)])

#define ACS_ULCORNER	NCURSES_ACS('l') /* upper left corner */

extern NCURSES_EXPORT(int) mvaddch (int, int, const chtype);
---------------------------------------------------------------

This is from /usr/include/ncurses.h on Ubuntu 8.04 running on a 32 bit
processor (AMD Athlon), but I think the corresponding Oberon
implementation should be the same for each Ubuntu or Debian version.
What I know for shure is that my attempt is wrong, because

res:= nc.mvaddch (y, x, nc.ACS_ULCORNER);

prints out an 'l' instead of an upper left corner.

Can anybody please give me a hint?
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

Hi Frank,

H2O shares some modules with the OOC compiler. These include the 
configuration modules, which are not part of the source distribution, 
but are generated automatically during configuration. Try doing the 
normal configuration process "./configure ..." before you run "oo2c -M 
TestH2O".  That should generate the required files for you.

Cheers,
   Stewart

Frank Hrebabetzky wrote:
> On Mon, 28 Sep 2009 22:33:02 +0800
> Stewart Greenhill <sgr...@us...> wrote:
> 
> Hi,
> 
> Thanks for your hints, but I am still trying the first step. Berhards
> instructions tell how to install the whole package, but I don't want to
> overwrite my working one. So I downloaded and extracted the
> tarball oo2c_32.2.1.11.tar.bz2, then I
> - mkdir temp
> - cp oo2c_32.2.1.11/src temp
> - cd temp
> - oo2c -M TestH2O
> This resulted in a bunch of warnings about unused objects and
> terminated with the line:
> 
> src/OOC/Config.Mod:6:54: Cannot locate module `OOC:Config:Autoconf'
> 
> Regards,

On Mon, 28 Sep 2009 22:33:02 +0800
Stewart Greenhill <sgr...@us...> wrote:

Hi,

Thanks for your hints, but I am still trying the first step. Berhards
instructions tell how to install the whole package, but I don't want to
overwrite my working one. So I downloaded and extracted the
tarball oo2c_32.2.1.11.tar.bz2, then I
- mkdir temp
- cp oo2c_32.2.1.11/src temp
- cd temp
- oo2c -M TestH2O
This resulted in a bunch of warnings about unused objects and
terminated with the line:

src/OOC/Config.Mod:6:54: Cannot locate module `OOC:Config:Autoconf'

Regards,
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

> Hi Frank and Bernhard,
> 
> Thanks Bernhard for putting that documentation together. H2O is 
> unfinished, but is still usable within limits.
> 
> The main problem it has is dealing with the myriad ways that the 
> language is mangled in header declarations. Many of the base files
> can include nasty bits of code including templates, assembler, etc. 
> Therefore, its usually necessary to eliminate these by a process of 
> trial and error. Unfortunately, the error reporting is often not as
> good as it could be, so it can be hard to find out where the
> translator runs into difficulty. Most APIs like GTK are reasonably
> stable; it tends to be the base files that change over time.
> 
> I've attached an example configuration file, which I last used to 
> translate GTK, probably on a Windows system. Note the list of
> "Excludes" and the definitions required to replace them.
> 
> If you're interested in GTK, I think its probably not too difficult
> to get something working. GTK is coded for a C-language interface,
> which makes it an easy target. Let me know if you need help with it.
> I've got a current Debian system I can test it out on.
> 
> Cheers,
>    Stewart
> 
> Treutwein Bernhard wrote:
> > Hi Frank,
> > 
> > hopefully Stewart also repllies, maybe a conjoint effort could
> > yield some more documentation for H2O, but to get started take a
> > look at the enclosed pdf. It contains my short notices about
> > installation and Stewart's hints about usage ...
> > 
> > 
> > I don't have a lot experiences with OOC and H2O it, but I
> > successfully compiled it under Cygwin and I have a setup for using
> > it with BlackBox/ComponentPascal. That is also the reason for the
> > pdf, I've written up my notes in standard format of the BlaxBox
> > Component Framework (see http://www.oberon.ch/BlackBox.html).
> > 
> > I've used it only very occasional.
> > 
> > regards
> > --
> >   Bernhard Treutwein
> >   e-mail: bernhard treutwein (at) verwaltung uni-muenchen de
> >  
> > 
> >> -----Original Message-----
> >> From: Frank Hrebabetzky [mailto:hr...@te...] 
> >> Sent: Saturday, September 26, 2009 7:52 PM
> >> To: ooc...@li...
> >> Subject: [ooc-compiler] How to use H2O?
> >>
> >>
> >> Hi all,
> >>
> >> For my actual toy program I need a GTK interface module which I am
> >> completing gradually. A (direct) response to my last
> >> question ([ooc-compiler] Bug? Record field changes on its own)
> >> convinced me that the error lies in the interface module. To 
> >> check it I
> >> would like to generate the interface with H2O and compare it with
> >> my own. My questions are:
> >>
> >> 1. How to install H2O?
> >>    I didn't compile oo2c from source, but installed the Ubuntu
> >> repository version. I would like to keep it like this because
> >> everything works well. I found the H2O sources on
> >> oo2c_32-21.9.tar.bz2. What do I do with them?
> >>
> >> 2. How to use H20?
> >> Will installation result in an executable which I call like
> >> 'h20 /usr/share/include' and out come miraculously the interfaces
> >> to all the header files on my machine?
> >> Or do I have to write my own program calling imported H2O 
> >> modules in the
> >> right manner (parsing the correct header files in the correct 
> >> sequence,
> >> then generating the interface code), or something like that?
> >> -- 
> >> Frank Hrebabetzky		+55 / 48 / 3235 1106
> >> Florianopolis, Brazil
> >>
> >> --------------------------------------------------------------
> >> ----------------
> >> Come build with us! The BlackBerry&reg; Developer Conference in
> >> SF, CA is the only developer event you need to attend this year. 
> >> Jumpstart your
> >> developing skills, take BlackBerry mobile applications to 
> >> market and stay 
> >> ahead of the curve. Join us from November 9&#45;12, 2009. 
> >> Register now&#33;
> >> http://p.sf.net/sfu/devconf
> >> _______________________________________________
> >> ooc-compiler mailing list
> >> ooc...@li...
> >> https://lists.sourceforge.net/lists/listinfo/ooc-compiler
> >>
> 

Hi Frank and Bernhard,

Thanks Bernhard for putting that documentation together. H2O is 
unfinished, but is still usable within limits.

The main problem it has is dealing with the myriad ways that the 
language is mangled in header declarations. Many of the base files can 
include nasty bits of code including templates, assembler, etc. 
Therefore, its usually necessary to eliminate these by a process of 
trial and error. Unfortunately, the error reporting is often not as good 
as it could be, so it can be hard to find out where the translator runs 
into difficulty. Most APIs like GTK are reasonably stable; it tends to 
be the base files that change over time.

I've attached an example configuration file, which I last used to 
translate GTK, probably on a Windows system. Note the list of "Excludes" 
and the definitions required to replace them.

If you're interested in GTK, I think its probably not too difficult to 
get something working. GTK is coded for a C-language interface, which 
makes it an easy target. Let me know if you need help with it. I've got 
a current Debian system I can test it out on.

Cheers,
   Stewart

Treutwein Bernhard wrote:
> Hi Frank,
> 
> hopefully Stewart also repllies, maybe a conjoint effort could yield some
> more documentation for H2O, but to get started take a look at the enclosed
> pdf. It contains my short notices about installation and Stewart's hints
> about usage ...
> 
> 
> I don't have a lot experiences with OOC and H2O it, but I  successfully
> compiled it under Cygwin and I have a setup for using it with
> BlackBox/ComponentPascal. That is also the reason for the pdf, I've written
> up my notes in standard format of the BlaxBox Component Framework (see
> http://www.oberon.ch/BlackBox.html).
> 
> I've used it only very occasional.
> 
> regards
> --
>   Bernhard Treutwein
>   e-mail: bernhard treutwein (at) verwaltung uni-muenchen de
>  
> 
>> -----Original Message-----
>> From: Frank Hrebabetzky [mailto:hr...@te...] 
>> Sent: Saturday, September 26, 2009 7:52 PM
>> To: ooc...@li...
>> Subject: [ooc-compiler] How to use H2O?
>>
>>
>> Hi all,
>>
>> For my actual toy program I need a GTK interface module which I am
>> completing gradually. A (direct) response to my last
>> question ([ooc-compiler] Bug? Record field changes on its own)
>> convinced me that the error lies in the interface module. To 
>> check it I
>> would like to generate the interface with H2O and compare it with my
>> own. My questions are:
>>
>> 1. How to install H2O?
>>    I didn't compile oo2c from source, but installed the Ubuntu
>> repository version. I would like to keep it like this because
>> everything works well. I found the H2O sources on
>> oo2c_32-21.9.tar.bz2. What do I do with them?
>>
>> 2. How to use H20?
>> Will installation result in an executable which I call like
>> 'h20 /usr/share/include' and out come miraculously the interfaces to
>> all the header files on my machine?
>> Or do I have to write my own program calling imported H2O 
>> modules in the
>> right manner (parsing the correct header files in the correct 
>> sequence,
>> then generating the interface code), or something like that?
>> -- 
>> Frank Hrebabetzky		+55 / 48 / 3235 1106
>> Florianopolis, Brazil
>>
>> --------------------------------------------------------------
>> ----------------
>> Come build with us! The BlackBerry&reg; Developer Conference in SF, CA
>> is the only developer event you need to attend this year. 
>> Jumpstart your
>> developing skills, take BlackBerry mobile applications to 
>> market and stay 
>> ahead of the curve. Join us from November 9&#45;12, 2009. 
>> Register now&#33;
>> http://p.sf.net/sfu/devconf
>> _______________________________________________
>> ooc-compiler mailing list
>> ooc...@li...
>> https://lists.sourceforge.net/lists/listinfo/ooc-compiler
>>

Hi Frank,

hopefully Stewart also repllies, maybe a conjoint effort could yield some
more documentation for H2O, but to get started take a look at the enclosed
pdf. It contains my short notices about installation and Stewart's hints
about usage ...

I don't have a lot experiences with OOC and H2O it, but I  successfully
compiled it under Cygwin and I have a setup for using it with
BlackBox/ComponentPascal. That is also the reason for the pdf, I've written
up my notes in standard format of the BlaxBox Component Framework (see
http://www.oberon.ch/BlackBox.html).

I've used it only very occasional.

regards
--
  Bernhard Treutwein
  e-mail: bernhard treutwein (at) verwaltung uni-muenchen de

> -----Original Message-----
> From: Frank Hrebabetzky [mailto:hr...@te...] 
> Sent: Saturday, September 26, 2009 7:52 PM
> To: ooc...@li...
> Subject: [ooc-compiler] How to use H2O?
> 
> 
> Hi all,
> 
> For my actual toy program I need a GTK interface module which I am
> completing gradually. A (direct) response to my last
> question ([ooc-compiler] Bug? Record field changes on its own)
> convinced me that the error lies in the interface module. To 
> check it I
> would like to generate the interface with H2O and compare it with my
> own. My questions are:
> 
> 1. How to install H2O?
>    I didn't compile oo2c from source, but installed the Ubuntu
> repository version. I would like to keep it like this because
> everything works well. I found the H2O sources on
> oo2c_32-21.9.tar.bz2. What do I do with them?
> 
> 2. How to use H20?
> Will installation result in an executable which I call like
> 'h20 /usr/share/include' and out come miraculously the interfaces to
> all the header files on my machine?
> Or do I have to write my own program calling imported H2O 
> modules in the
> right manner (parsing the correct header files in the correct 
> sequence,
> then generating the interface code), or something like that?
> -- 
> Frank Hrebabetzky		+55 / 48 / 3235 1106
> Florianopolis, Brazil
> 
> --------------------------------------------------------------
> ----------------
> Come build with us! The BlackBerry&reg; Developer Conference in SF, CA
> is the only developer event you need to attend this year. 
> Jumpstart your
> developing skills, take BlackBerry mobile applications to 
> market and stay 
> ahead of the curve. Join us from November 9&#45;12, 2009. 
> Register now&#33;
> http://p.sf.net/sfu/devconf
> _______________________________________________
> ooc-compiler mailing list
> ooc...@li...
> https://lists.sourceforge.net/lists/listinfo/ooc-compiler
> 

Hi all,

For my actual toy program I need a GTK interface module which I am
completing gradually. A (direct) response to my last
question ([ooc-compiler] Bug? Record field changes on its own)
convinced me that the error lies in the interface module. To check it I
would like to generate the interface with H2O and compare it with my
own. My questions are:

1. How to install H2O?
   I didn't compile oo2c from source, but installed the Ubuntu
repository version. I would like to keep it like this because
everything works well. I found the H2O sources on
oo2c_32-21.9.tar.bz2. What do I do with them?

2. How to use H20?
Will installation result in an executable which I call like
'h20 /usr/share/include' and out come miraculously the interfaces to
all the header files on my machine?
Or do I have to write my own program calling imported H2O modules in the
right manner (parsing the correct header files in the correct sequence,
then generating the interface code), or something like that?
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

I have a problem with this bug (?) report: (1) My program has grown
pretty large meanwhile, and (2) it has a GUI based on GTK and the
record field change mentioned above occurs between 2 mouse clicks. So it
is pretty difficult to isolate the error into a simple example.

The program is a game containing a variable of type

  State*      = POINTER TO StateD;
  StateD*     = RECORD [ABSTRACT]
                  nPl-, movPl-:   SHORTINT;  (* # players, moving pl.*)
		  active-, aiPl-: SET;       (* active, AI players *)
		END;

When I press the start button, state.movPl has the correct value, 0.
Immediately upon the next action, t.i. clicking on the playfield, the
value has changed to 62 without any appearent reason. Now when I change
the sequence of the 2 SHORTINTs to

  StateD*     = RECORD [ABSTRACT]
                  movPl-, nPl-:   SHORTINT;  (* # players, moving pl.*)

then the value changes to 40.

Is it a bug? How can I find that out?
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

On Fri, 07 Aug 2009 10:08:05 +0800
Stewart Greenhill <sgr...@us...> wrote:

Hi Stewart,

> The most common cause of segmentation errors is uninitialised
> pointers. NIL pointers are checked but the run-time system, but
> dereference checks can't differentiate between random garbage on the
> heap and a valid pointer.
> 
> Is is possible that "moves.first" has not been initialised?

That was indeed the problem. I relied on Oberon to initialize all
pointers to NULL, but this is not the case if the variable is local to
a procedure as in my appended example. It wouldn't make sense anyway to
initialize a local variable at the first call, when it can have an
arbitrary value on all further calls. But: Is that documented anywhere?

> The best way to get some error context is to use gdb.

I got lost in the C-translations of structures, pointers and their
extensions, so I fell back on the traditional Oberon debugger: MODULE
Out :->

Now for my example which reproduces the problem:

------------------------------------------------
MODULE test;

IMPORT Out;

TYPE
  Item  = POINTER TO ItemD;
  ItemD = RECORD next: Item END;
  Header= RECORD first, last: Item END;

PROCEDURE Foo;
VAR list: Header;
BEGIN
  IF list.first=NIL THEN Out.String("list.first is NIL")
  ELSE Out.String("list.first is NOT NIL")
  END;
  Out.Ln;
END Foo;

BEGIN
  Foo;
END test.
-----------------------------------------------------------

The output is 'list.first is NOT NIL'.

Thanks for the hints.
-- 
Frank Hrebabetzky		+55 / 48 / 3235 1106
Florianopolis, Brazil

2000	Jan	Feb	Mar	Apr	May	Jun (1)	Jul (34)	Aug (19)	Sep (33)	Oct (14)	Nov (4)	Dec (4)
2001	Jan (6)	Feb (11)	Mar (8)	Apr (1)	May (24)	Jun (12)	Jul (13)	Aug (16)	Sep (8)	Oct (6)	Nov	Dec (5)
2002	Jan	Feb (14)	Mar	Apr	May (1)	Jun	Jul (3)	Aug (8)	Sep	Oct (3)	Nov	Dec (6)
2003	Jan (6)	Feb (4)	Mar (1)	Apr (1)	May (11)	Jun	Jul (1)	Aug (1)	Sep (3)	Oct (12)	Nov (22)	Dec (3)
2004	Jan (11)	Feb (16)	Mar (8)	Apr	May (35)	Jun (3)	Jul (14)	Aug (3)	Sep (7)	Oct (4)	Nov (30)	Dec (3)
2005	Jan (7)	Feb (16)	Mar (2)	Apr	May (10)	Jun (2)	Jul (4)	Aug (5)	Sep (4)	Oct (11)	Nov (1)	Dec (14)
2006	Jan (15)	Feb (6)	Mar (3)	Apr	May (1)	Jun (7)	Jul (11)	Aug	Sep	Oct	Nov	Dec
2007	Jan	Feb (5)	Mar (6)	Apr	May	Jun (11)	Jul (2)	Aug	Sep (9)	Oct (4)	Nov (2)	Dec
2008	Jan (5)	Feb (4)	Mar (5)	Apr	May (11)	Jun (1)	Jul	Aug	Sep	Oct	Nov (4)	Dec (7)
2009	Jan (8)	Feb	Mar (1)	Apr	May	Jun	Jul	Aug (3)	Sep (6)	Oct (6)	Nov	Dec
2010	Jan	Feb	Mar (1)	Apr	May (1)	Jun (2)	Jul (28)	Aug (18)	Sep	Oct (9)	Nov	Dec
2011	Jan	Feb (2)	Mar	Apr	May	Jun (2)	Jul (16)	Aug (18)	Sep (1)	Oct	Nov (1)	Dec (1)
2012	Jan	Feb	Mar	Apr	May	Jun (2)	Jul	Aug	Sep (1)	Oct	Nov	Dec
2015	Jan	Feb	Mar	Apr	May	Jun (4)	Jul	Aug	Sep	Oct	Nov	Dec (2)
2016	Jan (4)	Feb (1)	Mar (3)	Apr (1)	May (1)	Jun	Jul (2)	Aug (4)	Sep	Oct	Nov	Dec