Re: [Oorexx-devel] RexxMutableBuffer, RexxBuffer, RexxString : some questions

[Rick McGuire]

On Mon, Aug 16, 2010 at 7:10 PM, Jean-Louis Faucher
 wrote:
> Thanks Rick
>
> That's more clear now... one more question :-)
> When creating a RexxString instance using new_string, there is no behaviour
> associated to the instance, right ? If I'm not wrong, only newRexx attach a
> behaviour.
> So this instance created by new_string can be used internally, by calling
> directly its C++ methods, but it can't be used with sendMessages.
> Is that correct ?

No, not correct.  This gets created with the default string class
behaviour and is a fully functional string object.  new_string() is
used in situations where a real string instance is always created
(i.e., there's no possibility of a subclass being used).  Most places
in the interpreter that is true.

Rick

>
> Can such an instance without behaviour be returned to the interpreter ?
> It seems that the instances of RexxBuffer are in this case... I can't test,
> but I suppose that none of the methods inherited from Object can be used ?
>
> Jean-Louis
>
> --
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> ___
> Oorexx-devel mailing list
> Oorexx-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/oorexx-devel
>
>

--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Re: [Oorexx-devel] RexxMutableBuffer, RexxBuffer, RexxString : some questions

[Jean-Louis Faucher]

Thanks Rick

That's more clear now... one more question :-)
When creating a RexxString instance using new_string, there is no behaviour
associated to the instance, right ? If I'm not wrong, only newRexx attach a
behaviour.
So this instance created by new_string can be used internally, by calling
directly its C++ methods, but it can't be used with sendMessages.
Is that correct ?

Can such an instance without behaviour be returned to the interpreter ?
It seems that the instances of RexxBuffer are in this case... I can't test,
but I suppose that none of the methods inherited from Object can be used ?

Jean-Louis
--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev ___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Re: [Oorexx-devel] RexxMutableBuffer, RexxBuffer, RexxString : some questions

[Rick McGuire]

On Mon, Aug 16, 2010 at 5:48 PM, Jean-Louis Faucher
 wrote:
> Hi,
>
> I saw those lines in RexxArray::toString :
> RexxMutableBuffer *mutbuffer = ((RexxMutableBufferClass*)
> TheMutableBufferClass)->newRexx(NULL, 0);
> ProtectedObject p1(mutbuffer);
> mutbuffer->append(_stringValue);
> newString = mutbuffer->makeString();
>
> Q1 :
> I don't understand the implementation of RexxMutableBufferClass::newRexx.
> There is a new with placement, but why allocate at the memory location of
> 'this' ? According to the lines above, 'this' is TheMutableBufferClass.
>     /* allocate the new object   */
>     newBuffer = new ((RexxClass *)this) RexxMutableBuffer(bufferLength,
> defaultSize);
>     newBuffer->dataLength = string->getLength();
>     /* copy the content  */
>     newBuffer->data->copyData(0, string->getStringData(),
> string->getLength());

This is sort of based on old code before C++ really called these
"placements".  This call maps to the following new operator:

void *RexxMutableBuffer::operator new(size_t size, RexxClass *bufferClass)
{
RexxObject * newObj = new_object(size, T_MutableBuffer);
newObj->setBehaviour(bufferClass->getInstanceBehaviour());
return newObj;
}

This uses the instance behavior from the class object to construct the
object instance.  This allows for a mutablebuffer object to be
constructed using a class object that might be a subclass of the
mutable object class.  The classes that get exported as Rexx language
classes should have a new operator that allows the source class to be
provided.

>
> Q2 :
> Why do we have bufferLength and dataLength in RexxMutableBuffer ? Those
> attributes are already available in the encapsulated RexxBuffer :
> data->bufferSize and data->dataLength

Completeness and style.  To me, using data->bufferSize directly breaks
the object encapsulation.  The MutableBuffer class should define its
base operations without exposing the encapsulated data object at all.


>
> Q3 :
> RexxBuffer and RexxString have some helpers to create instances from C++ :
> new_buffer and new_string
> But nothing similar for RexxMutableBuffer. Any reason ?

Part history, part "it's not used often enough to need one".
Originally, this code was written in C with a lot of macros to emulate
a C++ type of behavior.  When this was converted to C++ (in just two
weeks!), many of the original macros were just mapped to the C++
equivalents so we didn't have to go and change everything.  For many
new classes, the number of places where the native code instantiates
the new objects does not justify adding the convenience methods.


>
> Q4 :
> In programming guide, I see that NewBufferString lets create a
> RexxBufferStringObject which is a mutable String object. It's implemented by
> a RexxString, not by a RexxMutableBuffer. Any reason ? Is it because
> RexxMutableBuffer would need ProtectedObject which is not available from
> API, I think ?

Yes, this API is specifically designed for situations where it is
desired to construct a String object value in place without needing to
buffer the object.  For example, the charin() method uses this to
allocate a string object before performing the stream I/O, which
allows the data to be read directly into the target location.  Once
the data has been read, the String object can be chopped to the final
result size.

Rick
>
> Jean-Louis
>
>
>
>
> --
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> ___
> Oorexx-devel mailing list
> Oorexx-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/oorexx-devel
>
>

--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

[Oorexx-devel] RexxMutableBuffer, RexxBuffer, RexxString : some questions

[Jean-Louis Faucher]

Hi,

I saw those lines in RexxArray::toString :
RexxMutableBuffer *mutbuffer = ((RexxMutableBufferClass*)
TheMutableBufferClass)->newRexx(NULL, 0);
ProtectedObject p1(mutbuffer);
mutbuffer->append(_stringValue);
newString = mutbuffer->makeString();

Q1 :
I don't understand the implementation of RexxMutableBufferClass::newRexx.
There is a new with placement, but why allocate at the memory location of
'this' ? According to the lines above, 'this' is TheMutableBufferClass.
/* allocate the new object   */
newBuffer = new ((RexxClass *)this) RexxMutableBuffer(bufferLength,
defaultSize);
newBuffer->dataLength = string->getLength();
/* copy the content  */
newBuffer->data->copyData(0, string->getStringData(),
string->getLength());

Q2 :
Why do we have bufferLength and dataLength in RexxMutableBuffer ? Those
attributes are already available in the encapsulated RexxBuffer :
data->bufferSize and data->dataLength

Q3 :
RexxBuffer and RexxString have some helpers to create instances from C++ :
new_buffer and new_string
But nothing similar for RexxMutableBuffer. Any reason ?

Q4 :
In programming guide, I see that NewBufferString lets create a
RexxBufferStringObject which is a mutable String object. It's implemented by
a RexxString, not by a RexxMutableBuffer. Any reason ? Is it because
RexxMutableBuffer would need ProtectedObject which is not available from
API, I think ?

Jean-Louis
--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev ___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Re: [Oorexx-devel] Request for help on (Ubuntu) Linux (32- and64-bit)

[Rony G. Flatscher]

On 16.08.2010 20:38, David Ashley wrote:
>   Right now the build machine is in a sort of transition state. All the 
> nightly 
> builds are running (with the exception of Win XP) but I need to install a new 
> network card in the new machine before I can start giving access to VMs. I 
> should have the card this week sometime.
>
> What version of Ubuntu do you want to test on? I can install 10.04 easily and 
> I 
> have some older versions as well. Let me know and I will istall it and let 
> you 
> know how to access it.
>   
32-bit 10.4 would be just fine (that's actually the version I have been
working with since yesterday).
Many thanks for all of the efforts in advance!

Maybe Rick found the possible cause already, which would be just awesome!

Again, many thanks to everybody helping!

---rony


> On 08/16/2010 01:19 PM, Rony G. Flatscher wrote:
>   
>> On 16.08.2010 19:27, Mike Cowlishaw wrote:
>> 
 I might take a look at this, this weekend, if I have time.
 Unfortunately, the set up required is a deterrring factor.

 
>>> This would seem to be an excellent example of the use of a virtual machine
>>> ... with a standard virtual machine with Ubuntu on it, Rony could do the
>>> minimum setup (added packages) to recreate the problem, then hand it over
>>> to the expert debugger-person.
>>>
>>>   
>> Yes, that would be really great!
>>
>> Maybe the build-machine allows for that already?
>>
>> ---rony
>> 


--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

[Oorexx-devel] Additional info (Re: Request for help on (Ubuntu) Linux (32- and 64-bit)

[Rony G. Flatscher]

O.K., here is one more piece of information which may be helpful.

Running one of those Rexx programs that cause that error this way from
rexxtry.rex fails:

call testOOo.rxo

whereas the following works:

cmd="rexx testOOo.rxo"
say cmd
cmd


Here are additional variations of invoking "testOOo.rxo":

* invoking from the commandline causes the reported error:

  rexx testOOo.rxo

* creating a Rexx program "test1.rex" containing the following code,
  causing the error:

call "testOOo.rxo"

* creating a Rexx program "test2.rex" containing the following code,
  which runs successfully (!):

cmd="rexx testOOo.rxo"
say cmd/* displays the string above */
cmd

HTH,

---rony


On 16.08.2010 20:17, Rony G. Flatscher wrote:
> Hi Mark,
>
> On 16.08.2010 19:02, Mark Miesfeld wrote:
>   
>> I know how frustrating these types of things can be.  (But, it is very
>> satisfying if you solve them.)
>>   
>> 
> Indeed, at the moment it is "superfrustrating", probably because I am a
> little bit "outworn" working for weeks now on overhauling the
> installation procedures for BSF4ooRexx on all platforms, where I opened
> far too many cans of worms, than I had anticipated!
> (But then, one should gain full desktop integration including templates
> and the like on both, Windows and Linux, and including ooRexx files
> themselves as well.)
>
>   
>> I might take a look at this, this weekend, if I have time.
>> Unfortunately, the set up required is a deterrring factor.
>>   
>> 
> Thank you very much!
>
>   
>> Until then:
>>
>> I would try with a build from trunk.  David has the build machine
>> working well and you could easily pick up a pre-built deb package with
>> all the latest fixes.
>>   
>> 
> Just did that to no avail. The problem with the requires is still there.
> (This is rev. 6111.)
> :(
>
> One more piece of information: at setup and installation, where a shell
> script starts a Rexx program, which then will run other shell scripts
> and Rexx programs, will along the lines start a Rexx program (via
> address) that requires UNO.CLS and succeeds! (I have just not been able
> to find a pattern to report.) "Running other Rexx programs" in this
> context means that they are invoked via the environment, e.g. with a
> statement like: "rexx someRexxprogram.rex" on its own line (so no
> procedure/function call using the Rexx file name as the target).
>
>   
>> I would try putting all the required files in the same diretory you
>> are executing your script from, just to see what happens.
>>   
>> 
> That I did already, and unfortunately, it did not change anything.
> :(
>
> Regards,
>
> ---rony
>
>   
>> On Mon, Aug 16, 2010 at 9:31 AM, Rony G. Flatscher
>>  wrote:
>>   
>> 
>>> Hi there,
>>>
>>> after experimenting the whole day, I have not yet come to a conclusion about
>>> the cause of a runtime problem I have been experiencing with ooRexx 4.0.1 on
>>> 32- and 64-bit Ubuntu:
>>>
>>> when requiring "UNO.CLS" which requires other Rexx packages itself, an error
>>> along the following lines occurs:
>>>
>>> r...@ronylinux:/opt/BSF4ooRexx/install$ rexx /usr/bin/UNO.CLS
>>>   318 *-* ::requires UNO_XINTERFACES.RXO-- get the list of all of UNO
>>> XInterface classes
>>> EX0003E: Error 3 running /usr/bin/UNO.CLS line 318:  Failure during
>>> initialization
>>> EX0200E: Error 3.1:  Failure during initialization: File
>>> "/usr/bin/UNO_XINTERFACES.RXO" is unreadable
>>> r...@ronylinux:/opt/BSF4ooRexx/install$
>>>
>>>
>>> it seems to be positional, i.e. if changing the requires statement to
>>> "::requires "rgf_utli2.rex" then the same error occurs but now relating to
>>> "rgf_util2.rex" !
>>> this error occurs whenever Rexx programs requiring UNO.CLS are invoked from
>>> the command line using the "rexx" executable. Running programs requiring
>>> UNO.CLS via the OpenOffice macro menu works!
>>> all rights on the scripts are set to 755 (rwxr-xr-x), they reside in
>>> "/usr/bin" (as a matter of fact, it does not matter where they reside and
>>> whether they are owned by root or by the user who executes the scripts).
>>>
>>> I have not been able to create a standalone test-set which would exhibit
>>> this behaviour on Linux, unfortunately. :(
>>>
>>> Therefore a request for help/ideas to solve this puzzling problem!
>>>
>>> This would be needed to recreate the problem as I have been facing it:
>>>
>>> get Ubuntu,
>>> uninstall the Ubuntu-OOo via synaptic manager (they tamper with the genuine
>>> OpenOffice, unfortunately, harming the Java interface to OOo, which is
>>> exploited by BSF4ooRexx),
>>> get the matching genuine OOo package by pointing your browser (from within
>>> Linux) to: , this will place the
>>> genuine OOo installation package that matches your installation on top,
>>> download it, unpack and install it (to install change into the DEB directory
>>> and issue: ""
>

Re: [Oorexx-devel] Request for help on (Ubuntu) Linux (32- and64-bit)

[David Ashley]

  Right now the build machine is in a sort of transition state. All the nightly 
builds are running (with the exception of Win XP) but I need to install a new 
network card in the new machine before I can start giving access to VMs. I 
should have the card this week sometime.

What version of Ubuntu do you want to test on? I can install 10.04 easily and I 
have some older versions as well. Let me know and I will istall it and let you 
know how to access it.

David Ashley

On 08/16/2010 01:19 PM, Rony G. Flatscher wrote:
>
> On 16.08.2010 19:27, Mike Cowlishaw wrote:
>>> I might take a look at this, this weekend, if I have time.
>>> Unfortunately, the set up required is a deterrring factor.
>>>
>> This would seem to be an excellent example of the use of a virtual machine
>> ... with a standard virtual machine with Ubuntu on it, Rony could do the
>> minimum setup (added packages) to recreate the problem, then hand it over
>> to the expert debugger-person.
>>
> Yes, that would be really great!
>
> Maybe the build-machine allows for that already?
>
> ---rony
>
>
>
>
> --
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> ___
> Oorexx-devel mailing list
> Oorexx-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/oorexx-devel
>


--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Re: [Oorexx-devel] Request for help on (Ubuntu) Linux (32- and64-bit)

[Rick McGuire]

Hmmm, taking a quick look at the code and the places where the file
unreadable error can be triggered, I found the following lines in
FileSystem.cpp:


RexxBuffer *SystemInterpreter::readProgram(const char *file_name)
/***/
/* Function:  Read a program into a buffer */
/***/
{
FILE*handle; /* open file access handle   */
size_t   buffersize; /* size of read buffer   */
{
handle = fopen(file_name, "rb"); /* open as a binary file
   */
if (handle == NULL)
{ /* open error?   */
return OREF_NULL;  /* return nothing
 */
}

if (fileno(handle) == (FOPEN_MAX - 2))
{  /* open error?   */
return OREF_NULL;  /* return nothing
 */
}

fseek(handle, 0, SEEK_END);  /* seek to the file end
   */
buffersize = ftell(handle);  /* get the file size
   */
fseek(handle, 0, SEEK_SET);  /* seek back to the file
beginning   */
}
RexxBuffer *buffer = new_buffer(buffersize); /* get a buffer
object   */
ProtectedObject p(buffer);
{
UnsafeBlock releaser;

fread(buffer->getData(), 1, buffersize, handle);
fclose(handle);  /* close the file
   */
}
return buffer;   /* return the program buffer */
}


I don't understand the purpose of the "fileno(handle) == (FOPEN_MAX -
2)" test, but that certainly could be the cause of this error.  It's
consistent with the fact that this error occurs even if you change the
name of the file.  This test appears to be something pulled forward
from the 3.2.0 code, but I don't know of any reason why that should
have been there in the first place.

Given the number of files that the jvm generally has open, it is
certainly conceivable that we're hitting a window where that's true.

Rick

On Mon, Aug 16, 2010 at 2:19 PM, Rony G. Flatscher
 wrote:
>
>
> On 16.08.2010 19:27, Mike Cowlishaw wrote:
>>> I might take a look at this, this weekend, if I have time.
>>> Unfortunately, the set up required is a deterrring factor.
>>>
>> This would seem to be an excellent example of the use of a virtual machine
>> ... with a standard virtual machine with Ubuntu on it, Rony could do the
>> minimum setup (added packages) to recreate the problem, then hand it over
>> to the expert debugger-person.
>>
> Yes, that would be really great!
>
> Maybe the build-machine allows for that already?
>
> ---rony
>
>
>
>
> --
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> ___
> Oorexx-devel mailing list
> Oorexx-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/oorexx-devel
>

--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Re: [Oorexx-devel] Request for help on (Ubuntu) Linux (32- and64-bit)

[Rony G. Flatscher]

On 16.08.2010 19:27, Mike Cowlishaw wrote:
>> I might take a look at this, this weekend, if I have time.
>> Unfortunately, the set up required is a deterrring factor.
>> 
> This would seem to be an excellent example of the use of a virtual machine
> ... with a standard virtual machine with Ubuntu on it, Rony could do the
> minimum setup (added packages) to recreate the problem, then hand it over
> to the expert debugger-person.
>   
Yes, that would be really great!

Maybe the build-machine allows for that already?

---rony




--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Re: [Oorexx-devel] Request for help on (Ubuntu) Linux (32- and 64-bit)

[Rony G. Flatscher]

Hi Mark,

On 16.08.2010 19:02, Mark Miesfeld wrote:
> I know how frustrating these types of things can be.  (But, it is very
> satisfying if you solve them.)
>   
Indeed, at the moment it is "superfrustrating", probably because I am a
little bit "outworn" working for weeks now on overhauling the
installation procedures for BSF4ooRexx on all platforms, where I opened
far too many cans of worms, than I had anticipated!
(But then, one should gain full desktop integration including templates
and the like on both, Windows and Linux, and including ooRexx files
themselves as well.)

> I might take a look at this, this weekend, if I have time.
> Unfortunately, the set up required is a deterrring factor.
>   
Thank you very much!

> Until then:
>
> I would try with a build from trunk.  David has the build machine
> working well and you could easily pick up a pre-built deb package with
> all the latest fixes.
>   
Just did that to no avail. The problem with the requires is still there.
(This is rev. 6111.)
:(

One more piece of information: at setup and installation, where a shell
script starts a Rexx program, which then will run other shell scripts
and Rexx programs, will along the lines start a Rexx program (via
address) that requires UNO.CLS and succeeds! (I have just not been able
to find a pattern to report.) "Running other Rexx programs" in this
context means that they are invoked via the environment, e.g. with a
statement like: "rexx someRexxprogram.rex" on its own line (so no
procedure/function call using the Rexx file name as the target).

> I would try putting all the required files in the same diretory you
> are executing your script from, just to see what happens.
>   
That I did already, and unfortunately, it did not change anything.
:(

Regards,

---rony

> On Mon, Aug 16, 2010 at 9:31 AM, Rony G. Flatscher
>  wrote:
>   
>> Hi there,
>>
>> after experimenting the whole day, I have not yet come to a conclusion about
>> the cause of a runtime problem I have been experiencing with ooRexx 4.0.1 on
>> 32- and 64-bit Ubuntu:
>>
>> when requiring "UNO.CLS" which requires other Rexx packages itself, an error
>> along the following lines occurs:
>>
>> r...@ronylinux:/opt/BSF4ooRexx/install$ rexx /usr/bin/UNO.CLS
>>   318 *-* ::requires UNO_XINTERFACES.RXO-- get the list of all of UNO
>> XInterface classes
>> EX0003E: Error 3 running /usr/bin/UNO.CLS line 318:  Failure during
>> initialization
>> EX0200E: Error 3.1:  Failure during initialization: File
>> "/usr/bin/UNO_XINTERFACES.RXO" is unreadable
>> r...@ronylinux:/opt/BSF4ooRexx/install$
>>
>>
>> it seems to be positional, i.e. if changing the requires statement to
>> "::requires "rgf_utli2.rex" then the same error occurs but now relating to
>> "rgf_util2.rex" !
>> this error occurs whenever Rexx programs requiring UNO.CLS are invoked from
>> the command line using the "rexx" executable. Running programs requiring
>> UNO.CLS via the OpenOffice macro menu works!
>> all rights on the scripts are set to 755 (rwxr-xr-x), they reside in
>> "/usr/bin" (as a matter of fact, it does not matter where they reside and
>> whether they are owned by root or by the user who executes the scripts).
>>
>> I have not been able to create a standalone test-set which would exhibit
>> this behaviour on Linux, unfortunately. :(
>>
>> Therefore a request for help/ideas to solve this puzzling problem!
>>
>> This would be needed to recreate the problem as I have been facing it:
>>
>> get Ubuntu,
>> uninstall the Ubuntu-OOo via synaptic manager (they tamper with the genuine
>> OpenOffice, unfortunately, harming the Java interface to OOo, which is
>> exploited by BSF4ooRexx),
>> get the matching genuine OOo package by pointing your browser (from within
>> Linux) to: , this will place the
>> genuine OOo installation package that matches your installation on top,
>> download it, unpack and install it (to install change into the DEB directory
>> and issue: ""
>> get the installation package of BSF4ooRexx from this particular location:
>> 
>>
>> unzip it, change into "bsf4oorexx/install" and run "install.sh"; this will
>> copy the entire package to "/opt/BSF4ooRexx" and proceed from
>> "/opt/BSF4ooRexx/install/" where the installation and uninstall scripts get
>> created
>> after the installation BSF4ooRexx got installed, and now you need to install
>> the OOo support by running "/opt/BSF4ooRexx/install/reinstall.sh"
>>
>> currently it is possible that "unopkg" is not successfully carried out, due
>> to a leftover lock file in "~/.openoffice/3/.lock"
>> locate and delete that lock-file, then run "sudo
>> /opt/BSF4ooRexx/install/installOOo.sh"
>> check by starting the OOo writer, go to "Tools -> Macros -> Run" and pick
>> the "HelloWorld" example from the "OpenOffice/ooRexxExamples" macro folder.
>>
>> best logoff and logon again (this makes sure tha

Re: [Oorexx-devel] SLES 11 Tests

[Erico Mendonca]

>>> On 8/11/2010 at 04:12 PM, in message 
>>> <4c62a1a6022400055...@novprvlin0050.provo.novell.com>, "Erico Mendonca" 
>>>  wrote:



BUT, I'm getting error on TEST_006 on SLES 11SP1 if I try to use the same build 
as SLES 11:
 

Okay, apparently the test was failing because the networking on that specific 
VM wasn't setup correctly. I ran the tests 10 times without an error on SLES 11 
SP1. So far it's been tested on openSUSE 10.2, SLES 11 and SLES 11sp1 with 
success. 

I'll open the ticket and attach my proposed SPEC. 


--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev ___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Re: [Oorexx-devel] Request for help on (Ubuntu) Linux (32- and64-bit)

[Mike Cowlishaw]

 
> I might take a look at this, this weekend, if I have time.
> Unfortunately, the set up required is a deterrring factor.

This would seem to be an excellent example of the use of a virtual machine
... with a standard virtual machine with Ubuntu on it, Rony could do the
minimum setup (added packages) to recreate the problem, then hand it over
to the expert debugger-person.


--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Re: [Oorexx-devel] Request for help on (Ubuntu) Linux (32- and 64-bit)

[Mark Miesfeld]

Hi Rony,

I know how frustrating these types of things can be.  (But, it is very
satisfying if you solve them.)

I might take a look at this, this weekend, if I have time.
Unfortunately, the set up required is a deterrring factor.

Until then:

I would try with a build from trunk.  David has the build machine
working well and you could easily pick up a pre-built deb package with
all the latest fixes.

I would try putting all the required files in the same diretory you
are executing your script from, just to see what happens.

Of course, you may have already tried these things.

--
Mark Miesfeld

On Mon, Aug 16, 2010 at 9:31 AM, Rony G. Flatscher
 wrote:
> Hi there,
>
> after experimenting the whole day, I have not yet come to a conclusion about
> the cause of a runtime problem I have been experiencing with ooRexx 4.0.1 on
> 32- and 64-bit Ubuntu:
>
> when requiring "UNO.CLS" which requires other Rexx packages itself, an error
> along the following lines occurs:
>
> r...@ronylinux:/opt/BSF4ooRexx/install$ rexx /usr/bin/UNO.CLS
>   318 *-* ::requires UNO_XINTERFACES.RXO-- get the list of all of UNO
> XInterface classes
> EX0003E: Error 3 running /usr/bin/UNO.CLS line 318:  Failure during
> initialization
> EX0200E: Error 3.1:  Failure during initialization: File
> "/usr/bin/UNO_XINTERFACES.RXO" is unreadable
> r...@ronylinux:/opt/BSF4ooRexx/install$
>
>
> it seems to be positional, i.e. if changing the requires statement to
> "::requires "rgf_utli2.rex" then the same error occurs but now relating to
> "rgf_util2.rex" !
> this error occurs whenever Rexx programs requiring UNO.CLS are invoked from
> the command line using the "rexx" executable. Running programs requiring
> UNO.CLS via the OpenOffice macro menu works!
> all rights on the scripts are set to 755 (rwxr-xr-x), they reside in
> "/usr/bin" (as a matter of fact, it does not matter where they reside and
> whether they are owned by root or by the user who executes the scripts).
>
> I have not been able to create a standalone test-set which would exhibit
> this behaviour on Linux, unfortunately. :(
>
> Therefore a request for help/ideas to solve this puzzling problem!
>
> This would be needed to recreate the problem as I have been facing it:
>
> get Ubuntu,
> uninstall the Ubuntu-OOo via synaptic manager (they tamper with the genuine
> OpenOffice, unfortunately, harming the Java interface to OOo, which is
> exploited by BSF4ooRexx),
> get the matching genuine OOo package by pointing your browser (from within
> Linux) to: , this will place the
> genuine OOo installation package that matches your installation on top,
> download it, unpack and install it (to install change into the DEB directory
> and issue: ""
> get the installation package of BSF4ooRexx from this particular location:
> 
>
> unzip it, change into "bsf4oorexx/install" and run "install.sh"; this will
> copy the entire package to "/opt/BSF4ooRexx" and proceed from
> "/opt/BSF4ooRexx/install/" where the installation and uninstall scripts get
> created
> after the installation BSF4ooRexx got installed, and now you need to install
> the OOo support by running "/opt/BSF4ooRexx/install/reinstall.sh"
>
> currently it is possible that "unopkg" is not successfully carried out, due
> to a leftover lock file in "~/.openoffice/3/.lock"
> locate and delete that lock-file, then run "sudo
> /opt/BSF4ooRexx/install/installOOo.sh"
> check by starting the OOo writer, go to "Tools -> Macros -> Run" and pick
> the "HelloWorld" example from the "OpenOffice/ooRexxExamples" macro folder.
>
> best logoff and logon again (this makes sure that from that point on
> "/etc/profile" is honored which sets up the environment for BSF4ooRexx; this
> allows one to start BSF4ooRexx scripts also by double-clicking via the
> desktop interface of Linux).
>
> Now, if you open a command line window and go into "/opt/BSF4ooRexx/install"
> and run "rexx testOOo.rxo" you should get the above mentioned error!
>
> Hopefully, you can find a workaround or the cause of this error, which is a
> showstopper right now for the Linux platform.
>
> TIA,
>
> ---rony
>
>
>
>
> --
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> ___
> Oorexx-devel mailing list
> Oorexx-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/oorexx-devel
>
>

--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev 
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.ne

Re: [Oorexx-devel] Request for help on (Ubuntu) Linux (32- and 64-bit)

[Rony G. Flatscher]

... cut ...

> This would be needed to recreate the problem as I have been facing it:
>
> * get Ubuntu,
> * uninstall the Ubuntu-OOo via synaptic manager (they tamper with
>   the genuine OpenOffice, unfortunately, harming the Java
>   interface to OOo, which is exploited by BSF4ooRexx),
> * get the matching genuine OOo package by pointing your browser
>   (from within Linux) to:
>   , this will place the
>   genuine OOo installation package that matches your installation
>   on top, download it, unpack and install it (to install change
>   into the DEB directory and issue: ""
>
... cut ...

Sorry, overlooked the command to easily install the genuine OOo package,
by changing into the DEB directory and issuing a:

dpkg -i *.deb

---rony
--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev ___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

[Oorexx-devel] Request for help on (Ubuntu) Linux (32- and 64-bit)

[Rony G. Flatscher]

Hi there,

after experimenting the whole day, I have not yet come to a conclusion
about the cause of a runtime problem I have been experiencing with
ooRexx 4.0.1 on 32- and 64-bit Ubuntu:

* when requiring "UNO.CLS" which requires other Rexx packages
  itself, an error along the following lines occurs:
  o

r...@ronylinux:/opt/BSF4ooRexx/install$ rexx /usr/bin/UNO.CLS
  318 *-* ::requires UNO_XINTERFACES.RXO-- get the list of all 
of UNO XInterface classes
EX0003E: Error 3 running /usr/bin/UNO.CLS line 318:  Failure during 
initialization
EX0200E: Error 3.1:  Failure during initialization: File 
"/usr/bin/UNO_XINTERFACES.RXO" is unreadable
r...@ronylinux:/opt/BSF4ooRexx/install$ 
  

* it seems to be positional, i.e. if changing the requires statement
  to "::requires "rgf_utli2.rex" then the same error occurs but now
  relating to "rgf_util2.rex" !
* this error occurs whenever Rexx programs requiring UNO.CLS are
  invoked from the command line using the "rexx" executable. Running
  programs requiring UNO.CLS via the OpenOffice macro menu works!
* all rights on the scripts are set to 755 (rwxr-xr-x), they reside
  in "/usr/bin" (as a matter of fact, it does not matter where they
  reside and whether they are owned by root or by the user who
  executes the scripts).

I have not been able to create a standalone test-set which would exhibit
this behaviour on Linux, unfortunately. :(

Therefore a request for help/ideas to solve this puzzling problem!

This would be needed to recreate the problem as I have been facing it:

* get Ubuntu,
* uninstall the Ubuntu-OOo via synaptic manager (they tamper with
  the genuine OpenOffice, unfortunately, harming the Java interface
  to OOo, which is exploited by BSF4ooRexx),
* get the matching genuine OOo package by pointing your browser
  (from within Linux) to:
  , this will place the
  genuine OOo installation package that matches your installation on
  top, download it, unpack and install it (to install change into
  the DEB directory and issue: ""
* get the installation package of BSF4ooRexx from this particular
  location:
  

  o unzip it, change into "bsf4oorexx/install" and run
"install.sh"; this will copy the entire package to
"/opt/BSF4ooRexx" and proceed from
"/opt/BSF4ooRexx/install/" where the installation and
uninstall scripts get created
  o after the installation BSF4ooRexx got installed, and now you
need to install the OOo support by running
"/opt/BSF4ooRexx/install/reinstall.sh"
+ currently it is possible that "unopkg" is not
  successfully carried out, due to a leftover lock file
  in "~/.openoffice/3/.lock"
+ locate and delete that lock-file, then run "sudo
  /opt/BSF4ooRexx/install/installOOo.sh"
+ check by starting the OOo writer, go to "Tools ->
  Macros -> Run" and pick the "HelloWorld" example from
  the "OpenOffice/ooRexxExamples" macro folder.
* best logoff and logon again (this makes sure that from that point
  on "/etc/profile" is honored which sets up the environment for
  BSF4ooRexx; this allows one to start BSF4ooRexx scripts also by
  double-clicking via the desktop interface of Linux).

Now, if you open a command line window and go into
"/opt/BSF4ooRexx/install" and run "rexx testOOo.rxo" you should get the
above mentioned error!

Hopefully, you can find a workaround or the cause of this error, which
is a showstopper right now for the Linux platform.

TIA,

---rony



--
This SF.net email is sponsored by 

Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev ___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oorexx-devel