Some information that might help resolve the PTHREAD_* issues.
Rick
-- Forwarded message --
From: Robert Maynard robert.mayn...@kitware.com
Date: Tue, Sep 23, 2014 at 11:31 AM
Subject: Re: [CMake] Problems resolving PTHREAD_* symbols.
To: Rick McGuire object.r...@gmail.com
Cc
I've fixed the other problems and David committed a fix that might fix the
Mac build also that is other than the change to REQUIRED_LIBRARIES. I've
been able to test that fix on one of the platforms we need to build on and
it seemed ok, but I think David will need to check this out using the
I've started coding this up, and the , syntax for list creation looks
very doable. The meaning of the , depends on the context, as already
noted, but the contexts are absolutely unambiguous so it is always possible
to assign the correct meaning.
However, I want to circle back around to the
collections.
just my 2p worth
Jon
On 18 September 2014 14:24, Rick McGuire object.r...@gmail.com wrote:
I confess that I've been wanting a shortcut for creating arrays for a
long time and have even considered submitting this feature request myself.
This is not something difficult to implement
On Wed, Sep 24, 2014 at 11:41 AM, Mike Cowlishaw m...@speleotrove.com
wrote:
However, I want to circle back around to the discussion of the
short-circuited AND. We can't eliminate the usage on IF/WHEN, etc., but on
the other hand, there are certainly contexts where an and operator that
-both XOR
OTOH, if the ooRexx '' and '|' operators are already defined to
short-circuit, how about:
= short-circuit AND
+ = eval-both AND
| = short-circuit OR
|+ = eval-both OR
= eval-both
?
-Chip-
On 9/24/2014 10:17 AM, Rick McGuire wrote:
I've started coding this up
On Wed, Sep 24, 2014 at 12:14 PM, Mike Cowlishaw m...@speleotrove.com
wrote:
However, I want to circle back around to the discussion of the
short-circuited AND. We can't eliminate the usage on IF/WHEN, etc., but on
the other hand, there are certainly contexts where an and operator that
On Wed, Sep 24, 2014 at 1:06 PM, Gil Barmwater gbarmwa...@alum.rpi.edu
wrote:
Good point! I'm leaning toward - and -| as they kind of imply less
than a full or |.
So was I, then I hit a small snag on how to implement these, since they
would really fall outside of operators are handled in
problem rather
than a hard to implement problem. Sometimes things that are easy to do
are still wrong because of other implications.
Rick
On 9/24/2014 1:17 PM, Rick McGuire wrote:
On Wed, Sep 24, 2014 at 1:06 PM, Gil Barmwater gbarmwa...@alum.rpi.edu
wrote:
Good point! I'm leaning toward
I have an initial stab at implementing this now. I decide on ::RESOURCE for
the directive name. Here's a short sample of how it works now:
loop name over .resources~names
say name
end
loop name over .resources~namesToo
say name
end
::resource names
Rick
Mike
::END
::resource namesToo end
On Wed, Sep 24, 2014 at 4:39 PM, Michael Lueck mlu...@lueckdatasystems.com
wrote:
David Ashley wrote:
At this point it really looks like we will need to start with a clean
slate as far as platforms we support with the ooRexx 5.0 version.
I agree.
Our last client we were using Mod_Rexx
included in the array.
Rick
thanks,
Jon
On 24 September 2014 21:24, Rick McGuire object.r...@gmail.com wrote:
I have an initial stab at implementing this now. I decide on ::RESOURCE
for the directive name. Here's a short sample of how it works now:
loop name over .resources~names
say
What version of the compiler are you using? I had to add a check to
PlatformDefinitions.h to check for the MSC version. We used to have to use
map isnan to _isnan, but when I upgraded to VSC 13 it was necessary to add
a version check and disable the define. See line 133 in
w.david.ash...@gmail.com
wrote:
I use Visual Studio v9.0. The compiler is v15.00.30729.01.
On Thu, 2014-09-25 at 20:42 -0400, Rick McGuire wrote:
What version of the compiler are you using? I had to add a check to
PlatformDefinitions.h to check for the MSC version. We used to have
to use
The intent of this feature is to gather up all of the disparate pieces of
information that might effect the interpretation of a program and put them
in one place. This will be a combination of Rexx language information and
some platform-specific information. The object will be available in the
Decided to also add majorVersion, release, and revision values to this to
return the pieces of the version number.
Rick
On Sun, Sep 28, 2014 at 7:35 AM, Rick McGuire object.r...@gmail.com wrote:
The intent of this feature is to gather up all of the disparate pieces of
information that might
I take that back, I do not have an installed version. I just unpacked the
longstrings version from sourceforge into c:\NSIS and added it to my path.
I don't know why this isn't working for you.
Rick
On Sun, Sep 28, 2014 at 6:26 PM, Rick McGuire object.r...@gmail.com wrote:
David,
I think
One of the features I've been working on is a concept of namespaces that
can be used to disambiguate information pulled in from different ::requires
files. The basis if this is a new NAMESPACE option on the ::requires
directive where you give a simple name to a loaded requires file.
::requires
would use
x = .array~new
On Wed, Oct 1, 2014 at 9:10 AM, David Ashley w.david.ash...@gmail.com
wrote:
I prefer the namespace:name version. It seems more hierarchical.
David Ashley
On Tue, 2014-09-30 at 18:52 -0400, Rick McGuire wrote:
One of the features I've been working on is a concept
you might include.
Rick
On Wed, Oct 1, 2014 at 9:18 AM, Rick McGuire object.r...@gmail.com wrote:
The deeper I get into the details of implementing this, the more I think
it needs to be namespace;name too. I'm also going to de-couple this from
dot variables because the dot is not really
is a bit strange environment that tries to emulate a unix system. It would
be that it has its own cd command that expects unix file name conventions.
This is just a guess. cygwin always makes me nervous because it tries to
live in both worlds and does neither well.
Rick
On Wed, Oct 1, 2014 at
I think this might be something to ask on the cmake list. I need a few
more specifics. When you run this in the foreground, how are you accessing
the system? Are you still going via ssh? What is your background access
mechanism?
Rick
On Wed, Oct 1, 2014 at 9:31 AM, David Ashley
This article might be useful. Perhaps this is a problem with how you have
sshd configured...it might not have permission to access that directory.
Rick
On Wed, Oct 1, 2014 at 9:31 AM, David Ashley w.david.ash...@gmail.com
wrote:
Rick -
This is a strange one. When I run my script(s) that
This really is the wrong thing to do. This change would essentially mean
that the behavior of any program would become dependent upon what code is
using it rather than the preferences of the programmer.
The 5.0.0 version is offering something for a major cause of these sorts of
errors. The
Rats, it appears resource compiler version you are using doesn't support
enum symbols in the .rc file. I there any chance you could upgrade the
compiler version you are are using? You're already 4 release versions
behind what I'm using to build and I suspect problems like this will become
more
Ok, this looks like it is easier to fix than I thought. I'll commit the
change once I get the stuff I'm working on to build cleanly with this
change.
Rick
On Thu, Oct 2, 2014 at 9:19 AM, Rick McGuire object.r...@gmail.com wrote:
Rats, it appears resource compiler version you are using doesn't
I'm using the full 2013 version right now.
Rick
On Thu, Oct 2, 2014 at 10:42 AM, David Ashley w.david.ash...@gmail.com
wrote:
Are you using the 2013 express version?
David Ashley
On Thu, 2014-10-02 at 09:19 -0400, Rick McGuire wrote:
Rats, it appears resource compiler version you
-10-02 at 09:19 -0400, Rick McGuire wrote:
Rats, it appears resource compiler version you are using doesn't
support enum symbols in the .rc file. I there any chance you could
upgrade the compiler version you are are using? You're already 4
release versions behind what I'm using to build
.
Is there a tutorial or readily available documentation on the care and
feeding of the test framework? Sometimes, getting over the first
learning curb is the biggest hurdle.
-Chip-
On 10/4/2014 5:35 PM, Rick McGuire wrote:
We have the build machine issues more or less worked out so that we're
:07 PM, Rick McGuire object.r...@gmail.com
wrote:
Thanks for reporting this, I'll take a look at this later today. This
slipped through because there are currently no tests in the test suite for
the supplier class (hint, hint)
Rick
On Sun, Oct 5, 2014 at 1:25 PM, Erich Steinböck
Jon,
For people developing test cases, I really recommend not executing this by
running the entire test suite. I don't want to have to deal with other
failures or problems they might turn up. Running the test suite is NOT
necessary.
For me, all I do is add the root of the test tree and the
Don't bother adding the directives directory or any other directory you're
running tests from to your path. That shouldn't be necessary. You can run
any of the tests by either CDing to the directory or directly by specifying
the full directory name on the rexx command.
The directory you DO need
Definitely not a bug. Error 40 is Invalid call to routine and variants,
so having the [] method return that error would be a bug. The error
returned looks appropriate for this condition...it is an invalid position
argument for a message.
I'd like to keep development issues separate from the
The only error here is how this particular test case is just obscuring what
is going on. These two are NOT equivalent because of differences in
precedence involved. The [] only gets applied to the last term of the
expression, so the actual method call is just
699[6.99 - 6.00, 5*2]
or after the
...@chello.at
wrote:
Well, I took the substr testgroup as a base and translate it step by step
– learning things along this way
Sorry for bothering you
Here I learned about precedence…
*Von:* Rick McGuire [mailto:object.r...@gmail.com]
*Gesendet:* Montag, 06. Oktober 2014 19:13
*An:* Open Object
Ummm, what about the last character on the line. Looks like you are 0 for
3.
Rick
On Mon, Oct 6, 2014 at 2:58 PM, Walter Pachl christel.u.w.pa...@chello.at
wrote:
I hope that this is a valid one:
1 *-* Say 'AB BC CD DE EF'['1234567890'[1,2],22])
Error. 37 running.
in case test081 (commented out for the time being)
And test087 where I see a matching pair of parens
Walter ( still hoping for 1 of 3 )
PS That’s what you get by hiring unskilled testers J
*Von:* Rick McGuire [mailto:object.r...@gmail.com]
*Gesendet:* Montag, 06. Oktober 2014 21:03
That’s what you get by hiring unskilled testers J
*Von:* Rick McGuire [mailto:object.r...@gmail.com]
*Gesendet:* Montag, 06. Oktober 2014 21:03
*An:* Open Object Rexx Developer Mailing List
*Betreff:* Re: [Oorexx-devel] W-bug # 3
Ummm, what about the last character on the line. Looks
the mutablebuffer from scratch
And then one which you delegate (and where you think I might get a 1 after
all)
Over for today and thanks for all patience
Walter
*Von:* Rick McGuire [mailto:object.r...@gmail.com]
*Gesendet:* Montag, 06. Oktober 2014 21:37
*An:* Open Object Rexx
David,
The 5.0.0 interpreter is still compatible with the 4.2.0 daemon because
none of the structures used to pass data back and forth have changed. I'm
switching back and forth between 5.0.0 and 4.2.0 constantly without needing
to kill the rxapi process. Given that you needn't worry about
Not currently planned for the release.
Rick
On Tue, Oct 7, 2014 at 10:15 AM, r...@ntlworld.com wrote:
Will the new version include GTK and/or ncurses?
--
Meet PCI DSS 3.0 Compliance Requirements with EventLog
On Wed, Oct 8, 2014 at 7:27 AM, Walter Pachl christel.u.w.pa...@chello.at
wrote:
I have 4 testgroups for mutablebuffer
cntains
caselessContains
containsWord
caselessContainsWord
should they go with the same names into the class/mutablebuffer as tey are
in class/string?
yes
Do I
On Wed, Oct 8, 2014 at 6:12 PM, Walter Pachl christel.u.w.pa...@chello.at
wrote:
There is a testgroup
D:\_ooTest\trunk\ooRexx\base\bif\ERRORTEXT.testGroup
I had my fingers at THIS as early as 2005 !!! (didn't remember but found
my name in it :-)
When trying it now it gets many errors
)_64-bit 6.05 9 Oct
2014)
On Sun, Sep 28, 2014 at 3:18 PM, Rick McGuire object.r...@gmail.com
wrote:
Decided to also add majorVersion, release, and revision values to this to
return the pieces of the version number.
Rick
On Sun, Sep 28, 2014 at 7:35 AM, Rick McGuire object.r...@gmail.com
That's the same message we've been using for a long time...made necessary
by the people who chose to ignore gentler versions.
Rick.
On Thu, Oct 9, 2014 at 3:56 AM, Erich Steinböck erich.steinbo...@gmail.com
wrote:
During various 5.0.0 installs I've come across the
Yes, it is. It actually should have always been that way once array had a
makestring method added. That was supposed to cause automatic coercion of
the object into a string when required. There were some issues with how
makestring worked that made this dangerous, so I left the bug in place
with supplier being a subkeyword. That makes it very clear what is
happening.
David Ashley
On Fri, 2014-10-10 at 17:39 -0400, Rick McGuire wrote:
Having just coded about 3 supplier loops today where I ended up in an
infinite loop because I forgot to code the next method call at the end
I've been trying to improve the test coverage of some of the classes and am
currently working on tests for Object. At the moment, I'm writing some
tests for the RUN() method.
The RUN method is a private method that allows a method object to be run
dynamically on an object instance. Since this
occurred. I really, really don't expect this is going to ever be a
problem, but the behavior still needs to be well defined.
Rick
On Sat, Oct 11, 2014 at 4:14 PM, Rick McGuire object.r...@gmail.com wrote:
I've been trying to improve the test coverage of some of the classes and
am currently
Grrr, keep forgetting I'm not in SlickEdit when typing emails!
On Sun, Oct 12, 2014 at 5:48 AM, Rick McGuire object.r...@gmail.com wrote:
Because your loop is not valid. The assignment
d=r[c]
Is undetermined for relations. You are picking up the same (undetermined)
value every time
Not even going to look at something like this at this time.
Rick
On Sun, Oct 12, 2014 at 2:05 PM, Erich Steinböck erich.steinbo...@gmail.com
wrote:
rexx -e A=('word' .endofline)~copies(50)~makeArray()
*5.0.0:* 46.78 seconds elapsed time
*4.2.0:* 0.14 seconds elapsed time
That's one of the problems with DO OVER. The value assigned to the control
variable is not necessarily the value. The semantics here are to send a
makearray message to the expression object and then iterate over that. For
arrays, this results in iteration over a non-sparse version of the array,
with something that works on the latest
Windows and a handful of Linux versions, if there is an alternative.
best regards,
René.
On 13 okt. 2014, at 16:58, Rick McGuire object.r...@gmail.com wrote:
On Mon, Oct 13, 2014 at 10:45 AM, René Jansen rvjan...@xs4all.nl wrote:
David,
I
message --
From: Rick McGuire bigr...@users.sf.net
Date: Mon, Oct 13, 2014 at 1:09 PM
Subject: [oorexx:feature-requests] #638 Allow access to the main part of
a package object.
To: Ticket 638 6...@feature-requests.oorexx.p.re.sf.net
---
** [feature-requests:#638] Allow access to the main
in a clean state so that it can be reused, while still returning an
object that can be used to track execution.
Rick
Jon
On 14 October 2014 15:01, Rick McGuire object.r...@gmail.com wrote:
I've just started pounding on the .Message class and decided to revisit
this one. Looking at the code
On Tue, Oct 14, 2014 at 10:36 AM, Mike Cowlishaw m...@speleotrove.com
wrote:
I'd like to relax the reuse restriction a bit so that reuse is only
forbidden once the message object has been used asynchronously. Send() can
be called over and over. Start() can be called only once. However,
If the code is truly labeled as public domain, there should be no problem
with including it.
Rick
On Tue, Oct 14, 2014 at 3:50 PM, David Ashley w.david.ash...@gmail.com
wrote:
Rick, I guess this is mostly for you. I need advice.
PDCurses is a public domain version of the curses library (and
That is something that gets set by a debug build, which is the style the
build machine is creating. The release versions will not have this.
Rick
On Wed, Oct 15, 2014 at 6:12 AM, Sahananda (Jon) Wolfers
sahana...@windhorse.biz wrote:
Hi,
when testing the 5.0.0 version from the build
How recent is your interpreter version? There was a similar issue with
SQLLoadFuncs that has already been fixed.
Rick
On Wed, Oct 15, 2014 at 7:49 AM, Erich Steinböck erich.steinbo...@gmail.com
wrote:
5.0.0. will return 0 (failure) for
say
recent is your interpreter version?
It's most recent: REXX-ooRexx_5.0.0(MT)_64-bit 6.05 15 Oct 2014
Erich
On Wed, Oct 15, 2014 at 1:57 PM, Rick McGuire object.r...@gmail.com
wrote:
How recent is your interpreter version? There was a similar issue with
SQLLoadFuncs that has already been fixed
I was just defining an interface for the alarm class and was adding an
method that will be called when an alarm is canceled. As I typed the name
alarmCanceled, I was uncertain of the correct spelling of Canceled. It
turns out canceled and cancelled are both valid spellings...sigh. For
now, I'm
That doesn't really work in this case, because these are methods that get
called by the alarm objects, not methods that that the user calls.
Rick
On Wed, Oct 15, 2014 at 11:17 AM, Mike Cowlishaw m...@speleotrove.com
wrote:
I was just defining an interface for the alarm class and was adding
--
*From:* Rick McGuire [mailto:object.r...@gmail.com]
*Sent:* 15 October 2014 16:43
*To:* Open Object Rexx Developer Mailing List
*Subject:* Re: [Oorexx-devel] Synonyms for canceled...
It is a case where people provide an object that implements a particular
Nothing is currently planned in 5.0.0 for making any of that information
available. That stuff is the purposes of the interpreter to determine if
the package can be loaded.
Rick
On Mon, Oct 20, 2014 at 5:40 AM, Rony G. Flatscher rony.flatsc...@wu.ac.at
wrote:
The RexxPackageEntry structure
The annotations must consist of name/value pairs, and that is the only
thing that is going to be provided.
Rick
On Mon, Oct 20, 2014 at 5:45 AM, Rony G. Flatscher rony.flatsc...@wu.ac.at
wrote:
It is really great that annotations get added to the language!
It may be conceivable that there
Square brackets are just a method, not restricted to just collection
classes. Any class can choose to implement that method, so this problem is
no different than determining the effect of any method call.
Rick
On Mon, Oct 20, 2014 at 5:57 AM, Rony G. Flatscher rony.flatsc...@wu.ac.at
wrote:
Use SendMessage and send the appropriate operator message. For example,
result = context-SendMessage1(anObject, +, context-RexxTrue);
On Tue, Oct 21, 2014 at 4:30 PM, David Ashley w.david.ash...@gmail.com
wrote:
Rick -
If I am writing a C++ method for ooRexx and I want to perform an
grumble...gmail...grumble...not slickedit...grumble.
Anyway, that would add 1 to the object and return a new object.
Rick
On Tue, Oct 21, 2014 at 4:47 PM, Rick McGuire object.r...@gmail.com wrote:
Use SendMessage and send the appropriate operator message. For example,
result = context
Just did my first Linux build in a while, and I'm getting a compile error
because it can't find ncurses.h. What do I need to install to get this to
build?
Rick
--
___
Oops, forgot to do the svn add before committing.
Rick
On Thu, Oct 30, 2014 at 9:05 AM, David Ashley w.david.ash...@gmail.com
wrote:
Rick -
All the nightly builds failed withe the following error:
-- Looking for usersec.h
-- Looking for usersec.h - not found
-- Looking for ncurses.h
--
What phase are you getting this? During the initial build, or when trying
to build the installer?
Rick
On Fri, Oct 31, 2014 at 6:09 PM, David Ashley w.david.ash...@gmail.com
wrote:
Rick -
I am now getting the following Windows build error:
CMake Error at cmake_install.cmake:394 (file):
Just committed a fix for the one warning in Token.hpp and also a potential
fix for the compile error.
Rick
On Sat, Dec 6, 2014 at 3:35 PM, David Ashley w.david.ash...@gmail.com
wrote:
I got my used Mac Mini last Friday. It was pretty easy to set up. I
updated it to OSX 10.10.1 and then
The best way to make something like this available would be go generate an
SVN patch file and open a patch tracker item.
Rick
On Sat, Dec 6, 2014 at 5:28 PM, Enrico Sorichetti enricosoriche...@mac.com
wrote:
Hello,
I have had oorexx 5 built and working on Mavericks and Yosemite for a while
Lueck mlu...@lueckdatasystems.com
wrote:
Greetings Rick,
Thank you for engaging in my question.
Rick McGuire wrote:
The resulting interpreter would be a Windows-only thing, something I
have no interest in writing.
Well since Java is able to be managed, and Java is cross-platform
I'd really prefer that access be kept to a small core group at first. If
additional people start to get really involved, understand the rules for
what sort of changes are permitted, and appear to be trustworthy, they can
be easily added later.
I'm starting to think that maybe we should create a
Section 10.2.1, Power Operator.
On Tue, Jan 6, 2015 at 7:44 PM, Walter Pachl christel.u.w.pa...@chello.at
wrote:
Thank you very much!
Unfortunately I have TRL2 only (incompletely) in my head
I guess this cannot be found in the ooRexx Reference, can it?
Walter
*Von:* Rick McGuire
Ashley w.david.ash...@gmail.com wrote:
So why doesn't Jon's method throw an error? Are you not invoking a
method on a class that does not exist (the isa method)? Why does that not
throw an error?
David Ashley
On 03/11/2015 10:12 AM, Rick McGuire wrote:
Jon's method doesn't throw an error
On Wed, Mar 11, 2015 at 10:55 AM, David Ashley w.david.ash...@gmail.com
wrote:
That works for builtin classes, but not for user defined classes. I want
something that can query both.
That method works just fine for any class that is visible in the
environment. It is not just restricted to
or not the class actually exists. Your .context method does that.
David Ashley
On 03/11/2015 09:58 AM, Rick McGuire wrote:
On Wed, Mar 11, 2015 at 10:55 AM, David Ashley w.david.ash...@gmail.com
wrote:
That works for builtin classes, but not for user defined classes. I want
something that can
...@gmail.com wrote:
Just what I was looking for. Just could not find the correct place to
look.
David Ashley
On 03/11/2015 09:57 AM, Rick McGuire wrote:
On Wed, Mar 11, 2015 at 10:48 AM, David Ashley w.david.ash...@gmail.com
wrote:
Rick -
Is there a way to query ooRexx to discover if a class
The arguments passed to the init() method depends on the implementation of
the new method for the class. In the case of array, the new() method needs
to use the arguments to construct the array instance before init() is
called, so none of the arguments are passed along.
Rick
On Mon, Mar 30,
- Windows expert
- D - Project Lifecycle administrator
- E - Installer expert
- F - Builds
- G - The test framework
- H - The test suite (the actual tests written using the test
framework)
- I - The documentation processing (Docbook).
Jon
On 13 April 2015 at 10:48, Rick
Jon,
There's one additional important role that needs to be filled, that of the
language architect. That person's role is to be the final decision maker
on what language features get added to ooRexx. The architect's role is
largely to be the person who says no to features and is responsible
Jon,
This looks like it is a pretty good structure for handling details. I
might be willing to be included in the ARB for the arbitration process
since the current structure is a bit light on actual ooRexx skills.
Rick
On Sun, Apr 26, 2015 at 1:42 PM, Jon Wolfers sahana...@gmail.com wrote:
Some additional major areas:
The test framework
The test suite (the actual tests written using the test framework)
The documentation processing (Docbook).
Rick
On Mon, Apr 13, 2015 at 3:43 AM, Jon Wolfers sahana...@gmail.com wrote:
BCC: RexxLa Board
Dear Developers,
whilst I in no way
and would just cause grief to any developers who were wishing to make the
transition to 64-bitand they would get a crippled implementation that
would not be fully capable of exploiting the 64-bit addressing.
Rick
best regards,
René.
On 29 mei 2015, at 13:30, Rick McGuire object.r
It looks like you are using the GCC version with the namespace linker bug.
Rick
On Thu, Jun 4, 2015 at 3:23 PM, René Jansen rvjan...@xs4all.nl wrote:
Hi Enrico,
thanks for the pointer. Did that, downloaded source and built cmake 3.2.3.
Let it generate a new makefile in a clean directory.
Big Endian is also known as network order in some contexts, but the network
order usage does seem to include a term for non-network order.
Rick
On Fri, May 29, 2015 at 7:06 AM, René Jansen rvjan...@xs4all.nl wrote:
2) Also not sure I like the architecture name, since this is not
really a
wrote:
Where did Sam Detweiler fit in? Didn't he write the original OS/2 Rexx?
Rick McGuire object.r...@gmail.com wrote the following on 27/05/15
14:41:35:
Posting this information here in response to Michael's questions since I
don't have posting access to the RexxLA list:
On Wed
On Thu, May 28, 2015 at 7:55 AM, Michael Lueck mlu...@lueckdatasystems.com
wrote:
Rick McGuire wrote:
Simon developed the initial object-oriented concepts and the original
prototype code. He began working with the development group in Endicott
starting in 1989. There was a lot of work
the
environment just to make it similar.
Rick
Erich
On Thu, Oct 9, 2014 at 1:46 PM, Rick McGuire object.r...@gmail.com
wrote:
Oops, the method was implemented but I missed adding to the method
table. Its in now.
Rick
On Thu, Oct 9, 2014 at 5:39 AM, Erich Steinböck
erich.steinbo
On Sun, May 24, 2015 at 2:54 AM, Mike Cowlishaw m...@speleotrove.com wrote:
.
That is the default situation for a method unless an EXPOSE is used.
Basically, all variables are local unless identified as object scope on the
EXPOSE instruction. USE LOCAL is intended to give you the opposite
It most certainly IS an assignment. The stem variable rr. is set to
contain a reference to the same stem object referenced by the stem variable
r. ooRexx does not have a concept of variable aliasing. This is the same
assignment mechanism that allows USE ARG to work or returning a stem object
How much from scratch? What do you have installed already (svn, cmake,
visual studio of some flavor, nsis, etc.)?
I tend to build things without building the installer and just run things
from the build directory using a simple batch file that sets up paths
appropriately. That might be simpler
? (I have set a few
ones to closed after my commit)
Yes I would. Among other things, it makes it easier to gather up the list
of changes when a new release is cut.
Rick
Erich
On Fri, May 22, 2015 at 8:34 PM, Rick McGuire object.r...@gmail.com
wrote:
Erich,
You might not want to make
On Fri, May 22, 2015 at 4:04 PM, Mike Cowlishaw m...@speleotrove.com wrote:
::method init
USE LOCAL x y z
The variable list should be optional to specify that everything is an
object variable.
::method init
use local -- everything is an object variable.
Slightly concerned
On Sat, May 23, 2015 at 3:24 AM, Mike Cowlishaw m...@speleotrove.com wrote:
::method init
USE LOCAL x y z
The variable list should be optional to specify that everything is an
object variable.
::method init
use local -- everything is an object variable.
Slightly concerned
that is the
inverse of EXPOSE
To: [oorexx:feature-requests] 6...@feature-requests.oorexx.p.re.sf.net
- **status**: accepted -- closed
---
** [feature-requests:#654] Add an instruction that is the inverse of
EXPOSE**
**Status:** closed
**Milestone:** 5.0.0
**Created:** Wed Oct 29, 2014 01:02 PM UTC by Rick
I needed to build the docs or just relied on David's build
machine. David had some instructions for how to setup a machine, but I'm
not finding it right now. Maybe David can chip in with some comments here.
Erich
On Tue, May 26, 2015 at 2:50 PM, Rick McGuire object.r...@gmail.com
wrote
it could not choose a suitable overload for the
reportException call.
Erich, this looks like your change 10916 for setMethod exception
reporting. Do you need help with this?
Jon
On 26 May 2015 at 16:02, Rick McGuire object.r...@gmail.com wrote:
Jon, no. When you do the checkout, all
1301 - 1400 of 2203 matches
Mail list logo
