Not sure what you mean by 'attach to process'?
With the menu Debug/Attach to process
you can select a running process from a list, and when a breakpoint is
reached, the process is stopped on this breakpoint.
In your case, when the exception is triggered, you should have the call
stack in
Hi Rony
2011/6/5 Rony G. Flatscher rony.flatsc...@wu-wien.ac.at
would it be possible to have the REPLY keyword statement for routines?
If I remove the test context-inMethod() then reply *seems* to work from a
::routine (tested with a tiny script).
But this is not a proof that it would work
Hi Rainer
2011/3/14 Rainer Tammer tam...@tammer.net
Hello,
I changed the extension a bit:
I think that this version should be better.
Comments??
Looks good to me. I see a similar code in PHP sources,so...
Let's hope that will solve the problem :-)
Regards
Jean-Louis
Hi
Still learning about concurrency in ooRexx...
Why do I have a deadlock in the green line if I remove the yellow line ?
Before the green line, I do guard off. So the yellow line should not have an
impact on the green line...
It seems that the lock is reacquired by guard on, even if the
2011/3/14 Rick McGuire object.r...@gmail.com
Guard conditions nest. So the guard off in the m1 method does not
completely remove the locking from that thread. It only removes the
guarded condition caused by entry to that method, so the object is
still locked by that thread.
ah ! ok... I
Hi
I had a look at
https://sourceforge.net/projects/oorexx/forums/forum/408477/topic/4404931
I see no difference in code beetween 320 and 410 which could explain the
failure.
But there is a difference in test case : 64 bits vs 32 bits. And I wonder if
the declaration of addr is ok :
long
The PNG images are needed for the HTML doc.
Jean-Louis
2011/3/7 Mark Miesfeld miesf...@gmail.com
I haven't kept up with the image changes in the rest of the docs, so
I'm not sure. In the oodialog section some of them are still used.
But, I agree that unused ones should be removed.
--
Hi
Currently, we don't have a uniform way to get the current thread id from a
rexx script.
Under Windows : SysQueryProcess(TID)
if (stricmp(option, TID) == 0)
{
return context-WholeNumber(GetCurrentThreadId());
}
Under Unix : Sysgettid()
pthread_t tid = pthread_self();
2011/3/5 Rick McGuire object.r...@gmail.com
I guess I'd need to know a reason. If the purpose is to somehow map
this back to the Rexx thread, then I'm not comfortable with exposing a
system-specific concept such as a thread id to any ooRexx construct.
The issue of Rexx thread identifiers
2011/3/3 Rick McGuire object.r...@gmail.com
A second guess here. If SIGNAL ON ANY is used, then this is probably
trapping the NOMETHOD condition. If nothing is trapping the NOMETHOD
condition, then the SYNTAX error is raised. There are no error codes
or message associated with NOMETHOD.
Obviously, I forgot to update the comment in the test case...
Here is the test case with the right comments for 'signal on nomethod'
Regards
Jean-Louis
condition.rex
Description: Binary data
--
Free Software Download:
2011/3/3 Rick McGuire object.r...@gmail.com
Because NOMETHOD is not a SYNTAX error, so there's no message or error
code associated with it. It has information specific to it's
particular situation, but it is not a SYNTAX error.
ok, thanks.
After changing my catch all to both 'signal on
There will be a portable ooRexxTry-GUI called ooRexxTry.rxj from a
student, which I will enclose in the next release of BSF4ooRexx, which looks
like the Windows ooRexxTry.rex, but has a few bells and whistles more, and
most importantly, it runs on all platforms BSF4ooRexx is available (i.e.
the hex id value should still be communicated somehow;
either turned on by a switch or a table mapping at the start/end of
tracing where the symbolic values to id values are given.
On 23.02.2011 13:54, Jean-Louis Faucher wrote:
Hi
I'm thinking of modifying my sandbox to add informations about
Hi
I'm thinking of modifying my sandbox to add informations about current
thread, activation and scope lock when tracing.
But maybe I miss the point and a better (simpler) technique exists... Let me
know.
To illustrate my problem (T1, A1, P1 added by me):
coroutine~start -- guarded
Hi
I was about to add the File class at the begining of chapter 5 in rexxpg.
But I see that many other classes are missing, and that's maybe wanted (I
mean to show only a subset of the classes)
So, I made a review of what's missing, let me know if I should update the
list with the lines marked
ok, then I will update the list
Regards
Jean-Louis
2011/2/20 Rick McGuire object.r...@gmail.com
On Sun, Feb 20, 2011 at 8:57 AM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Hi
I was about to add the File class at the begining of chapter 5 in rexxpg.
But I see that many other
Hi
After reading this sentence in the doc of 'forward' :
If you do not specify the CONTINUE option, the language processor
immediately exits the current
method before forwarding the message
I expected to benefit of a so-called tail call elimination, but that's not
the case... I get a stack
removed from the ooRexx call stack.
Rick
On Sun, Feb 20, 2011 at 3:11 PM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Hi
After reading this sentence in the doc of 'forward' :
If you do not specify the CONTINUE option, the language processor
immediately exits the current
method
, 2011 at 3:53 AM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Hi
Without the makeArray method below, I get an error.
It's because of the test
if (!isOfClass(Method, methobj))
in RexxObject::run
m = .MyMethod~new
STATES OF AMERICA]
a~initCap=[United States Of America]
Is that correct?
If so, then this would really solve most such needs and problems in a very
easy and elegant manner and opens up very interesting new opportunities for
ooRexx!
---rony
On 18.02.2011 20:37, Jean-Louis Faucher wrote
but a professional tool on which they
count. So I understand totally your negative vote.
Rick
On Sat, Feb 19, 2011 at 12:02 PM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Rony
That's correct.
I can run your script in my sandbox, and I get the result you are
expecting.
Just
not
a
good argument, but ooRexx is first of all a hobby for me :-)
For some people, ooRexx is not a hobby but a professional tool on which
they
count. So I understand totally your negative vote.
Rick
On Sat, Feb 19, 2011 at 12:02 PM, Jean-Louis Faucher
jfaucher...@gmail.com wrote
on
which
they
count. So I understand totally your negative vote.
Rick
On Sat, Feb 19, 2011 at 12:02 PM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Rony
That's correct.
I can run your script in my sandbox, and I get the result you are
expecting
Hi
Without the makeArray method below, I get an error.
It's because of the test
if (!isOfClass(Method, methobj))
in RexxObject::run
m = .MyMethod~new(, 'say hello self')
.MyString~new(john)~run(m)
::class MyString subclass String
the significant overhead associated
with needing to use late binding on every method call used with these
items.
Rick
On Fri, Feb 18, 2011 at 3:53 AM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Hi
Without the makeArray method below, I get an error.
It's because of the test
Hi
When debugging, I'd like to find quickly the calls to a given method, while
keeping all the details displayed by trace i before the call... A local
trace activated only in the method is not enough.
Is there a way to get this kind of output without adding the yellow lines ?
8 *-* ::class
, Feb 17, 2011 at 3:40 AM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Hi
When debugging, I'd like to find quickly the calls to a given method,
while
keeping all the details displayed by trace i before the call... A local
trace activated only in the method is not enough
architectural reasons why this shouldn't be pursued?
Jean-Louis Faucher wrote:
Hi
Are there some reasons to not support :
do c over string
do index, item over supplier -- currently we don't support two
variables, but...
?
I was also thinking to generators... For fun, I wrote
Hi
From the doc, I understand that 'guard on' can wait until a change is made
to an exposed variable.
According to my tests, it works when the variable is exposed in the current
method, like that :
::method guardGenerate
expose status stopped
guard on when status ==
Hi
Are there some reasons to not support :
do c over string
do index, item over supplier -- currently we don't support two variables,
but...
?
I was also thinking to generators... For fun, I wrote a generator class
whose spec is
::class generator
::method generate : does a reply and start the
Hi Rony,
Have a look at platform/unix/makepkg.rex :
* Prerequisites:
* make DESTDIR=`pwd`/tmp install
*has been run before this
* Invocation:
* Should not be executed directly, but via:
* make DESTDIR=`pwd`/tmp package
*in the directory where ooRexx is built.
I followed
like
to tackle building an installable package? grin
--
Mark Miesfeld
On Mon, Feb 7, 2011 at 2:51 PM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Hi Rony,
Have a look at platform/unix/makepkg.rex :
* Prerequisites:
* make DESTDIR=`pwd`/tmp install
*has been run before
There is a skeleton InternalClasses in the wiki, currently empty.
Is there other sites I'm not aware of ?
I have accumulated a few notes coming from :
- QR in the developper list,
- how some bugs have been fixed,
- notes taken on the fly, while debuging.
It's raw material and certainly not a
The qualified path is memoized by all methods except init, and I think that
init should memoize too.
See the value returned by the last expression :
'cd' value(windir,,ENVIRONMENT)
system_ini = .File~new(system.ini)
win_ini = .File~new(win.ini)
say system_ini~canRead -- memoization of qualified
Rick
I did not yet study the implementation of the File class, but the methods
look very similar to the Java File class.
http://download.oracle.com/javase/6/docs/api/java/io/File.html
If this is it the case, then I'm inclined to copy-paste some portions of the
Java File documentation... I'm not
would *strongly* advise against copying and pasting the Java
documentation (or any other copyrighted documentation or code) into the
ooRexx project, unless Oracle was asked for permission. (One argument in
the case Oracle vs. Google is right about copyrighted material, it seems
to even
The private methods are not documented, right ?
Ex :
deleteDir
deleteFile
normalizePathSyntax
qualifiedPath
Jean-Louis
--
Nokia and ATT present the 2010 Calling All Innovators-North America contest
Create new apps games
Mark,
I finished the review of changes for 4.1.0.
Feel free to fix any error/omission I may have done.
CHANGES :
Section Patches is empty, to remove ?
ReleaseNotes :
remember : TODO correct file list with actual files at release
Some sections are empty, to remove ?
New platforms
Some
I believe I have the bug, documentation, RFE, and patches tracker
database items pretty cleaned up. Every thing going in the 4.1.0
release is marked as 'pending' and 'next release.' If you do a filter
on those, you should see all the relevant items. Feel free to update
the CHANGES file.
I placed the Draft 2 docs here.
http://build.oorexx.org/builds/release-candidates/
Please review them and let me know if another draft is necessary.
David, I checked the tables modified with pgwide, it's ok.
readme : still version 4.0.0
The list of platforms is to update ? ex : Windows
ok, I started the work.
But while doing that, I had some doubts for the left position of the
table... And I discovered a problem that I didn't see at first sight in
winextensions.pdf :
With pgwide=1, the table 4.2 is left-aligned on the column of the
varlistentry Event Log Record.
Without pgwide,
Since the build I made under Cygwin looks good, I decided to commit both in
trunk and in branch 4.1.0.
David, you can create a draft2 version of the docs.
Jean-Louis
--
Nokia and ATT present the 2010 Calling All
Hi
I committed an example in my sandbox which illustrates the performance
problem of MutableBuffer~replaceAt.
http://oorexx.svn.sourceforge.net/viewvc/oorexx/sandbox/jlf/samples/mutablebuffer/
As i said already, a possible workaround could be to pass the tailSize from
replaceAt when calling
Hi Mark
I need your help to understand initPropSheetHeader...
The PROPSHEETHEADER structure contains LPCTSTR strings, so when built in
byte char mode, the strings are made of byte char.
But I see that you do a conversion ansi -- unicode and store the result in
pcpsd-caption and psh-pszCaption,
Hi
.mutablebuffer~new('abcdef')~replaceat('XXX', 1, 16)
makes the interpreter crash.
I plan to open a bug and apply the fix I made in sandbox.
See adjustGap in
http://oorexx.svn.sourceforge.net/viewvc/oorexx/sandbox/jlf/trunk/interpreter/classes/BufferClass.hpp?revision=6144view=markup
Is this
thanks Mark
Now, I understand... a good caption is indeed better than garbage :-)
Jean-Louis
2010/9/7 Mark Miesfeld miesf...@gmail.com
On Mon, Sep 6, 2010 at 2:49 PM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
I need your help to understand initPropSheetHeader...
Jean-Louis,
I
ok, I do nothing for the moment.
Let me know if I have something to do later (that will be tomorrow)
Jean-Louis
2010/9/7 Rick McGuire object.r...@gmail.com
On Mon, Sep 6, 2010 at 6:37 PM, Jean-Louis Faucher
jfaucher...@gmail.com wrote:
Hi
.mutablebuffer~new('abcdef')~replaceat('XXX', 1
Hi
Good to know : when running
rexx testOORexx.rex -R ooRexx\base\bif -f charout
from an USB hard drive, under Win XP service pack 3
I have three failures in :
TEST_MANY_CHARS
TEST_MANY_CHARS_BIN
TEST_MANY_CHARS_NB
because charout can't write the string and returns the string's size instead
of
Hi Bruce
In unix/rexxutil.cpp :
#ifdef __APPLE__ __MACH
generates this warning under Linux
./extensions/rexxutil/platform/unix/rexxutil.cpp:170:18: warning: extra
tokens at end of #ifdef directive
Don't know if this expression is allowed by your compiler, so I forward for
verification :-)
sources.
If we keep this possibility for ooRexx, maybe that could let deliver a
byte-char version totally compatible, while leaving an open door for
necessary evolutions in the wide-char version ?
Jean-Louis
2010/8/30 Rick McGuire object.r...@gmail.com
On Mon, Aug 30, 2010 at 4:51 PM, Jean-Louis
We use oorexx at work, under Citrix. No admin rights.
The administrator has disabled the cmd shell, but oorexxshell works
perfectly. I hope he's not reading that :-)
Jean-Louis
--
Sell apps to millions through the
Workaround (?) Try to configure directly from trunk, not from trunk/tmp
I remember I had problems with make install a few years ago, when
configuring from a directory different from trunk (was 32 bits).
I never tried again, so I can't tell if the problem is still there or not...
Jean-Louis
If I'm not wrong, a closing quote is missing in trunk/Makefile.am
Replace
sed -e s/^VER=/VER=$(ORX_MAJOR)/ -e s/^REL=/REL=$(ORX_AGE)/ -e
s/^MOD=/MOD=$(ORX_REVISION) \
by
sed -e s/^VER=/VER=$(ORX_MAJOR)/ -e s/^REL=/REL=$(ORX_AGE)/ -e
s/^MOD=/MOD=$(ORX_REVISION)*/* \
I can't test so I prefer to not
Hi,
Some weeks ago, I fixed two crashes I had in my sandbox, and maybe those
fixes are candidate to go in trunk...
Let me know...
6053
sandbox\jlf\trunk\interpreter\parser\SourceFile.cpp
Fix a crash in GC (markObjectsMain) : a string popped from liveStack has a
corrupted behaviour's vtable
On the reader side, we have the status ERROR at the end because of that :
void StreamInfo::readBuffer(char *data, size_t length, size_t bytesRead)
{
if (!fileInfo.read(data, length, bytesRead))
{
fprintf(stderr, \n*** StreamInfo::readBuffer\n);
notreadyError();
}
...
sorry, no experience with fop.
at work, we use xep, maybe you can download the free personal
editionhttp://www.renderx.com/download/personal.htmland see if you
have a more clear error message, or even no error at all.
Jean Louis
Hi Mark
Some problems I had when running from trunk :
editrex.rex : Error 98.900: the storage allocated for the dialog template
is too small
Idem with ftyperex.rex, oostddlg.rex, oostdfct.rex, ...
oobandit.rex : GETCONTROLDATAATTRIBUTE is not a method of a BANDITDLG
oodraw.rex : the dialog
I'm testing RexxGtk and I get an assert error in test2-1.rex.
Gtk-CRITICAL **: gtk_widget_show: assertion `GTK_IS_WIDGET (widget)' failed
David, is it working for you ?
Maybe I'm totally wrong, but it seems that the CSELF assignment should be
moved to GtkWidget.
When debugging show, I see that
One year ago, I started to investigate Valgrind (under Linux) because I had
poor performances with csvStream, compared with stream + parse.
The datas, scripts and results are available here :
http://jean-louis.faucher.perso.neuf.fr/ooRexx/profiling/csv
If I remember correctly, the numbers you see
Nothing to recompile, you just run Valgrind by passing the path to the
executable. But the execution is very slow...
Jean-Louis
--
___
Oorexx-devel mailing list
Hi
I can't compile rexxutil.cpp under PuppyLinux because xattr/xattr.h is not
found.
Maybe I need to install something, I did not investigate.
Temporarily bypassed by surrounding with
#if defined( HAVE_ATTR_XATTR_H)
I don't mind to stay like that, I have just to be careful to not commit that
in
yes, it's a minimalist distribution.
I can live with that, no problem.
Thanks !
Jean-Louis
--
___
Oorexx-devel mailing list
Oorexx-devel@lists.sourceforge.net
Hi,
I'd like to create a sandbox to experiment changes that will not go to
official ooRexx.
I'm reading the tutorial written by Rick two years ago :
http://sourceforge.net/mailarchive/forum.php?thread_name=33bd05330711150446l30ee70b2m43978a99e5426068%40mail.gmail.comforum_name=oorexx-devel
Is it
Thanks Rick
Yes, I plan to merge sometimes (say once per week), to keep my sandbox
synchronised with trunk.
Since the procedure is good, I will start the creation of the sandbox now.
I'm not very used to run SVN from the command line because I use Tortoise
SVN under Windows. I did not find how to
If I modify RexxClass::defineMethod to no create a copy of instanceBehaviour
then 123~reduce works as expected (reduce is found).
I understand from the comment in the code that it's a design decision to
make a copy :
*/* make a copy of the instance */
/* behaviour so any previous
Thanks Rick,
My understanding of the situation :
Internally, the class of 123 is Integer, not String.
Externally, we see it as a String class, but there are some differences :
identityHash returns a value derived from the C++ instance address, so we
have two distinct values for
You will say I'm looking for troubles, but I'm curious to know why
.Class~define is not available on Rexx defined classes.
I removed the check in RexxClass::defineMethod to see what happens... My toy
script works good, but I had to bypass a problem with RexxInteger : when the
unknown method is
I tested under Linux : the reader gets all the datas sent by the writer.
It's because the eof condition is managed by the ::read method (attribute
fileeof)
Don't know if we can emulate the windows behavior under Linux, but I prefer
the current Linux behavior...
Jean-Louis
Rick
With the current fix for pipes, we don't catch all the datas sent by the
writer below.
But i don't know if it's a bug or a feature...
Indeed, lines() and char() returns 0 when the transient stream has no more
data. It doesn't call _read, and thus has no chance to see new incoming
datas.
The
I suspected that the rules would be more complicated :-)
Ok, I don't investigate more on the automatic transformation.
Maybe I can still help by generating the list of links having several words
in their child text.
Can't attach the resulting file, so I will commit the file links with
several
The number of links to update seems high (I found more than 2000 linkend in
oodialog, assuming all are candidate for update). Proposition : I can
modify the script transformdir in DocMusings to apply these transformations
for endterm.
By default, this script creates a clone of the directory
Hi Rony
Maybe you downloaded the sdl version of pdcurses ? (pdc34dlls.zip)
This one depends on SDL.dll, that could explain the problem...
If you have depends.exe (http://dependencywalker.com), try : depends
orxncurses.dll
I downloaded pdc34dll.zip.
Jean-Louis
David,
I was curious to see orxcurses in action under Windows, so I wrote a
makefile.win which depends on PDCurses.
Looks good and working well !
All the tests are ok (positions, colors, scrolling...) except the following
points :
The tilde is echoed ^@ and I must hit another key to see it :
Assuming you have the file address.rex which contains :
if arg() 0 then say -
say main: address()
call myProcedure
call myRoutine
.myClass~myMethod
if arg() == 0 then call address.rex again
return
myProcedure: procedure
say myProcedure: address()
return
::routine myRoutine
say myRoutine:
ok, then I open a bug ?
Do you plan to fix it or do I try ? For the routine calls, I think that
RoutineClass::callRexx must pass the default environment when doing
code-call(...). Did not look yet at the method calls.
Jean-Louis
SysFileCopy and SysFileMove are now available for Unix.
Is it ok to update the doc accordingly ? (basically remove Windows only)
There is a script for the tests :
test/trunk/ooRexx/base/rexxutil/platform/unix/SysFileCopyMove.sh
Currently not integrated in the test framework.
Usage :
(copy to text editor for proper alignment)
While parsing the rexxgtk documentation, I had an error (now fixed) because
of this syntax diagram :
treelist.xml
section id=gtltreestoresetvalue
.*+--, col, val--+*
.V |
It works for me, with a fresh rebuild of ooRexx.
This script displays 81 when run from hostemu directory :
address hostemu 'execio * diskr readme.txt (finis stem in.'
say in.0
::requires hostemu LIBRARY
Jean-Louis
--
Some questions about stream sharing options (just trying to understand why,
under Windows, my log file is not updated by the callee when both the caller
and the callee open a stream on the same log file)
Question 1)
Documentation
.stream~open
SHARED
Enables another process to work with the
[Resend the body of the previous mail, because seems not visible from
mailing list]
Assuming that the syntax diagrams will not be replaced soon by something
else (let me know :-), I'd like to apply the fixes described in the previous
mail.
Those fixes are 100% compatible with the current toolset.
Thanks Mark
Yes, there is no impact on build, no impact on delivery.
I wait for the agreement of David before starting the work :-)
Jean-Louis
--
This SF.Net email is sponsored by the Verizon Developer Community
Take
Great ! Thanks David
Jean-Louis
--
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app
David,
I tested align=center and without any align.
In both cases, the HTML looks good, no text around the image.
But with align=center, the image is centered in the pdf, whereas it's left
aligned when no attribute aligned.
So removing the align attribute seems the most appropriate to keep the
While working on rexxref doc, I see that most of the class images are left
aligned.
Because of that, we have some text displayed on the right of the images in
the HTML documentation.
Is it intended ? I don't think so, but...
Let me know if I can remove these align=left (also in rxsock, and maybe
Hi,
Maybe this is already a known problem by the group, but I was unaware of
that till today...
Under win32 (and probably win64), call SysFileTree *.odg, files, FO
will return all the following files :
f.odga
f.odg1
.~lock.dlgArea.odg#
This is because the search made by FindFirstFile (and
Jon
Thanks for the odg files.
You can have a look at the build 5396 on the build machine : in oodialog,
dlgArea looks great page 423 :-)
For sample1, I extracted the png and refer it directly : the pdf format does
not bring more quality here (bitmap).
Jean-Louis
This proposition has no impact on the HTML documentation.
By removing the .png suffix of the images in the sgml files, the system
(Jade) can decide which format to look for : .pdf when target is print, .png
when target is html.
This is parameterized in the customation layer oorexx.print.dsl (to
I propose to do the changes on rxsock, and test the result with the build
machine.
Then I will apply the changes to the rest of the documentation, if it works.
Here are my questions :-)
Q1
In which directory do I apply the changes : docs/trunk ? (I think it's
mandatory for the build machine)
Jon
They should be in docs/trunk/oodialog.
I'm working there (trunk).
Jean-Louis
--
Return on Information:
Google Enterprise Search pays you back
Get the facts.
http://p.sf.net/sfu/google-dev2dev
Tested with the build machine, rxsock.pdf looks good except the empty area
under the figures.
I think it's the format of the page that must be adjusted in OODraw, to be
limited to the image area.
In the html doc, I see that some text is displayed on the right of the
figures.
Is it intended ?
Ok, thanks.
I'm currently playing with the transformation of the ooRexx doc, and
(naively) tried to reduce the duration by creating one thread per sgml file.
From what I read in several posts about Python/Ruby, it seems that the best
way to exploit 100% of the multicores is to use a fork instead
Hi Rony,
On my system, WC_ERR_INVALID_CHARS is defined when WINVER = 0x0600 (in
winnls.h)
If I understand correctly
http://msdn.microsoft.com/en-us/library/aa383745(VS.85).aspx, this is the
case for Window 7, Windows Server 2008, Vista.
It's strange it works for me because I have WinXP SP2 and
I'm curious to know why different rules apply for arguments to methods
(rexxref p 104)
In case of method call, why the string message is not sent when
request(string) returns .nil ?
d = .directory~new
say(d) -- a Directory
.stdout~say(d) -- error argument 1 must have a string value
This works
trace ?i
...some intermediate code
r=myobj~mymeth
When I see that a result is wrong, I can rexecute the previous statement
with =
but how can I step into the called method ?
I could stop the debug, put trace ?i inside the called method and restart
the debug, but is there a more direct technique
Rick
I see that MAXIMUM_PATH_LENGTH has no value in last #else.
This piece of code supports the case where PATH_MAX is not defined.
But we have some files where PATH_MAX is used inconditionally.
Seems not a problem on the current platforms, but I suppose that should be
replaced by
ok, I see your points.
My attempt to use expose in a ::routine was because, to me (beginner), a
:routine is similar to a procedure or a method (it's a piece of code that I
can execute). I see expose allowed in procedure and in method (as a
beginner, I'm not supposed to master the difference
You can see an example of PDF generated by XEP personal edition :
http://jean-louis.faucher.perso.neuf.fr/ooRexx/rexxref-XEP_Personal_Edition.pdf
The stamp is at the bottom of each page.
Generated with the default stylesheet (no ooRexx custom files).
I downloaded the personal edition from :
David,
Thanks for these precisions, I understand your concerns and to reassure
everybody, I don't plan to change anything in the official ooRexx
documentation (in relation with the current subject :-). And I would be
happy to work on RR, if a consensus is reached by the group. Or any other
Hi,
Mark made a call to volunteers for documentation improvements in a comment
of defect 2808222.
I'm interested to work on the following changes, if the group agrees :
Convert all the sgml files to XML DocBook (was started by David in tryxsl).
Use DocBook XSL to generate final doc
PassiveTeX is
101 - 200 of 220 matches
Mail list logo
