I commented out mean by putting NB. at the front of the line of its definition. 
The script that I ran is the one that is attached to the email below and I see 

>> NB.mean =: +/ % #  NB. definition of mean - VERB

immediately above the definition of trajectory. 

I am running 

iMac Pro (2017)
Processor 3 GHz Intel Xeon W
Memory 64 GB 2666 MHz DDR4
Graphics Radeon Pro Vega 64 16 GB

macOS Mojave Version 10.14.6

Cheers, bob

> On Oct 17, 2019, at 4:17 PM, Henry Rich <[email protected]> wrote:
> 
> It doesn't fail for me under Windows.  The script you sent didn't have 'mean' 
> commented out, but when I commented it out it still didn't fail.
> 
> Henry Rich
> 
> On 10/17/2019 7:03 PM, 'robert therriault' via Beta wrote:
>> Okay Henry,
>> 
>> I started a fresh session
>> 
>>     JVERSION
>> Engine: j901/j64avx2/darwin
>> Beta-m: commercial/2019-10-02T11:28:59
>> Library: 9.01.10
>> Qt IDE: 1.8.3/5.12.3
>> Platform: Darwin 64
>> Installer: J901 install
>> InstallPath: /users/bobtherriault/j901
>> Contact: www.jsoftware.com
>> 
>> Opened debug (no verbs)
>> 
>> loaded the following script from a temp file
>> 
>>  cocurrent 'base'
>> require 'plot'
>> MAXITER=: 50    NB. safety long-stop
>> S=: 100         NB. height above ground (m) - UPDATED
>> T=: 0           NB. time at end of current epoch (s) - UPDATED
>> e=: 0.1         NB. time interval of simulation epoch (s) - CONST
>> U=: 0           NB. starting velocity (m/s) - CONST
>> V=: U           NB. velocity at end of current epoch (s) - UPDATED
>> g=: 9.81        NB. acceleration due to gravity at earth surface (m/s^2) - 
>> CONST
>> NB.mean =: +/ % #  NB. definition of mean - VERB
>> 
>> trajectory=: monad : 'u F: v y'
>> 
>> u=: monad define
>>   NB. Trajectory of solid object in free-fall
>> _3 Z: MAXITER             NB. Stop Fold when there have been too many 
>> iterations
>> _2 Z: S<:0                NB. Stop Fold when object hits the ground
>>   NB. Recurrence relation is based on free-fall formula: V = U + gt
>> V=: y + g*e               NB. free-fall: vertical velocity at end of epoch
>> S=: S - e*mean y,V        NB. free-fall: vertical height at end of epoch
>> T=: T + e                 NB. time of flight at end of epoch
>> if. S<:0 do. smoutput '+++ LANDED: T=',":T end.
>> V return.
>> )
>> 
>> v=: monad : 'S'
>> 
>> plot trajectory U
>> 
>> 
>> Note that 'mean' is commented out, and when I loaded this script the session 
>> crashed.
>> 
>> I am emailing the crash report directly to you.
>> 
>> Let me know if there is anything else you require.
>> 
>> Cheers, bob
>> 
>>> On Oct 17, 2019, at 3:51 PM, Henry Rich <[email protected]> wrote:
>>> 
>>> Certainly, if you have a crash I'd like to know about it.  What I need is a 
>>> script that will crash when run from a fresh session.
>>> 
>>> Henry Rich
>>> 
>>> On 10/17/2019 6:47 PM, 'robert therriault' via Beta wrote:
>>>> Sorry about that Henry, I have updated now.
>>>> 
>>>> On the last run, I ended up having to go into the Activity Monitor to 
>>>> force quit the J process.
>>>> 
>>>> When I reopened the session, I updated the addons (including base) and 
>>>> restarted and the reloaded the project that the trajectory script was in. 
>>>> When I loaded that script the first time:
>>>> 
>>>>     load '/users/bobtherriault/j901-user/projects/enhanced/jig6.ijs'
>>>> |fold limit: trajectory
>>>> |       u F:v y
>>>> |[-346] /users/bobtherriault/j901-user/projects/enhanced/jig6.ijs
>>>> 
>>>> Subsequent loading of the script results in everything running smoothly 
>>>> even after executing clear '' plots display as expected
>>>> 
>>>> Loading with mean commented out and debug off I get
>>>> 
>>>>     load '/users/bobtherriault/j901-user/projects/enhanced/jig6.ijs'
>>>> |value error: trajectory
>>>> |       u F:v y
>>>> |[-346] /users/bobtherriault/j901-user/projects/enhanced/jig6.ijs
>>>> 
>>>> Turning debug on after executing clear '' of course revealed no verbs in 
>>>> the base locale. Loading the script again caused J to crash ( I can send 
>>>> the crash report if that is useful)
>>>> 
>>>> Restarting J and running the script I once again get the expected:
>>>> 
>>>>     load '/users/bobtherriault/j901-user/projects/enhanced/jig6.ijs'
>>>> |value error: trajectory
>>>> |       u F:v y
>>>> |[-346] /users/bobtherriault/j901-user/projects/enhanced/jig6.ijs
>>>> 
>>>> Then I clear '' and this time load the script first and then start the 
>>>> debugger (the default is trajectory). Loading script crashes again ( crash 
>>>> report saved for this one as well)
>>>> 
>>>> Let me know if you want the crash reports.
>>>> 
>>>> Cheers, bob
>>>> 
>>>>> On Oct 17, 2019, at 3:17 PM, Henry Rich <[email protected]> wrote:
>>>>> 
>>>>> You have not updated the dev/fold addon to 1.0.5 level.
>>>>> 
>>>>> Henry Rich
>>>>> 
>>>>> On 10/17/2019 6:14 PM, 'robert therriault' via Beta wrote:
>>>>>> Hi Gilles,
>>>>>> 
>>>>>> When I comment out the definition of mean and then clear the base locale 
>>>>>> and run with debug on I get the following in the top pane of the 
>>>>>> debugger I get 59 lines with a '>' indicating line 41 (would be simpler 
>>>>>> if I could copy the results that show up in the pane but don't know how).
>>>>>> 
>>>>>> In the second pane I get:
>>>>>> value error
>>>>>> Fold_j_[41] 13!:8 ] 13!:11''[FoldZv_j_=:fzv
>>>>>> trajectory[0] u F: v y
>>>>>> script[0] {: 3 : '0!:0 y [ 4!:55<''y''' jpath_z_&.:>
>>>>>> fn[tacit] script
>>>>>> load[7] fn fl
>>>>>> load[0] 0 load y
>>>>>> 
>>>>>> I am not sure how to interpret that, but is the result that I get. I am 
>>>>>> running it by loading a script with the definitions and final line:
>>>>>> 
>>>>>> plot trajectory U
>>>>>> 
>>>>>> I get the same result running that line in the Qt IDE
>>>>>> 
>>>>>> And now the debugger has become non operational. Buttons do highlight, 
>>>>>> but do not do anything and the debugger will not clear or exit. The 
>>>>>> script window can be edited and loaded but the execution window only 
>>>>>> responds with the load command and does not update. Toggling between 
>>>>>> windows does not include the debug window, although it can be selected 
>>>>>> when clicked on.
>>>>>> 
>>>>>> Hope this helps!
>>>>>> 
>>>>>> Cheers, bob
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>>> On Oct 17, 2019, at 2:26 PM, Gilles Kirouac <[email protected]> wrote:
>>>>>>> 
>>>>>>> Bob
>>>>>>> 
>>>>>>>   The trajectory script uses *mean* in the definition of u. If you run
>>>>>>> trajectory *without* the  definition of mean, it should result in a
>>>>>>> value error. Right?
>>>>>>> 
>>>>>>>   Now run the same under the J debugger to see if it can pinpoint that
>>>>>>> the value error is on *mean*.
>>>>>>> 
>>>>>>>   I will do the same on Win10 shortly.
>>>>>>> 
>>>>>>>   ~ Gilles
>>>>>>> 
>>>>>>> Le 2019-10-17 à 17:05, 'robert therriault' via Beta a écrit :
>>>>>>>> Trajectory runs fine for me on a Mac.
>>>>>>>> 
>>>>>>>>    JVERSION
>>>>>>>> Engine: j901/j64avx2/darwin
>>>>>>>> Beta-m: commercial/2019-10-02T11:28:59
>>>>>>>> Library: 9.01.08
>>>>>>>> Qt IDE: 1.8.3/5.12.3
>>>>>>>> Platform: Darwin 64
>>>>>>>> Installer: J901 install
>>>>>>>> InstallPath: /users/bobtherriault/j901
>>>>>>>> Contact: www.jsoftware.com
>>>>>>>> 
>>>>>>>> I note that the Qt IDE and the libraries differ from Gilles, but this 
>>>>>>>> would be natural for a different operating system, right?
>>>>>>>> 
>>>>>>>> Cheers, bob
>>>>>>>> 
>>>>>>>>> On Oct 17, 2019, at 1:51 PM, Henry Rich <[email protected]> wrote:
>>>>>>>>> 
>>>>>>>>> I don't have Linux.  It runs correctly on Windows for me.
>>>>>>>>> 
>>>>>>>>> Henry Rich
>>>>>>>>> 
>>>>>>>>> On 10/17/2019 4:30 PM, Gilles Kirouac wrote:
>>>>>>>>>>  I am now getting an interpreter crash.
>>>>>>>>>> 
>>>>>>>>>>  I updated three scripts through PacMan. dev/fold is now at level
>>>>>>>>>> 1.0.5. Then ran the same Trajectory script. Without the debugger, I 
>>>>>>>>>> get
>>>>>>>>>> the normal value error: trajectory. Doing the same with the J 
>>>>>>>>>> debugger
>>>>>>>>>> but with the J session run under gdb here is the result:
>>>>>>>>>> 
>>>>>>>>>> 
>>>>>>>>>> gilles@T530gk ~/j901 $ gdb bin/jqt
>>>>>>>>>> ...
>>>>>>>>>> (gdb) run
>>>>>>>>>> Starting program: /home/gilles/j901/bin/jqt
>>>>>>>>>> [Thread debugging using libthread_db enabled]
>>>>>>>>>> Using host libthread_db library 
>>>>>>>>>> "/lib/x86_64-linux-gnu/libthread_db.so.1".
>>>>>>>>>> [New Thread 0x7fffe4b49700 (LWP 8058)]
>>>>>>>>>> 
>>>>>>>>>> Thread 1 "jqt" received signal SIGSEGV, Segmentation fault.
>>>>>>>>>> 0x00007fffda5cd1ed in suspset () from /home/gilles/j901/bin/libj.so
>>>>>>>>>> (gdb)
>>>>>>>>>> 
>>>>>>>>>>  I did it twice to make sure I was not doing something silly.
>>>>>>>>>> 
>>>>>>>>>>  ~ Gilles
>>>>>>>>>> 
>>>>>>>>>> Le 2019-10-17 à 11:51, Henry Rich a écrit :
>>>>>>>>>>> Gilles,
>>>>>>>>>>> 
>>>>>>>>>>>  I have modified the Fold script so that it allows messages from
>>>>>>>>>>> execution of the user's verb to be displayed and debugged.  Use
>>>>>>>>>>> Package Manager to update dev/fold - no need for a new beta.
>>>>>>>>>>> 
>>>>>>>>>>>  I hope that suits better.
>>>>>>>>>>> 
>>>>>>>>>>> Henry Rich
>>>>>>>>>>> 
>>>>>>>>>>> On 10/11/2019 11:27 AM, Gilles Kirouac wrote:
>>>>>>>>>>>> I installed beta-k then m on Linuxmint 64b
>>>>>>>>>>>> 
>>>>>>>>>>>> My desktop directory is named Bureau, not Desktop. The installer 
>>>>>>>>>>>> is thus
>>>>>>>>>>>> unable to install shortcuts on Desktop. Minor annoyance. I created 
>>>>>>>>>>>> a
>>>>>>>>>>>> link to Bureau for next time. May be suggest to do so in install
>>>>>>>>>>>> instructions. I added a few qt5 missing packages.
>>>>>>>>>>>> 
>>>>>>>>>>>> With the current IDE, to find the syntax of std lib utilities (e.g.
>>>>>>>>>>>> dates, files, strings, text), I have to go through Help, Help, Usr,
>>>>>>>>>>>> Standard Library, ... More importantly I have to KNOW that the std 
>>>>>>>>>>>> lib
>>>>>>>>>>>> doc is there.
>>>>>>>>>>>> 
>>>>>>>>>>>> Do you think a newcomer will easily explore to that level? The 
>>>>>>>>>>>> Library
>>>>>>>>>>>> is a selling point. It should be easy to get to. And IMO the Lib 
>>>>>>>>>>>> is more
>>>>>>>>>>>> useful to anyone than Foreigns.
>>>>>>>>>>>> 
>>>>>>>>>>>> Two years ago, I proposed to add a line to the Help Menu pointing 
>>>>>>>>>>>> to the
>>>>>>>>>>>> std lib doc. IIRC the proposal was accepted, but ... so I 
>>>>>>>>>>>> reiterate.
>>>>>>>>>>>> 
>>>>>>>>>>>> Having launched jqt through a terminal, I received the following:
>>>>>>>>>>>> 
>>>>>>>>>>>> QPaintDevice::metrics: Device has no metric information
>>>>>>>>>>>> 
>>>>>>>>>>>> Should I worry?
>>>>>>>>>>>> 
>>>>>>>>>>>> I briefly experimented with Fold after reading the Release Notes. 
>>>>>>>>>>>> Those
>>>>>>>>>>>> could include a link to 
>>>>>>>>>>>> https://code.jsoftware.com/wiki/Vocabulary/fcap,
>>>>>>>>>>>> which I found through a web search.
>>>>>>>>>>>> 
>>>>>>>>>>>> I tried to run the Trajectory example from that page and got a 
>>>>>>>>>>>> value
>>>>>>>>>>>> error. Where? A visual look at the script revealed a susceptible
>>>>>>>>>>>> culprit: mean. Added require 'stats/base' and it ran.
>>>>>>>>>>>> 
>>>>>>>>>>>> While trying the Trajectory, I ran the script with the debugger. 
>>>>>>>>>>>> [The
>>>>>>>>>>>> debugger did not fail, great!] The value error on mean was not 
>>>>>>>>>>>> obvious.
>>>>>>>>>>>> The stack is given as:
>>>>>>>>>>>> value error
>>>>>>>>>>>> Fold_j_[40] 13!:8]13!:11''[FoldZv_j_=:fzv
>>>>>>>>>>>> trajectory[0] u F: v y
>>>>>>>>>>>> 
>>>>>>>>>>>> This Trajectory example is a simple case. I suspect that errors in 
>>>>>>>>>>>> more
>>>>>>>>>>>> complex cases may not be easy to analyze.
>>>>>>>>>>>> 
>>>>>>>>>>>> From a notational point of view, the Fold family is a great 
>>>>>>>>>>>> addition, I
>>>>>>>>>>>> think.
>>>>>>>>>>>> 
>>>>>>>>>>>>    JVERSION
>>>>>>>>>>>> Engine: j901/j64avx/linux
>>>>>>>>>>>> Beta-m: commercial/2019-10-02T09:15:44
>>>>>>>>>>>> Library: 9.01.10
>>>>>>>>>>>> Qt IDE: 1.8.1/5.5.1
>>>>>>>>>>>> Platform: Linux 64
>>>>>>>>>>>> Installer: J901 install
>>>>>>>>>>>> InstallPath: /home/gilles/j901
>>>>>>>>>>>> Contact: www.jsoftware.com
>>>>>>>>>>>>      ~ Gilles
>>>>>>>>>>>> 
>>>>>>>>>>>> ----------------------------------------------------------------------
>>>>>>>>>>>> For information about J forums see 
>>>>>>>>>>>> http://www.jsoftware.com/forums.htm
>>>>>>>>>> ----------------------------------------------------------------------
>>>>>>>>>> For information about J forums see 
>>>>>>>>>> http://www.jsoftware.com/forums.htm
>>>>>>>>> -- 
>>>>>>>>> This email has been checked for viruses by AVG.
>>>>>>>>> https://www.avg.com
>>>>>>>>> 
>>>>>>>>> ----------------------------------------------------------------------
>>>>>>>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>>>>>>> ----------------------------------------------------------------------
>>>>>>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>>>>>> ----------------------------------------------------------------------
>>>>>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>>>>> ----------------------------------------------------------------------
>>>>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>>>> ----------------------------------------------------------------------
>>>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>>> ----------------------------------------------------------------------
>>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>> ----------------------------------------------------------------------
>>> For information about J forums see http://www.jsoftware.com/forums.htm
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
> 
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to