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
