Hi, two things: (1) *PLEASE* submit a real bug report following the suggestions in the BUGS file in the PDL distribution including the requested 'perldl -V' output and a short, specific way to reproduce the problem along with generated output and what should have happened. Without real reports there is no way to track the status of PDL performance or to determine what to fix.
(2) There is a similar problem with TriD windows being closed by hand which results in perldl dying the next plot attempt. To work around the problem of losing my interactive work which I use as the starting point for program development, I added the following to my perldl work: * Edited the perldl script to change the number of lines saved in history from 50 to 250 * Added a periodic flush of the history every 10 perldl commands so when a crash occurs I don't lose lots of typing that has to be reproduced by hand. In the .perldlrc: push @PERLDL::AUTO, 'flhist() unless ($perldl_hist_flush++ % 10);'; where the flhist() command is just a cut and paste of the existing function in the perldl program which I've attached below. --Chris ----- Original Message ----- From: Kåre Edvardsen <[EMAIL PROTECTED]> Date: Monday, May 7, 2007 10:57 pm Subject: [Perldl] More plplot issues... To: Perl <[email protected]> > There is a very unfortunate thing with plplot and PDL; when making > mistakes which makes PDL barf, I often experience the whole PDL > sessionto terminate with segmentation fault. If I for instance > get a warning > like:
BUGS
Description: Binary data
flhist.pdl
Description: Binary data
_______________________________________________ Perldl mailing list [email protected] http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
