Re: [sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Michael Orlitzky
On 06/20/2018 04:21 PM, Nils Bruin wrote: > > Hm, if you intend to use that in sage library code, do remember to > change back the state to domain:complex when you're done Better yet, set it back to whatever it was originally. -- You received this message because you are subscribed to the

[sage-devel] Re: sage gets killed on startup on ubuntu

2018-06-20 Thread Dima Pasechnik
Either you updated the system, or you updated the (shell) environment, or your disk/swap is almost full, or your hardware is dying, or worse... On Wednesday, June 20, 2018 at 7:34:22 PM UTC+1, Christian Stump wrote: > > Hi, > > since today, I cannot start sage anymore, not even on a freshly

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Nils Bruin
On Wednesday, June 20, 2018 at 1:45:16 PM UTC-7, Richard_L wrote: > > Is this potentially the same bug as Maxima's #3324? > (to which both Robert and Nils contributed last year) > > Advice please: Should I start a new ticket at Maxima, or add to the old > one? > The causes could well be related,

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Richard_L
Is this potentially the same bug as Maxima's #3324? (to which both Robert and Nils contributed last year) Advice please: Should I start a new ticket at Maxima, or add to the old one? Thanks On Wednesday, June 20, 2018 at 11:57:58 AM UTC-7, Richard_L wrote: > > Thank you. > Now to report the

[sage-devel] Re: sage gets killed on startup on ubuntu

2018-06-20 Thread Christian Stump
> some interesting env var set? not actively, looking through them doesn't show anything that seems sage related. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Nils Bruin
On Wednesday, June 20, 2018 at 12:01:50 PM UTC-7, Eric Gourgoulhon wrote: > > Le mercredi 20 juin 2018 00:22:07 UTC+2, Nils Bruin a écrit : >> >> On Tuesday, June 19, 2018 at 2:30:46 PM UTC-7, Richard_L wrote: >>> >>> It would be useful if one could override Sage's default >>> "domain:complex",

[sage-devel] sage gets killed on startup on ubuntu

2018-06-20 Thread Dima Pasechnik
some interesting env var set? -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Eric Gourgoulhon
Le mercredi 20 juin 2018 00:22:07 UTC+2, Nils Bruin a écrit : > > On Tuesday, June 19, 2018 at 2:30:46 PM UTC-7, Richard_L wrote: >> >> It would be useful if one could override Sage's default "domain:complex", >> at least until the Maxima fix comes through. >> >> One can:

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Richard_L
Thank you. Now to report the bug. On Wednesday, June 20, 2018 at 11:15:36 AM UTC-7, Nils Bruin wrote: > > On Wednesday, June 20, 2018 at 10:12:24 AM UTC-7, Richard_L wrote: >> >> ..."domain:complex" is not the default for which code: Maxima or Sage? >> > > it's a maxima command, so: default for

[sage-devel] sage gets killed on startup on ubuntu

2018-06-20 Thread Christian Stump
Hi, since today, I cannot start sage anymore, not even on a freshly compiled development version (or any other of my older sage versions): Testing that Sage starts... [2018-06-20 17:33:33] SageMath version 8.2, Release Date: 2018-05-05 This looks like the first time you are running Sage.

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Nils Bruin
On Wednesday, June 20, 2018 at 10:12:24 AM UTC-7, Richard_L wrote: > > ..."domain:complex" is not the default for which code: Maxima or Sage? > it's a maxima command, so: default for maxima. Sage uses maxima with non-default settings. > > Sage's documentation [ >

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Richard_L
..."domain:complex" is not the default for which code: Maxima or Sage? Sage's documentation [https://doc.sagemath.org/html/en/reference/calculus/sage/calculus/var.html] states By default, var returns a complex variable. To define real or positive > variables we can specify the domain as:

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Nils Bruin
On Wednesday, June 20, 2018 at 7:15:11 AM UTC-7, Richard_L wrote: > > Thank you! maxima_calculus.eval("domain:real;") solves my immediate > problem. > Of course, the underlying bug remains. > BTW, I had heard years ago to avoid declaring sage variables with > domain='positive', as this often

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Nils Bruin
On Tuesday, June 19, 2018 at 10:35:46 PM UTC-7, Ralf Stephan wrote: > > On Wednesday, June 20, 2018 at 12:22:07 AM UTC+2, Nils Bruin wrote: >> >> Of course, another solution for your computation to succeed is to not put >> the assumptions in. Inequalities will not affect matrix algebra >>

[sage-devel] Re: Re; Symbolic matrix inverse still gives seg-fault, core-dump

2018-06-20 Thread Richard_L
Thank you! maxima_calculus.eval("domain:real;") solves my immediate problem. Of course, the underlying bug remains. BTW, I had heard years ago to avoid declaring sage variables with domain='positive', as this often triggered maxima failures. I wonder if I have stumbled on the same bug. On

Re: [sage-devel] Patchbot trouble with flint/fmpz.h

2018-06-20 Thread John Cremona
On 20 June 2018 at 09:23, Erik Bray wrote: > On Tue, Jun 19, 2018 at 5:13 PM Jeroen Demeyer wrote: > > > > Several patchbots are showing this error: > > > > [eclib-20171002.p0] libtool: compile: g++ -std=gnu++11 > > -DPACKAGE_NAME=\"eclib\" -DPACKAGE_TARNAME=\"eclib\" > >

Re: [sage-devel] Patchbot trouble with flint/fmpz.h

2018-06-20 Thread Erik Bray
On Tue, Jun 19, 2018 at 5:13 PM Jeroen Demeyer wrote: > > Several patchbots are showing this error: > > [eclib-20171002.p0] libtool: compile: g++ -std=gnu++11 > -DPACKAGE_NAME=\"eclib\" -DPACKAGE_TARNAME=\"eclib\" > -DPACKAGE_VERSION=\"20171002\" "-DPACKAGE_STRING=\"eclib 20171002\"" >

Re: [sage-devel] Re: ImportError: cannot import name _remove_dead_weakref

2018-06-20 Thread Erik Bray
On Tue, Jun 19, 2018 at 6:53 PM Anand Shekar wrote: > > This worked, thanks! > > I was trying to do a 'make' after the untar and ran against all those > errors. Am very new to this - so thanks for the help and pointing it out. Concerning the actual error you got, make sure you have a clean

Re: [sage-devel] Re: trac error again

2018-06-20 Thread Erik Bray
On Tue, Jun 19, 2018 at 10:26 PM John H Palmieri wrote: > > I fixed my own problem: I had pushed without committing anything first. After > committing, I can view the ticket. Yes, it happens when you add a branch that has the same HEAD as the branch it's being compared against (i.e. develop).

[sage-devel] Re: segfault sage using Maxima integrate

2018-06-20 Thread 'Nasser M. Abbasi' via sage-devel
Update. I found the cause of the segfault. Posted answer here FYI https://ask.sagemath.org/question/42664/segmentation-fault-core-dumped-sagemath-version-82 So I am all set. No more issues. Case closed. Thank you --Nasser On Tuesday, June 19, 2018 at 9:40:56 PM UTC-5, Nasser M. Abbasi wrote: