You are not logged in.
I'm experiencing problems with the sagemath package. If I just launch `sage` from the terminal it gives this stack trace. The same happens if I try to do a `from sage.all import *` from ipython, from a python file or from sage -python.
I also tried a full removal with `pacman -Rns sagemath`, followed by a reinstall, but the problem is still here.
Does anyone have an idea on how to solve this?
Sage conversation on google groups
https://groups.google.com/g/sage-support/c/38P9AiJ1Xt4
and on ask sagemath
https://ask.sagemath.org/question/55533 … ol-st_new/
Offline
I had this same problem. I was able to work around it by downgrading the lrcalc package from 2.0 to 1.2.
Maybe sage needs to be re-linked against this library
Offline
I had this same problem. I was able to work around it by downgrading the lrcalc package from 2.0 to 1.2.
Maybe sage needs to be re-linked against this library
Both packages supply the same soname version of liblrcalc, liblrcalc.so.1.0.1. The lack of soname bump used to indicate version incompatibility would seem to be an upstream issue.
Offline
Upstream bug is here https://bugs.archlinux.org/task/69500
The proposed solution is the same: downgrade librlcalc.
Offline
tigerjack you do not consider it ABI breaking to remove functions such the following?
git show f2678531d14f6ad3b72a1cbfe574087f004db6f7 -- src/symfcn.h
commit f2678531d14f6ad3b72a1cbfe574087f004db6f7 (HEAD, tag: lrcalc-2.0, origin/master, origin/HEAD, master)
Author: Anders Skovsted Buch <asbuch@math.rutgers.edu>
Date: Sat Jan 30 21:10:04 2021 -0500
Version 2.0
diff --git a/src/symfcn.h b/src/symfcn.h
deleted file mode 100644
index 29bb00d..0000000
--- a/src/symfcn.h
+++ /dev/null
@@ -1,43 +0,0 @@
-#ifndef _SYMFCN_H
-#define _SYMFCN_H
-
-#include "hashtab.h"
-#include "vector.h"
-
-int part_itr_sz(vector *part);
-int part_itr_sub(vector *part, vector *outer);
-int part_itr_between(vector *part, vector *inner, vector *outer);
-
-int part_length(vector *p);
-vector *part_conjugate(vector *p);
-int part_subset(vector *p1, vector *p2);
-
-long long lrcoef(vector *outer, vector *inner1, vector *inner2);
-
-hashtab *skew(vector *outer, vector *inner, int maxrows);
-hashtab *mult(vector *sh1, vector *sh2, int maxrows);
-hashtab *coprod(vector *part, int all);
-
-hashtab *schur_lc_mult(hashtab *lc1, hashtab *lc2, int maxrows);
-
-list *quantum_reduce(hashtab* s, int rows, int cols);
-void fusion_reduce(hashtab *lc, int rows, int cols, int opt_zero);
-
-
-typedef struct {
- vector *outer;
- vector *inner;
- vector *conts;
- int maxrows;
- vector *conjugate;
- int rows;
- int cols;
- int matrix[1];
-} skewtab;
-
-skewtab *st_new(vector *outer, vector *inner, vector *conts, int maxrows);
-int st_next(skewtab *st);
-void st_print(skewtab *st);
-void st_free(skewtab *st);
-
-#endif
Why is Arch considered the upstream and not https://bitbucket.org/asbuch/lrcalc/ ?
Offline
I see how can it be totally bad to change the interface on their side this way. I am not assuming it's arch developer's fault, I'm just trying to find a workaround for this. I mean, I didn't even know about the existence of lrcalc before this bug, I'm just using sage since 2 weeks and came across this error right now; reporting it to both sage and arch community seemed the right way to help and get help.
Offline
Sorry for my over reaction. You are completely right Arch has to fix the issue however it got there.
Offline
Apparently the bug has been fixed.
https://bugs.archlinux.org/task/69500
I still didn't have the chance to test the new update thought, I'll mark the thread solved as soon as I have more info.
Offline
When doing a full system upgrade today, I got the gripe:
:: Starting full system upgrade...
resolving dependencies...
looking for conflicting packages...
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: lrcalc1 and lrcalc are in conflict
It's a bit cryptic; for me, the easy solution was to uninstall sagemath, do the upgrade and re-install sagemath. As noted above, lrcalc version 2.0 introduced an incompatibility with sagemath, and the maintainers chose to make sagemath require the earlier version, now renamed lrcalc1 for the purpose -- hence the conflict.
Offline
Same here PMR21.
Apparently, the resolution had some side effects on general system upgrade.
simple pacman -Syu results in "error lrcalc1 and lrcalc are in conflict".
I would prefer, if possible, not to proceed with complete removal of sagemath.
Bug opened so the dependencies can get fixed (in case this is the real issue) : https://bugs.archlinux.org/task/69539
Quite new at this, feel free to pitch in on the bug report.
Offline
Solution was provided by Anthony Rojas.
pacman -Suy --ignore lrcalc
resolve the conflict in my case.
Apparently the bug was with pacman.
https://bugs.archlinux.org/task/69539
Offline