Summary: | dev-lisp/clisp with USE=pari with sci-mathematics/pari-2.5.0 - pari.c:136:37: error: ‘gpi’ undeclared (first use in this function) | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Andrey Grozin <grozin> |
Component: | Current packages | Assignee: | Common Lisp Bugs <common-lisp> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | bircoph, frp.bissey, jstein, sci-mathematics |
Priority: | Normal | Keywords: | NeedPatch |
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Andrey Grozin
2011-07-28 03:34:48 UTC
The same happens for dev-lisp/clisp-2.48-r2 in the main tree. So, I removed [lisp-overlay]. Some changes in the header files we may need to include some more if memory serves me right. I am having a closer look, it would have nice to mention the bug numbers when you sent that email over. That will be more work than I expected. This a bit all over the place, for example we have allocatemoremem which has been replaced by allocatemem (see the summary of http://trac.sagemath.org/sage_trac/ticket/11321). Quite a few of these were alias or declared external. I suspect a number of these shouldn't have been called directly in the first place. Should we mask the USE flag pari? Masking will be an overkill, just add <sci-mathematics/pari-2.5.0 requirement. On some systems (e.g. mine) only clisp uses this package. Upstream bug: http://sourceforge.net/p/clisp/bugs/659/ (In reply to Andrew Savchenko from comment #5) > Masking will be an overkill, just add <sci-mathematics/pari-2.5.0 > requirement. That has been done, thanks, clisp-2.49-r8.ebuild has: pari? ( <sci-mathematics/pari-2.5.0 ) =dev-lisp/clisp-2.49-r8 is out of the tree. Current =dev-lisp/clisp-2.49.90 contains pari? ( <sci-mathematics/pari-2.5.0 ), closing |