[sage-devel] Re: Error building openblas-0.2.19

You will probably need a variation of the following: https://groups.google.com/d/msg/sage-release/3QJoAgg9bgo/5uvnTLV1AwAJ Martin -- 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, s

[sage-devel] Error building openblas-0.2.19

I just compiled sage-7.5.beta3 on a brand-new computer (also tried 7.4 with the same outcome) and openblas-0.2.19 failed to build with the attached error message (apparently different from previous problems with a missing library): Makefile:123: *** OpenBLAS: Detecting CPU failed. Please set TA

[sage-devel] Re: OpenSSL as a new systemwide Sage dependency ?

On Monday, November 21, 2016 at 5:26:12 PM UTC, Dima Pasechnik wrote: > > > > On Monday, November 21, 2016 at 11:21:31 AM UTC, Emmanuel Charpentier > wrote: >> >> Dear list, >> >> The fact that we can't ship openSSL (see uncountable theads in sage-devel >> and others) seems to pose more and mor

Re: [sage-devel] Re: OpenSSL as a new systemwide Sage dependency ?

On Monday, November 21, 2016 at 7:10:49 PM UTC, Emmanuel Charpentier wrote: > > Le lundi 21 novembre 2016 à 10:17 -0800, Volker Braun a écrit : > > Actually OSX is foobar'ed even then, Apple's ancient openssl just doesn't > support TLSv1.2. Some sites are already requiring that: > > osx:~ vbraun

[sage-devel] Implements change_ring() for BooleanPolynomialRing (#21817) -- Needs Review

Hi, This patch (#21817 ) implements a dedicated change_ring() for BooleanPolynomialRing. Previous change_ring() uses the one implemented in MPolynomialRing_generic as the parent of BooleanPolynomialRing. This makes change_ring() always return a multivariate

Re: [sage-devel] Re: OpenSSL as a new systemwide Sage dependency ?

Le lundi 21 novembre 2016 à 10:17 -0800, Volker Braun a écrit : > Actually OSX is foobar'ed even then, Apple's ancient openssl just > doesn't support TLSv1.2. Some sites are already requiring that: > osx:~ vbraun$ openssl s_client -connect www.kernel.org:443 > CONNECTED(0003) > write:errno=54

Re: [sage-devel] OpenSSL as a new systemwide Sage dependency ?

Le lundi 21 novembre 2016 à 10:15 -0800, Samuel Lelievre a écrit : > Emmanuel Charpentier: > > > Dima Pasechnik: > > > > > I was referring to the fact that > > > https://github.com/sagemath/binary-pkg > > > does not work on OSX 10.12, due to this SSL blues. > > > > > > Unless I misunderstand, we c

[sage-devel] Re: OpenSSL as a new systemwide Sage dependency ?

Actually OSX is foobar'ed even then, Apple's ancient openssl just doesn't support TLSv1.2. Some sites are already requiring that: osx:~ vbraun$ openssl s_client -connect www.kernel.org:443 CONNECTED(0003) write:errno=54 On Monday, November 21, 2016 at 12:21:31 PM UTC+1, Emmanuel Charpentie

Re: [sage-devel] OpenSSL as a new systemwide Sage dependency ?

Emmanuel Charpentier: > Dima Pasechnik: > > > I was referring to the fact that > > https://github.com/sagemath/binary-pkg > > does not work on OSX 10.12, due to this SSL blues. > > > > Unless I misunderstand, we currently aren't able > > to build distributable Sage binaries on OSX 10.12. > > OK. T

Re: [sage-devel] OpenSSL as a new systemwide Sage dependency ?

On Monday, November 21, 2016 at 5:50:29 PM UTC, Emmanuel Charpentier wrote: > > Le lundi 21 novembre 2016 à 09:34 -0800, Dima Pasechnik a écrit : > > > > On Monday, November 21, 2016 at 4:09:41 PM UTC, Thierry > (sage-googlesucks@xxx) wrote: > > Hi, > > On Mon, Nov 21, 2016 at 03:21:31AM -0800,

Re: [sage-devel] OpenSSL as a new systemwide Sage dependency ?

Le lundi 21 novembre 2016 à 09:34 -0800, Dima Pasechnik a écrit : > On Monday, November 21, 2016 at 4:09:41 PM UTC, Thierry (sage-googles > ucks@xxx) wrote: > > Hi, > > > > > > > > On Mon, Nov 21, 2016 at 03:21:31AM -0800, Emmanuel Charpentier > > wrote: > > > > > Dear list, > > > > >  > > >

Re: [sage-devel] OpenSSL as a new systemwide Sage dependency ?

On Monday, November 21, 2016 at 4:09:41 PM UTC, Thierry (sage-googlesucks@xxx) wrote: > > Hi, > > On Mon, Nov 21, 2016 at 03:21:31AM -0800, Emmanuel Charpentier wrote: > > Dear list, > > > > The fact that we can't ship openSSL (see uncountable theads in > sage-devel > > and others) seems t

Re: [sage-devel] Re: OpenSSL as a new systemwide Sage dependency ?

Le 21 nov. 2016 18:26, "Dima Pasechnik" a écrit : > > > > On Monday, November 21, 2016 at 11:21:31 AM UTC, Emmanuel Charpentier wrote: >> >> Dear list, >> >> The fact that we can't ship openSSL (see uncountable theads in sage-devel and others) seems to pose more and more difficulties. See for exam

[sage-devel] Re: OpenSSL as a new systemwide Sage dependency ?

On Monday, November 21, 2016 at 11:21:31 AM UTC, Emmanuel Charpentier wrote: > > Dear list, > > The fact that we can't ship openSSL (see uncountable theads in sage-devel > and others) seems to pose more and more difficulties. See for example this > thread

Re: [sage-devel] OpenSSL as a new systemwide Sage dependency ?

Should we really add OpenSSL as a dependency to fix a few very specific issues? I'm not saying that I'm against the proposal, but we should really weigh the pros against the cons. The main "con" is: adding an extra dependency which might make it even harder for ordinary users to compile Sage f

Re: [sage-devel] OpenSSL as a new systemwide Sage dependency ?

Hi, On Mon, Nov 21, 2016 at 03:21:31AM -0800, Emmanuel Charpentier wrote: > Dear list, > > The fact that we can't ship openSSL (see uncountable theads in sage-devel > and others) seems to pose more and more difficulties. See for example this > thread

Re: [sage-devel] Re: What is the correct way to patch packages that use autotools?

On Thu, Nov 17, 2016 at 5:28 PM, Jean-Pierre Flori wrote: > > > On Thursday, November 17, 2016 at 5:27:15 PM UTC+1, Erik Bray wrote: >> >> Hmm, okay. I am using my system's autoreconf. For the sage autotools >> is that just an optional package I need to install? > > Yes. Okay, well, it turns ou

Re: [sage-devel] Re: What is the correct way to patch packages that use autotools?

On Mon, Nov 21, 2016 at 4:09 PM, Erik Bray wrote: > On Thu, Nov 17, 2016 at 5:28 PM, Jean-Pierre Flori wrote: >> >> >> On Thursday, November 17, 2016 at 5:27:15 PM UTC+1, Erik Bray wrote: >>> >>> Hmm, okay. I am using my system's autoreconf. For the sage autotools >>> is that just an optional p

Re: [sage-devel] Would it be correct to check for a (sufficient) systemwide installation before compiling a package ?

On Mon, Nov 21, 2016 at 2:52 PM, Emmanuel Charpentier wrote: > Le lundi 21 novembre 2016 14:28:11 UTC+1, Jeroen Demeyer a écrit : >> >> On 2016-11-21 12:37, Emmanuel Charpentier wrote: >> > What do you think ? >> >> I think we should simply NOT install the package at all instead of >> faking an in

Re: [sage-devel] OpenSSL as a new systemwide Sage dependency ?

On Mon, Nov 21, 2016 at 12:21 PM, Emmanuel Charpentier wrote: > Dear list, > > The fact that we can't ship openSSL (see uncountable theads in sage-devel > and others) seems to pose more and more difficulties. See for example this > thread on sage-support, and especially Dima's answer, as well as t

Re: [sage-devel] Checking out old branches

On Fri, Nov 18, 2016 at 8:18 AM, William Stein wrote: > > Ideas: > > > What about changing Cython to optionally use sha1 or md5 hashes instead of > timestamps? I've had this thought as well--it would be fairly straightforward to output a list of hashes for all Cython sources and compare those whe

Re: [sage-devel] Would it be correct to check for a (sufficient) systemwide installation before compiling a package ?

Le lundi 21 novembre 2016 14:28:11 UTC+1, Jeroen Demeyer a écrit : > > On 2016-11-21 12:37, Emmanuel Charpentier wrote: > > What do you think ? > > I think we should simply NOT install the package at all instead of > faking an install in spkg-install. We currently do for gcc and git OK so far

Re: [sage-devel] Would it be correct to check for a (sufficient) systemwide installation before compiling a package ?

On 2016-11-21 12:37, Emmanuel Charpentier wrote: What do you think ? I think we should simply NOT install the package at all instead of faking an install in spkg-install. We currently do for gcc and git using some logic in the top-level ./configure. -- You received this message because you

[sage-devel] Would it be correct to check for a (sufficient) systemwide installation before compiling a package ?

Dear list, We currently ship a number of packages that might duplicate the functionality of systemwide installed libraries. My case in point : xz, which provides support for lzma-compressed archives and a command-line tool. What I think is that, in such cases, unless we require a very specific

[sage-devel] OpenSSL as a new systemwide Sage dependency ?

Dear list, The fact that we can't ship openSSL (see uncountable theads in sage-devel and others) seems to pose more and more difficulties. See for example this thread on sage-support, and especially Dima's answer

[sage-devel] Re: A 404 in the Download Source page

I've made this ticket: https://github.com/sagemath/website/issues/97 and yes, the changelog is simply missing because the script generating it is lost. -- h On Mon, Nov 21, 2016 at 9:56 AM, TB wrote: > Dear sagemath.org website admins, > > In http://www.sagemath.org/download-source.html the R

[sage-devel] A 404 in the Download Source page

Dear sagemath.org website admins, In http://www.sagemath.org/download-source.html the README.txt link leads to a 404 page. The correct link might be https://git.sagemath.org/sage.git/tree/README.md Same for the changelog link (but versions prior to 7.4 works). Best regards, TB -- You recei