Re: [sage-devel] Should __hash__ be constant across Sage invocations?

On Friday, October 7, 2016 at 7:17:40 AM UTC-5, Johan S. R. Nielsen wrote: > > Jeroen Demeyer writes: > > $ python3 -c "print(hash('hello'))" > > -74281615791272169 > > $ python3 -c "print(hash('hello'))" > > 447613799834866972 > > $ python3 -c "print(hash('hello'))" > >

[sage-devel] Re: Failed source build

It looks like GCC 4.9.3 won't build with GCC 6.2 (probably due to an increase in compiler strictness or something). The reason Sage is building its own copy of GCC that is you need to have the same versions for gcc, gfortran, and g++. Partial builds with different compilers seems like

[sage-devel] error while building Sage 7.3 from source while installing numpy-1.11.1

I get the following error when it is trying to install numpy ImportError: ~/sage-7.3/local/lib/liblapack.so.3: undefined symbol: zunmqr_ my system is CentOS7 x86_64 -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group

Re: [sage-devel] Upstream links for purely sage package

Hi, On 07/10/2016 12:19, Jeroen Demeyer wrote: On 2016-10-06 12:58, 'Julien Puydt' via sage-devel wrote: For those sage-is-upstream packages, I was pointing the helper to http://files.sagemath.org/spkg/upstream/ but now it looks like those are not available anymore on the sagemath.org

Re: [sage-devel] Re: Finishing spkg upgrade

On 2016-10-07 15:04, Simon King wrote: Nice! I see many checkmarks there. So, is stuff ready to use already It's very much work in progress... We have only really just started working on this. -- You received this message because you are subscribed to the Google Groups "sage-devel" group.

[sage-devel] Re: Finishing spkg upgrade

Hi Jeroen, On 2016-10-07, Jeroen Demeyer wrote: > On 2016-10-07 13:59, Simon King wrote: >> That's why I ask here how to do it right... > > Well, that's partially why we have > https://github.com/OpenDreamKit/OpenDreamKit/issues/87 Nice! I see many checkmarks there. So,

Re: [sage-devel] Should __hash__ be constant across Sage invocations?

I've created #21661 reporting that @func_persist is currently broken for this reason. Best, Johan -- 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

Re: [sage-devel] Should __hash__ be constant across Sage invocations?

Jeroen Demeyer writes: > $ python3 -c "print(hash('hello'))" > -74281615791272169 > $ python3 -c "print(hash('hello'))" > 447613799834866972 > $ python3 -c "print(hash('hello'))" > 5466708731786648082 > > I think the only logical conclusion is that you should *not* rely on > comparing hashes

Re: [sage-devel] Re: Finishing spkg upgrade

On 2016-10-07 13:59, Simon King wrote: That's why I ask here how to do it right... Well, that's partially why we have https://github.com/OpenDreamKit/OpenDreamKit/issues/87 -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from

Re: [sage-devel] Should __hash__ be constant across Sage invocations?

On 2016-10-07 14:02, Johan S. H. Rosenkilde wrote: Would it be extremely difficult to attempt to satisfy this? Python 3 would certainly make it a lot harder. Compare: $ python2 -c "print(hash('hello'))" 840651671246116861 $ python2 -c "print(hash('hello'))" 840651671246116861 $ python2 -c

Re: [sage-devel] Should __hash__ be constant across Sage invocations?

> On 2016-10-07 13:35, Johan S. H. Rosenkilde wrote: >> Should __hash__ be constant across Sage invocations? > > No. I see no reason why it should be. I presume there's nothing in the Python spec that says so, since object() doesn't satisfy this. Is that what you mean? To me it seems obviously

[sage-devel] Re: Finishing spkg upgrade

Hi Jeroen, On 2016-10-07, Jeroen Demeyer wrote: > If you want to use autodoc with Cython code, you will probably need > parts of the Sage docbuilder (and I don't think that ripping this out of > Sage has even been done). Well, I did rip it out of sage, in my old-style

[sage-devel] Re: Should __hash__ be constant across Sage invocations?

Hi! On 2016-10-07, Jeroen Demeyer wrote: > On 2016-10-07 13:35, Johan S. H. Rosenkilde wrote: >> The implementation of __hash__ on finite fields claims to be the same as >> for 'object'. > > Could be: > > $ sage -c 'print(hash(object()))' > 8790924175895 > $ sage -c

Re: [sage-devel] Should __hash__ be constant across Sage invocations?

On 2016-10-07 13:35, Johan S. H. Rosenkilde wrote: The implementation of __hash__ on finite fields claims to be the same as for 'object'. Could be: $ sage -c 'print(hash(object()))' 8790924175895 $ sage -c 'print(hash(object()))' 8771665328407 -- You received this message because you are

Re: [sage-devel] Should __hash__ be constant across Sage invocations?

On 2016-10-07 13:35, Johan S. H. Rosenkilde wrote: Should __hash__ be constant across Sage invocations? No. I see no reason why it should be. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving

Re: [sage-devel] patchbot Internal Server Error

Le vendredi 7 octobre 2016 13:29:55 UTC+2, Thierry (sage-googlesucks@xxx) a écrit : > > Hi, > > the log acknowledged an error, but it was very unclear, i relaunched the > services and it seems to work now. > > Thank you Thierry! Eric. -- You received this message because you are

[sage-devel] Should __hash__ be constant across Sage invocations?

I just got bit by the following behaviour, which is surprising to me: sage: hash(GF(17)) -9223363248217694901 sage: hash(GF(17)) -9223363251803121333 So the hash of the very-much-the-same-every-time object GF(17) is not constant across invocations of Sage. Is this expected behaviour? Is this

Re: [sage-devel] patchbot Internal Server Error

Hi, the log acknowledged an error, but it was very unclear, i relaunched the services and it seems to work now. Ciao, Thierry On Fri, Oct 07, 2016 at 12:51:11AM -0700, Eric Gourgoulhon wrote: > Hi, > > Since yesterday, whenever I try to read some patchbot report about a > ticket, I'm getting

[sage-devel] Failed source build

- I'm using Debian 8 (jessie) - have instsalled GCC 6.2 - git cloned the source from today and just ran make (after installing some prerequisites). - After some minutes I arrived at this issue: [gcc-4.9.3.p1] In file included from ../../src/gcc/cp/except.c:1013:0:

[sage-devel] Re: Finishing spkg upgrade

1. sphinx Makefile To create the Makefile, I used sphinx-quickstart inside of sage -sh which ask many questions including this one: """ A Makefile and a Windows command file can be generated for you so that you only have to run e.g. `make html' instead of invoking sphinx-build directly. >

Re: [sage-devel] Finishing spkg upgrade

On 2016-10-07 12:35, Simon King wrote: Also, I'd like to build documentation. Sébastien has pointed me to https://github.com/seblabbe/slabbe as an example of code that uses Sage's doc formatting, and people here have pointed me to google docs. However, my question is: Is the Makefile that I find

[sage-devel] gamma(QQbar(...))

I was surprised by the following behaviour: sage: gamma(QQbar(sqrt(2))) 0.886581428719259 sage: gamma(QQbar(sqrt(2))).parent() Complex Field with 53 bits of precision (I would have preferred to have some symbolic answer or at least an answer in CIF, but not in CC) Furthermore: sage:

[sage-devel] Finishing spkg upgrade

Hi! At https://trac.sagemath.org/ticket/18514 I am in the process of upgrading my old-style group cohomology spkg to a more modularised new-style spkg. I've put it on "needs info", and perhaps some of you can give me the information that I need to finish the upgrade. The branch at #18514 is

Re: [sage-devel] Upstream links for purely sage package

On 2016-10-06 12:58, 'Julien Puydt' via sage-devel wrote: For those sage-is-upstream packages, I was pointing the helper to http://files.sagemath.org/spkg/upstream/ but now it looks like those are not available anymore on the sagemath.org server. I believe that this is a temporary outage

Re: [sage-devel] Notebook stuck on "N?" with a worksheet

On Fri, 7 Oct 2016, Jori Mäntysalo wrote: Could someone try SageNB with http://www.sis.uta.fi/~jm58660/Broken.sws More about this: the computation works, but getting output back to the browser window does not work. So if I reload the page, it seems to be OK. But with other worksheets

[sage-devel] Notebook stuck on "N?" with a worksheet

Could someone try SageNB with http://www.sis.uta.fi/~jm58660/Broken.sws For me that stucks when I try to evaluate the first calculation cell, the one with content #Katso komennon N helppi, erityisesti se, miten desimaalien määrää voidaan säädellä N? It is somehow related to some phrases or

[sage-devel] Upstream links for purely sage package

Hi, I'm the maintainer of a few Debian packages, for which sage is considered upstream: rubiks sagemath-database-mutually-combinatorial-designs sagemath-database-conway-polynomials sagemath-database-elliptic-curves sagemath-database-graphs sagemath-database-polytopes To

[sage-devel] patchbot Internal Server Error

Hi, Since yesterday, whenever I try to read some patchbot report about a ticket, I'm getting the message Internal Server Error from https://patchbot.sagemath.org/ Is it only me ? Eric. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To