[sage-devel] submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
{{{
sage: V = span(QQ,[(1,1),(1,0)])
sage: W = span(QQ,[(0,1),(1,0)])
sage: V
Vector space of degree 2 and dimension 2 over Rational Field
Basis matrix:
[1 0]
[0 1]
sage: W
Vector space of degree 2 and dimension 2 over Rational Field
Basis matrix:
[1 0]
[0 1]
sage: V is W
False
sage: V==W
True
sage: type(V)
sage: type(W)
}}}
They are parents. Yet they are not unique.
--
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@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
It is a violation of what!? Where did you read that parents should be
unique?
On 09/10/2017 09:26, Simon Brandhorst wrote:
{{{
sage: V = span(QQ,[(1,1),(1,0)])
sage: W = span(QQ,[(0,1),(1,0)])
sage: V
Vector space of degree 2 and dimension 2 over Rational Field
Basis matrix:
[1 0]
[0 1]
sage: W
Vector space of degree 2 and dimension 2 over Rational Field
Basis matrix:
[1 0]
[0 1]
sage: V is W
False
sage: V==W
True
sage: type(V)
sage: type(W)
}}}
They are parents. Yet they are not unique.
--
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@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
"While *parents* are unique, equal *elements* of a parent in Sage are not
necessarily identical. "
in
http://doc.sagemath.org/html/en/tutorial/tour_coercion.html
OK. In the parent class it says
"...this is useful if parents are unique, or element_constructor is a bound
method (this latter case can be detected automatically). ..."
So uniqueness is not strict. But when is a parent supposed to be unique and
when not?
On Monday, October 9, 2017 at 9:48:29 AM UTC+2, vdelecroix wrote:
>
> It is a violation of what!? Where did you read that parents should be
> unique?
>
> On 09/10/2017 09:26, Simon Brandhorst wrote:
> > {{{
> > sage: V = span(QQ,[(1,1),(1,0)])
> > sage: W = span(QQ,[(0,1),(1,0)])
> > sage: V
> >
> > Vector space of degree 2 and dimension 2 over Rational Field
> > Basis matrix:
> > [1 0]
> > [0 1]
> > sage: W
> >
> > Vector space of degree 2 and dimension 2 over Rational Field
> > Basis matrix:
> > [1 0]
> > [0 1]
> > sage: V is W
> > False
> > sage: V==W
> > True
> > sage: type(V)
> > 'sage.modules.free_module.FreeModule_submodule_field_with_category'>
> > sage: type(W)
> > 'sage.modules.free_module.FreeModule_submodule_field_with_category'>
> > }}}
> >
> > They are parents. Yet they are not unique.
> >
>
--
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@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
[sage-devel] Please help with #23967 Coercion pushout for FGP_modules
Would someone please guide me through this? I would like to learn now to implement such a thing in sage. Yet I wouldn't know where to start here. It is mathematically quite simple. -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] cannot push to trac
On Sat, Oct 7, 2017 at 3:23 PM, 'Martin R' via sage-devel
wrote:
> I was trying to push to #23847, but get an unhelpful error message... please
> help!
>
> martin@Martin-Laptop:~/sage-patchbot$ git trac push
> Pushing to Trac #23847...
> Guessed remote branch:
> u/mantepse/make_the_experimental_fricas_package_optional
> Traceback (most recent call last):
> File "/usr/local/bin/git-trac", line 18, in
> cmdline.launch()
> File "/usr/local/lib/python2.7/dist-packages/git_trac/cmdline.py", line
> 223, in launch
> app.push(ticket_number, remote=args.remote, force=args.force)
> File "/usr/local/lib/python2.7/dist-packages/git_trac/app.py", line 216,
> in push
> self.repo.push(remote, force)
> File "/usr/local/lib/python2.7/dist-packages/git_trac/git_repository.py",
> line 197, in push
> self.git.echo.push('trac', refspec)
> File "/usr/local/lib/python2.7/dist-packages/git_trac/git_interface.py",
> line 341, in meth
> return self.execute(git_cmd, *args, **kwds)
> File "/usr/local/lib/python2.7/dist-packages/git_trac/git_interface.py",
> line 98, in execute
> popen_stderr=subprocess.PIPE)
> File "/usr/local/lib/python2.7/dist-packages/git_trac/git_interface.py",
> line 263, in _run
> raise GitError(result)
> git_trac.git_error.GitError:
>
> martin@Martin-Laptop:~/sage-patchbot$ git remote -v
> origin git://github.com/sagemath/sage.git (fetch)
> origin git://github.com/sagemath/sage.git (push)
> trac git://trac.sagemath.org/sage.git (fetch)
> trac g...@trac.sagemath.org:sage.git (push)
Try doing it 'manually' (i.e. without using "git trac") and see if you
get a more helpful error message.
--
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@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
[sage-devel] switching back to readline-based IPython tab completion? (or live with segfaults etc?)
In view of https://trac.sagemath.org/ticket/22766 (segfaults at certain tab completions), caused by IPython doing multi-threaded tab completion and module importing (via prompt_toolkit), the only way to fix this at the moment is to switch to "rlipython" - a backport to IPython 5 of the old tab completion behaviour. Specifically, see https://trac.sagemath.org/ticket/22766#comment:32 and https://github.com/ipython/ipython/issues/10364/#issuecomment-300829008 One could also re-work ecllib and maxima_lib to get rid of the particular crashes mentioned on https://trac.sagemath.org/ticket/22766 but this does not remove the multi-threaded module importing, which in principle might cause more, hard to track down, crashes and errors. Dima -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
The reason why unique parents were introduced is for speed and memory reasons. Having a million copies of the same parent around is a bit of a waste, but this is something that happens quite easily if one does not make parents unique. I also noticed that submodules violate the "equal => equal hash principle" since: X = QQ^2 V = X.span([[1,0],[0,1]]) W = X.span_of_basis([[1,0],[0,1]]) Are all equal but have distinct hashes. -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] switching back to readline-based IPython tab completion? (or live with segfaults etc?)
Did you report this to IPython? What did they have to say about it? -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
Hi Simon, On 2017-10-09, Simon Brandhorst wrote: > "While *parents* are unique, equal *elements* of a parent in Sage are not > necessarily identical. " > in > http://doc.sagemath.org/html/en/tutorial/tour_coercion.html Indeed that's too strict, IMHO. In http://doc.sagemath.org/html/en/thematic_tutorials/coercion_and_categories.html, I wrote "You are encouraged to make your parent “unique”." and "Making parents unique can be quite important for an efficient implementation, because the repeated creation of “the same” parent would take a lot of time." Moreover, when using non-unique parents, one has to take care to avoid problems with coercion. > So uniqueness is not strict. But when is a parent supposed to be unique and > when not? There were cases where unique parents gave rise to memory leaks: If parents are unique, they need to be stored somewhere. And even when using a weak value dictionary for that (which we of course do), it can be tricky to avoid reference chains that prevent stuff from being garbage collected. As a rule of thumb: If the uniqueness condition is easy, then it is a good idea to have unique parents. But certainly you should avoid that a very costly isomorphism test of some sort is performed whenever you create a parent. Cheers, Simon -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: Please help with #23967 Coercion pushout for FGP_modules
On 2017-10-09, Simon Brandhorst wrote: > Would someone please guide me through this? > I would like to learn now to implement such a thing in sage. Yet I wouldn't > know where to start here. > It is mathematically quite simple. Does http://doc.sagemath.org/html/en/thematic_tutorials/coercion_and_categories.html help? Cheers, Simon -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: switching back to readline-based IPython tab completion? (or live with segfaults etc?)
On Monday, October 9, 2017 at 11:30:11 AM UTC+1, Maarten Derickx wrote: > > Did you report this to IPython? What did they have to say about it? Sure, I asked them. See e.g. https://github.com/ipython/ipython/issues/10364/#issuecomment-334993192 -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
Hi Simon, thank you for your explanation. Given the concrete example. Over exact PIDs/fields one can use echelonized matrices/hermite normal form to get unique Hashes. As far as I can see they are computed anyways at the time of initalization of a submodule. So the upshot is that unless someone notices speed problems when working with submodules, this is not worth to be changed? -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] Re: submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
+1 for correcting the hash of submodules in these cases! If the computation is expensive it can be done at the time __hash__ is called. Note that for "submodules with basis" it is a bit different since you specify a given set of generators. Vincent On 09/10/2017 14:35, Simon Brandhorst wrote: Hi Simon, thank you for your explanation. Given the concrete example. Over exact PIDs/fields one can use echelonized matrices/hermite normal form to get unique Hashes. As far as I can see they are computed anyways at the time of initalization of a submodule. So the upshot is that unless someone notices speed problems when working with submodules, this is not worth to be changed? -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: Please help with #23967 Coercion pushout for FGP_modules
Thank you Simon :). Probably, it will take me some time to read/understand though. Would you review the ticket when it is done? On Monday, October 9, 2017 at 12:45:17 PM UTC+2, Simon King wrote: > > On 2017-10-09, Simon Brandhorst > wrote: > > Would someone please guide me through this? > > I would like to learn now to implement such a thing in sage. Yet I > wouldn't > > know where to start here. > > It is mathematically quite simple. > > Does > http://doc.sagemath.org/html/en/thematic_tutorials/coercion_and_categories.html > > help? > > Cheers, > Simon > > -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
A case where this may be relevant: A,B,C,D submodules of the same ambient vector space, e.g. QQ^n A/B and C/D have a good pushout (A+C)/(B+D) this is where the sum of two elements (a+B) + (c +D) lives. So I have to teach the coercion framework how to create (A + C) / (B + D). Now if this is not unique I might end up creating a xillion parents just by adding elements. Let's see how it goes. If it happens, I will post it here. -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: submodules subvectorspaces violate the unique parent condition. Why? Should we change that?
On Monday, October 9, 2017 at 12:40:50 PM UTC+2, Simon King wrote: > > I wrote "You are encouraged to make your parent “unique”." > and > "Making parents unique can be quite important for an efficient > implementation, >because the repeated creation of “the same” parent would take a lot of > time." > Moreover, when using non-unique parents, one has to take care to avoid > problems with coercion. > > It is also important to keep in mind (and this is a good design principle) that if it's possible to just pass the required parent around, there's obviously no reason to recreate it, regardless of whether it's "unique". The cost of "uniqueness" is significant: it costs us a global cache, which is very sensitive to memory leaks, and it adds a significant cost to creation, because to construction parameters need to be normalized. It is MUCH cheaper to just pass the correct parent along if you already have your hands on it than to have to recreate the construction parameters and look it up in the global cache. The reason why the coercion framework needs globally unique parents is that it sometimes will construct a new parent for the result to live in, just based on elements living in other parents. THAT's the place where globally unique parents are quite necessary. It also helps the coercion framework in tracing more complicated coercion paths (because comparison for globally unique parents is very cheap). Modules are actually a problematic example: There are scenarios where lots of non-equal modules will be constructed. "Unique" stuff would give a very unwelcome overhead there. On the other hand, module elements could be involved in relatively complicated coercions, where uniqueness could help a bit. Since the coercion situations can be worked around by explicitly constructing the required morphisms, I would lean towards not going too far in putting "unique" stuff in, because it would inflict an unavoidable cost. -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] sage downloads
On Thu, Sep 28, 2017 at 2:58 PM, kcrisman wrote: > 1) At http://www.sagemath.org/download-windows.html it still only has the > virtual machine solution. Should Erik Bray's solution be there as well? This has since been updated! :) -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: Please help with #23967 Coercion pushout for FGP_modules
It helps a bit. Though my problem right now is that fgp_modules do not have a construction defined in the tutorial this is the case. So I have to give fgp_modules a construction functor? Or is there an easier way to get a pushout? On Monday, October 9, 2017 at 12:45:17 PM UTC+2, Simon King wrote: > > On 2017-10-09, Simon Brandhorst > wrote: > > Would someone please guide me through this? > > I would like to learn now to implement such a thing in sage. Yet I > wouldn't > > know where to start here. > > It is mathematically quite simple. > > Does > http://doc.sagemath.org/html/en/thematic_tutorials/coercion_and_categories.html > > help? > > Cheers, > Simon > > -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] Re: Please help with #23967 Coercion pushout for FGP_modules
On Mon, Oct 9, 2017 at 1:38 PM, Simon Brandhorst wrote: > It helps a bit. Though my problem right now is that fgp_modules do not > have a construction defined in the tutorial this is the case. So I have to > give fgp_modules a construction functor? Or is there an easier way to get a > pushout? > Nope, that's how you get pushouts. You can take a look at the QuotientFunctor in sage/categories/pushout.py for inspiration. David > On Monday, October 9, 2017 at 12:45:17 PM UTC+2, Simon King wrote: >> >> On 2017-10-09, Simon Brandhorst wrote: >> > Would someone please guide me through this? >> > I would like to learn now to implement such a thing in sage. Yet I >> wouldn't >> > know where to start here. >> > It is mathematically quite simple. >> >> Does http://doc.sagemath.org/html/en/thematic_tutorials/coercion_ >> and_categories.html help? >> >> Cheers, >> Simon >> >> -- > 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@googlegroups.com. > Visit this group at https://groups.google.com/group/sage-devel. > For more options, visit https://groups.google.com/d/optout. > -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] Re: Please help with #23967 Coercion pushout for FGP_modules
I see. I have found an easy solution by defining a method ._pushout_ It works. I have uploaded the branch so you can take a look. On Monday, October 9, 2017 at 7:55:13 PM UTC+2, David Roe wrote: > > > > On Mon, Oct 9, 2017 at 1:38 PM, Simon Brandhorst > wrote: > >> It helps a bit. Though my problem right now is that fgp_modules do not >> have a construction defined in the tutorial this is the case. So I have to >> give fgp_modules a construction functor? Or is there an easier way to get a >> pushout? >> > > Nope, that's how you get pushouts. You can take a look at the > QuotientFunctor in sage/categories/pushout.py for inspiration. > David > > > >> On Monday, October 9, 2017 at 12:45:17 PM UTC+2, Simon King wrote: >>> >>> On 2017-10-09, Simon Brandhorst wrote: >>> > Would someone please guide me through this? >>> > I would like to learn now to implement such a thing in sage. Yet I >>> wouldn't >>> > know where to start here. >>> > It is mathematically quite simple. >>> >>> Does >>> http://doc.sagemath.org/html/en/thematic_tutorials/coercion_and_categories.html >>> >>> help? >>> >>> Cheers, >>> Simon >>> >>> -- >> 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+...@googlegroups.com . >> To post to this group, send email to sage-...@googlegroups.com >> . >> Visit this group at https://groups.google.com/group/sage-devel. >> For more options, visit https://groups.google.com/d/optout. >> > > -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] recipe for target 'ecl-16.1.2.p4' fail
What version of ffi (possibly libffi) is one the system? François > On 10/10/2017, at 07:57, Johannes Martin wrote: > > I tried to install sage-8.0 from source on opensuse linux 42.3 > following the standard procedure > configure > make > > Any help will be appreciated, > Johannes 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, send an > email to sage-devel+unsubscr...@googlegroups.com. > To post to this group, send email to sage-devel@googlegroups.com. > Visit this group at https://groups.google.com/group/sage-devel. > For more options, visit https://groups.google.com/d/optout. > -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Many "optional - dot2tex graphviz" doctests pass without those packages
While working on some dot2tex/graphviz oddities, I noticed something strange: many of the doctests marked "optional - dot2tex graphviz" actually pass even when neither dot2tex nor graphviz is installed. There are 28 doctests which have "# optional - dot2tex" or "# optional - dot2tex graphviz" (and which are not marked "not tested"). Of those 28 doctests, only 17 fail when the packages are not installed. Since I don't really know those packages, I wonder what the reason is: (A) Somebody added "optional - dot2tex graphviz" by mistake, possibly copying/pasting it. (B) The doctest doesn't actually test what the author of the doctest was expecting: the intention was to write a doctest depending on dot2tex/graphviz but this got messed up. (C) Some Sage code doesn't properly raise an error when dot2tex/graphviz is not installed. -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] Many "optional - dot2tex graphviz" doctests pass without those packages
I forgot to say that I created https://trac.sagemath.org/ticket/24001 for this. On 2017-10-09 21:15, Jeroen Demeyer wrote: While working on some dot2tex/graphviz oddities, I noticed something strange: many of the doctests marked "optional - dot2tex graphviz" actually pass even when neither dot2tex nor graphviz is installed. -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] recipe for target 'ecl-16.1.2.p4' fail
libffi4 5.3.1+r233831-10.1 Am Montag, 9. Oktober 2017 21:09:54 UTC+2 schrieb François Bissey: > > What version of ffi (possibly libffi) is one the system? > > François > > > On 10/10/2017, at 07:57, Johannes Martin > wrote: > > > > I tried to install sage-8.0 from source on opensuse linux 42.3 > > following the standard procedure > > configure > > make > > > > Any help will be appreciated, > > Johannes 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, send > an email to sage-devel+...@googlegroups.com . > > To post to this group, send email to sage-...@googlegroups.com > . > > Visit this group at https://groups.google.com/group/sage-devel. > > For more options, visit https://groups.google.com/d/optout. > > > > -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] recipe for target 'ecl-16.1.2.p4' fail
Strange version number. Can you install something called just libffi version 3.x? > On 10/10/2017, at 08:58, Johannes Martin wrote: > > libffi4 5.3.1+r233831-10.1 > > Am Montag, 9. Oktober 2017 21:09:54 UTC+2 schrieb François Bissey: > What version of ffi (possibly libffi) is one the system? > > François > > > On 10/10/2017, at 07:57, Johannes Martin wrote: > > > > I tried to install sage-8.0 from source on opensuse linux 42.3 > > following the standard procedure > > configure > > make > > > > Any help will be appreciated, > > Johannes 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, send an > > email to sage-devel+...@googlegroups.com. > > To post to this group, send email to sage-...@googlegroups.com. > > Visit this group at https://groups.google.com/group/sage-devel. > > For more options, visit https://groups.google.com/d/optout. > > > > > -- > 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@googlegroups.com. > Visit this group at https://groups.google.com/group/sage-devel. > For more options, visit https://groups.google.com/d/optout. -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] recipe for target 'ecl-16.1.2.p4' fail
There are no older versions offered for downgrade or newer versions for update. It is the regular version installed by the operating system. Am Montag, 9. Oktober 2017 22:14:17 UTC+2 schrieb François Bissey: > > Strange version number. Can you install something called just libffi > version 3.x? > > > On 10/10/2017, at 08:58, Johannes Martin > wrote: > > > > libffi4 5.3.1+r233831-10.1 > > > > Am Montag, 9. Oktober 2017 21:09:54 UTC+2 schrieb François Bissey: > > What version of ffi (possibly libffi) is one the system? > > > > François > > > > > On 10/10/2017, at 07:57, Johannes Martin wrote: > > > > > > I tried to install sage-8.0 from source on opensuse linux 42.3 > > > following the standard procedure > > > configure > > > make > > > > > > Any help will be appreciated, > > > Johannes 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, send > an email to sage-devel+...@googlegroups.com. > > > To post to this group, send email to sage-...@googlegroups.com. > > > Visit this group at https://groups.google.com/group/sage-devel. > > > For more options, visit https://groups.google.com/d/optout. > > > > > > > > > -- > > 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+...@googlegroups.com . > > To post to this group, send email to sage-...@googlegroups.com > . > > Visit this group at https://groups.google.com/group/sage-devel. > > For more options, visit https://groups.google.com/d/optout. > > -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] recipe for target 'ecl-16.1.2.p4' fail
I am not sure where you get that libffi version from - upstream and upstream on their github repo haven’t cut a release for a while. However on the basis that your version could be based on a fork or snapshot of upstream, I see that FFI_SYSV is defined in headers that are now completely internal and specifically not distributed. So I’d say this a ecl bug where they access something they shouldn’t. I don’t know if it is fixed in a later upstream release at this stage. But upgrading ecl is on hold for a variety of reasons (unsolved breakages). François > On 10/10/2017, at 10:00, Johannes Martin wrote: > > There are no older versions offered for downgrade or newer versions for > update. > It is the regular version installed by the operating system. > > Am Montag, 9. Oktober 2017 22:14:17 UTC+2 schrieb François Bissey: > Strange version number. Can you install something called just libffi version > 3.x? > > > On 10/10/2017, at 08:58, Johannes Martin wrote: > > > > libffi4 5.3.1+r233831-10.1 > > > > Am Montag, 9. Oktober 2017 21:09:54 UTC+2 schrieb François Bissey: > > What version of ffi (possibly libffi) is one the system? > > > > François > > > > > On 10/10/2017, at 07:57, Johannes Martin wrote: > > > > > > I tried to install sage-8.0 from source on opensuse linux 42.3 > > > following the standard procedure > > > configure > > > make > > > > > > Any help will be appreciated, > > > Johannes 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, send an > > > email to sage-devel+...@googlegroups.com. > > > To post to this group, send email to sage-...@googlegroups.com. > > > Visit this group at https://groups.google.com/group/sage-devel. > > > For more options, visit https://groups.google.com/d/optout. > > > > > > > > > -- > > 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+...@googlegroups.com. > > To post to this group, send email to sage-...@googlegroups.com. > > Visit this group at https://groups.google.com/group/sage-devel. > > For more options, visit https://groups.google.com/d/optout. > > > -- > 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@googlegroups.com. > Visit this group at https://groups.google.com/group/sage-devel. > For more options, visit https://groups.google.com/d/optout. -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] recipe for target 'ecl-16.1.2.p4' fail
Surely libffi4 is something different. How about this: https://software.opensuse.org/package/libffi https://download.opensuse.org/repositories/devel:/libraries:/c_c++/openSUSE_Leap_42.3/x86_64/libffi-devel-3.2.1.git259-11.1.x86_64.rpm On Monday, October 9, 2017 at 10:00:47 PM UTC+1, Johannes Martin wrote: > > There are no older versions offered for downgrade or newer versions for > update. > It is the regular version installed by the operating system. > > Am Montag, 9. Oktober 2017 22:14:17 UTC+2 schrieb François Bissey: >> >> Strange version number. Can you install something called just libffi >> version 3.x? >> >> > On 10/10/2017, at 08:58, Johannes Martin wrote: >> > >> > libffi4 5.3.1+r233831-10.1 >> > >> > Am Montag, 9. Oktober 2017 21:09:54 UTC+2 schrieb François Bissey: >> > What version of ffi (possibly libffi) is one the system? >> > >> > François >> > >> > > On 10/10/2017, at 07:57, Johannes Martin >> wrote: >> > > >> > > I tried to install sage-8.0 from source on opensuse linux 42.3 >> > > following the standard procedure >> > > configure >> > > make >> > > >> > > Any help will be appreciated, >> > > Johannes 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, >> send an email to sage-devel+...@googlegroups.com. >> > > To post to this group, send email to sage-...@googlegroups.com. >> > > Visit this group at https://groups.google.com/group/sage-devel. >> > > For more options, visit https://groups.google.com/d/optout. >> > > >> > >> > >> > -- >> > 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+...@googlegroups.com. >> > To post to this group, send email to sage-...@googlegroups.com. >> > Visit this group at https://groups.google.com/group/sage-devel. >> > For more options, visit https://groups.google.com/d/optout. >> >> -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: recipe for target 'ecl-16.1.2.p4' fail
Offhand this could be https://trac.sagemath.org/ticket/21811 and Johannes could try to install with SAGE_GCC_INSTALL=yes Regards, -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: problem with abs on integration
This is https://trac.sagemath.org/ticket/13773 Sage cannot translate Maxima's "if" construct. However, since beta7 there is the cases function in Sage which can now represent such condition/expression pairs, so just an interface is needed. The explanation for the difference between x+2 and x-2 is probably that if you (Maxima) know x is positive then you know x+2 is so as well and you don't need cases. Regards, -- 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@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
