, but #23066 is closed in 8.0 and has title "sagenb
update to 1.0". It upgrades sagenb from 0.13 to 1.0.1.
Uh, too much code to check.
--
Jori Mäntysalo
t versions you have.
$ ./sage --pip list | egrep -i 'sage|flask'
Flask (0.10.1)
Flask-AutoIndex (0.6)
Flask-Babel (0.9)
Flask-OldSessions (0.10)
Flask-OpenID (1.2.5)
Flask-Silk (0.2)
sage (8.1b1)
sagenb (1.0.1)
sagenb-export (3.2)
sagetex (3.0)
--
Jori Mäntysalo
On Thu, 24 Aug 2017, Dima Pasechnik wrote:
although it very well might be due to its dependencies, like flask,
etc., updated to newer versions, e.g. for the purpose of python3
compatibility.
OK. What packages or parts have been updated lately?
--
Jori Mäntysalo
://stackoverflow.com/questions/9023488/build-error-with-variables-and-url-for-in-flask
but still can not see what bit I should flip.
--
Jori Mäntysalo
we know that product of eigenvalues
equals to determinant etc, product of invertible matrices is invertible
etc. What else could be done?
--
Jori Mäntysalo
On Tue, 22 Aug 2017, kcrisman wrote:
What happens if you ignore that and directly do sage --notebook=sagenb?
Tested, no help. The same internal error, but the publishing and
unpublishing works.
--
Jori Mäntysalo
I have been doing some quality check for finite lattice code. This is a
work in progress, but already can give some ideas for others, so I make it
public for few weeks:
https://sage.sis.uta.fi/home/pub/153/
--
Jori Mäntysalo
t first modify --notebook=sagenb to this form.
--
Jori Mäntysalo
-uploading-notebooks-in-sage/
and/or
https://ask.sagemath.org/question/38468/internal-server-error-after-deleting-or-suspending-a-user/
Yes, there seems to be the same error type "Could not build url for
endpoint - -". But I did not see how to correct this on those pages.
--
Jori Mäntysalo
ner mode" for a
program is and idea that has been abandoned before. I think we do not have
resources to add another level of metadata.
--
Jori Mäntysalo
d you mean 'worksheet.worksheet_publish'
instead?
I have no clue about how to debug this.
--
Jori Mäntysalo
s set of neighbors and a polynomial has
a set of roots, yet we return a list of neighbors and a list of roots. I
guess that the user can figure out what happened if an example says
sage: my_func(inputs)
[1, 2]
and he/she got
[2, 1]
--
Jori Mäntysalo
at startup? I.e. what poset (or lattice) is in the
global namespace?
--
Jori Mäntysalo
ticket closed, you can think if you now have got something ready for
highlights.
--
Jori Mäntysalo
r viewpoint: It
may be that a backend function for version 8.1 takes much work and a
interface function for version 8.2 is quite trivial; then the developer
should say nothing in the highlights of 8.1.
--
Jori Mäntysalo
that are closed when I am NOT the
reporter and author. ...which is kind of funny.
--
Jori Mäntysalo
if a lattice is regular, uniform or isoform.
But in reality part of that is already on 7.6.
--
Jori Mäntysalo
lse is
welcome!
I think 1 would be most natural.
--
Jori Mäntysalo
, and every
lattice ever created in a session is saved in memory.
--
Jori Mäntysalo
is
essential)
+1 from me too. Something printed (a, b, c) might actually be something
more complicated that a plain tuple.
--
Jori Mäntysalo
On Mon, 22 May 2017, Jeroen Demeyer wrote:
On 2017-05-19 14:16, Jori Mäntysalo wrote:
The only thing I want to see in the OUTPUT block is the *type* of the
objects and how many of them.
Why that? What's wrong with being verbose in the OUTPUT block?
This is a misquote. I support something
to comment this. We can have
longer INPUT and shorter OUTPUT, or vice versa.
--
Jori Mäntysalo
.
- If ``certificate=False`` return either (True, None) or (False, XX),
where XX is...
--
Jori Mäntysalo
into the developer manual.
Here is my opinion. But a metaquestion: Why can't someone (like Kwankyu)
just got direct emails and make a summary?
It is of course different when I have untold arguments to tell, not just
+1 or -1.
--
Jori Mäntysalo
-support.
And if that fails too, we will give an official dollar coin to William. He
will then toss a coin to make an arbitrary decision.
(By guideline I do not mean a rule.)
--
Jori Mäntysalo
INPUT? In some more complicated situations, mostly constructors, we
also could have examples inside input blocks.
--
Jori Mäntysalo
On Thu, 18 May 2017, Vincent Delecroix wrote:
Yep, seems to be nicer. But maybe we should first see if we want to have
"test if..." -structure in function docstrings.
How is this different from the poll G3 where everybody seems to agree - -
I think that the poll is still open
should first see if we want to have
"test if..." -structure in function docstrings.
--
Jori Mäntysalo
nformation, as
is_sectionally_complemented() -> Return True if the lattice is sectionally
complemented
sounds informationless explanation. What others think about this?
(Of course this only works when there a short explanation is possible;
more often it is not.)
--
Jori Mäntysalo
no guideline at all, we will end
having
add_vertex() -> Add a vertex to this graph.
add_edge() -> Add an edge to ``self``.
or something similar. We could at least reduce that.
--
Jori Mäntysalo
ys returns a pair (i.e. a tuple of two
elements) where first value is True or False, and the second value is the
certificate or None.
But that won't make the question about docstring easier. What to do if
INPUT and OUTPUT are strongly related? See for example
automorphism_group() of a graph.
Having lists of weaker and stronger properties is my idea. It can be
discussed too.
13) When should we use the format LatticePoset({1: [...]...}) in the
examples, and when LatticePoset(DiGraph('GTE3_#H'))?
14) What are good examples anyway?
At least these questions can be discussed, and this was only for a very
small and clearly defined function.
--
Jori Mäntysalo
quite massive output
of failing tests...
What kind of testing we are talking now? As an example we can create a
random graph g and check that g.is_planar() == (g.genus() == 0) etc.
Should we have that kind of randomized test? (I do those, but have not
included them to Sage code.)
--
Jori
looking for example
is_isoform() at finite lattices. Could we have an exact format for those?
--
Jori Mäntysalo
have to represent the exact Sage/Python types." with an example, and I
have added "Start with assumptions of the object, if there are any." etc.
And about that: what should be the place of SEEALSO-block? Or in general,
order of blocks?
--
Jori Mäntysalo
ES::
sage: pass # Actually tested in foobar()
? Then we could automatically check the code and see that every function
has the examples-block.
--
Jori Mäntysalo
that someday there could be more structured processing
of help, maybe foldable HTML help or something like that.
--
Jori Mäntysalo
On Mon, 8 May 2017, Dima Pasechnik wrote:
Works for me. A particular link you have problem there?
Seems to work again. At least these had problems:
https://trac.sagemath.org/ticket/22773
https://trac.sagemath.org/ticket/22674
--
Jori Mäntysalo
Trac gives the Internal server error, and even that after waiting a
minute.
--
Jori Mäntysalo
Not directly related, but I got notifications with one exception: closing
tickets. Would be nice to have those too.
--
Jori Mäntysalo
Having more structured docstring would of course be a good thing. Then we
could, as an example, hide the AUTHORS-section if wanted etc.
--
Jori Mäntysalo
On Tue, 21 Mar 2017, David Roe wrote:
currently not stored anywhere. The place you'll need to change is line 566
of sage/doctest/parsing.py, where you'll need to do something like
Thanks for pointing right files. I opened #22661 for this and also put in
some example code.
--
Jori
Is it possible to modify doctesting framework to have different time for
tests marked with # long time? Something like sage --warn-long 1,10 could
give warning for only doctests taking more than 10 seconds if they are
explicitly marked as slow.
--
Jori Mäntysalo
"hidden", and maybe some larger code snippets
available behind a click and so on.
--
Jori Mäntysalo
/ticket/18545 , but now I am unsure
after a comment from Kevin.
--
Jori Mäntysalo
On Mon, 13 Feb 2017, Daniel Krenn wrote:
+1 for crosslinks of any kind
An example: https://trac.sagemath.org/ticket/22377
--
Jori Mäntysalo
vers or are
covered by the Boolean property in a poset of property implications.
--
Jori Mäntysalo
am
unsure aftera a comment from Kevin.
--
Jori Mäntysalo
to be
sure).
Yes, that was it.
Ubuntu installer had made 1 GB swap, so I had total of 2 GB. I made a 5 GB
swapfile, and now it seems to compile. Very very slowly, but compiling
anyways.
--
Jori Mäntysalo
this, as it might be a real bug. The hardware should be OK, I
just wiped HD and installed 32-bit Ubuntu 16.04.1 LTS.
* * *
Btw, what is the oldest and slowest computer where SageMath currently can
be compiled (without docs, as docbuilding should preferably be done in a
supercomputer near you...)?
--
Jori
Could someone close all 49 wontfix-positive_review -tickets?
IMO it would be nice to have this done after every stable release.
--
Jori Mäntysalo
not
be listed in the high-level view.
If #1 adds foo() to graphs and #2 adds bar(), then the list should have
something like "Graph enchancements: foo() and bar()." Which ticket should
contain that information?
Hard question is this.
--
Jori Mäntysalo
like ZZ which is very common and have quite
long "real" name.)
Btw, is there an easy way for a user to add own aliases? So that it would
go to the right place in hierarchy.
--
Jori Mäntysalo
run
git diff --name-only develop your-new-branch | xargs python3...
--
Jori Mäntysalo
when making patches. That will save
time later.
--
Jori Mäntysalo
s_parallel() on
src/sage/combinat/posets/posets.py.
(IMO tab completion could even show linkage() when g.l is pressed and
show something like "linkage (alias to degeneracy)", but that's not a
decision that Sage should do, it should be a property of IPython.)
--
Jori Mäntysalo
your opinion.
* * *
I think we all agree that this should be consistent, and something like
is_induced_subposet() should also behave in the same way.
--
Jori Mäntysalo
{0: [0.50239985054502, 0], 1: [0.5037689172710763, 1]}
{0: [0.5001236499103481, 0], 1: [0.5075885512735917, 1]}
So should the save_pos do something when layout is not the default? It
seems that at least it is not strictly deterministic.
--
Jori Mäntysalo
kind of heuristic to guess what will be the fastest way.
--
Jori Mäntysalo
posets.py:
- "Dushnik-Miller" is mentioned in dimension()
- Eponymous "slender" is said in is_slender()
- "Semiorder" etc. are named in is_incomparable_chain_free()
--
Jori Mäntysalo
solution is to
make it _is_simple_deterministic(), so it won't show up in
-completion or html documentation.
is_cayley_directed() has no explanation of what is a Cayley graph.
I haven't read the paper, so I can not comment about the idea.
--
Jori Mäntysalo
examples? Is it on a worksheet,
or already integrated as a part of Sage code?
Have you tested that it is faster than what Sage currently have?
--
Jori Mäntysalo
get a wrong result from stack trace.
--
Jori Mäntysalo
ve snippet:
a = list(Subsets(range(13), 5))
And suddenly the first code snippet will never print anything.
Tell me that's not a bug!
To me it seems to be worst than a bug: intentionally broken design.
Something like this is almost impossible to teach in classroom.
--
Jori Mäntysalo
On Fri, 28 Oct 2016, Dima Pasechnik wrote:
Well, I suppose you can install more "usual" jupyter kernels, e.g. R one,
GAP one, etc.
Of course, but then profit for using Sage fades away.
On the plus side I could have "kernels" for, say, last two stable
versions and
stributive lattice is modular.)
etc.
--
Jori Mäntysalo
), which requires much more
ressources (I have checked with
giac and aborted after half an hour and more than 8G of RAM required).
OK and thanks! This explains clearly why this happens.
--
Jori Mäntysalo
(n) with given n, it is not that random.
--
Jori Mäntysalo
For other purposes Jupyterhub seems to be a great piece of software.
--
Jori Mäntysalo
, and then I also used Sage to
generate random polynomials.
--
Jori Mäntysalo
. It is much harder to debug. It is not that hard
to add some if M.ncols()==... before .is_similar().
--
Jori Mäntysalo
the algorithm Singular uses, and
the memory requirements. So this may be just a feature.
(Actually there is at least a minor bug: at mimum I should have gotten an
error message about running out of memory.)
--
Jori Mäntysalo
to get a random number. I don't
know if it plays well with Sage random state.
But in any case, this is just a test code to see how well Singular works.
--
Jori Mäntysalo
F = factor(P*Q)
and this got stuck. Might be that it just needs more memory than I have; I
don't know how much it normally uses.
--
Jori Mäntysalo
test this hypothesis?". Maybe some
day...)
--
Jori Mäntysalo
last look for worksheet sharing.
Thanks for help!
--
Jori Mäntysalo
F-8"
dpkg-reconfigure locales
pip3 install jupyterhub
Noticed that it used IPv6, got a headache when googling, found the
instruction to put GRUB_CMDLINE_LINUX_DEFAULT="ipv6.disable=1 quiet" on
/etc/default/grub, said update-grub and got the login prompt.
--
Jori Mäntysalo
you give me some kind of Hello world! -example of Jupyter? ./sage
--notebook=jupyter will open a server to port --- how to change the
port or, more general, is there some help file? First I would like to see
a way to have two local user using Sage and not seeing each other files.
--
Jori
to try this (yet).
How others teach Sage? Here I make a Sage account of the form
"Coursecode-Year", and the teacher asks students to share worksheet with
that account.
--
Jori Mäntysalo
) and so.
(Besides that, University of Tampere, Technical University of Tampere and
University of Applied Sciences of Tampere will be united from 1.1.2018.
So check out what other two wants and be ready for that too. Maple to
same server maybe?)
--
Jori Mäntysalo
uot;1+2", press Shift+Enter. Got "3".
Press arrow up, correct text to "1+3", again Shift+Enter and arrow up. And
now how to correct the line to "1+4" without mouse?
This is 64-bit Ubuntu, Firefox from repository, all up-to-date.
(OK, can do that, but is not quick and easy.)
--
Jori Mäntysalo
angle_free() and is_cayley() fails.
known_kaboom is needed as for example is_prime() don't just throw an
exception, it stops the whole workseet.)
--
Jori Mäntysalo
Anybody?
One -- and only one -- of my users if having really big problems with
cells not responding. Somehow computation seems to work, but the result is
not got back to worksheet. Pressing refresh helps sometimes, not always.
How to debug this?
--
Jori Mäntysalo
-- Forwarded
he line
I sent, and you can see that it will fail in memset(). (But, as said, with
different error message.)
I used 64-bit Ubuntu 16.04 LTS with most recent kernel from Ubuntu
packages. This kernel should not have any tweaks.
--
Jori Mäntysalo
("Failure allocating memory.")
but due to overcommit there is no exception raised. Setting (as root) the
value of /proc/sys/vm/overcommit_memory will not help, but may change
error message given.
--
Jori Mäntysalo
in advance when the new version will be available.
--
Jori Mäntysalo
that having two branches is just too much. In perfect world we
could have LTS with bugfixes only and so on, but we have not enought
manpower for that. But we could have some kind of decision like "7.6rc0
will be out at the end of April 2017."
--
Jori Mäntysalo
is
Sorry, I was unclear. They "see" it in techical sense, but mostly don't
read or notice.
("Hear but not listen", I don't know if there is same kind of verb pair
for visual reception.)
--
Jori Mäntysalo
-15).
In principle doable. It would mean that Volker won't change beta to rc
until some predefined date, and will make the change unless there is a
very good reason to release still one beta.
--
Jori Mäntysalo
On Mon, 17 Oct 2016, Erik Bray wrote:
Does Sage have *any* kind of roadmap planning?
No.
What kind of roadmap it could be? If some developers are interested in
graph theory, how to make them to add more linear algebra code to Sage?
--
Jori Mäntysalo
ogin -page.
--
Jori Mäntysalo
left author-field empty if I am not sure that I'll make patch
myself.
* * *
But we don't have any plans like "Version 8 will be out about q1/2018 and
will have mostly more support for numerical linear algebra.", and so
milestones are not really used.
--
Jori Mäntysalo
I opened https://trac.sagemath.org/ticket/2169 to make some example.
I guess you mean that for example
ValueError("%s and %s must be positive integers." % (m, n))
should be as it is, but
ValueError("the poset is not ranked")
changed to ArithmeticError.
--
Jori Mäntysalo
amming errors (an
input which a priori does not make sense).
Documentation for ValueError says "- - and the situation is not described
by a more precise exception - -". Maybe there should be something like
MathematicalError as a base class instead of ArithmeticError, but we can't
change that.
--
Jori Mäntysalo
a *method*
(i.e., a function with one argument "self"), and "self" of course has
the correct type but has a wrong value.
Yep, mostly those functions should raise ValueError. (Or, maybe, some
error derived from it; but that might be really, really bikeshedding.)
--
Jori Mäntysalo
ere.
So we have now a common view that 'type' in TypeError should (mostly?)
refer to types in wrong class, wrong category etc; and so for functions
having no input it should (almost?) never happen.
--
Jori Mäntysalo
t a
DiGraph with some special properties?
--
Jori Mäntysalo
is bad (but not nearly as bad as a bare
"except:")
OK, I'll correct those when I see one.
There are 56 on combinat, 50 on rings and 30 on graphs.
--
Jori Mäntysalo
out of memory or something like that?
--
Jori Mäntysalo
On Sun, 9 Oct 2016, Emmanuel Charpentier wrote:
sage.symbolic.expression.Expression, you jus have to patch the source and
recompile. That means a turnaround time of 10-40 minutes each time.
How? ./sabe -b takes less than a minute even on quite old machine.
--
Jori Mäntysalo
101 - 200 of 508 matches
Mail list logo
