Re: [sage-devel] Temporary files problems

[Andrey Novoseltsev]

On Tuesday, 27 September 2022 at 11:11:00 UTC-6 dim...@gmail.com wrote:

> Basically, we should deprecate and remove tmp_dir() and tmp_filename() 
> from Sage. 
> Does Sagecell use them? It should not, Python3 has perfectly good 
> replacements...
>

The problems that surfaces so far that I believe these tools may not 
address on their own:
1) tmp directory is used for "too long" i.e. while Sage process is 
running,  which may be a problem if it gets automatically cleaned
2) tmp stuff is deleted "too fast" e.g. as soon as it was generated, while 
some interfaces may rely on access to generated content a bit after Sage is 
done with it, e.g. to show plots in a browser
3) default location may not work with our interfaces as well

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/faca7769-4469-4c98-aace-04ab902a370dn%40googlegroups.com.

Re: [sage-devel] Temporary files problems

[Andrey Novoseltsev]

On Tuesday, 27 September 2022 at 03:22:47 UTC-6 dim...@gmail.com wrote:

> On Tue, Sep 27, 2022 at 3:58 AM Andrey Novoseltsev  
> wrote: 
> > this temporary directory is created and then used ever after. But what 
> if it gets deleted? 
> I think it's meant to be used in a context manager, i.e. with with 
> statement, e.g., citing python docs: 
>
> >>> with tempfile.TemporaryDirectory() as tmpdirname: 
> ... print('created temporary directory', tmpdirname) 
> >>> 
> # directory and contents have been removed 
>
> Or with a callback (from Sage source) 
>
> TMP_DIR_FILENAME_BASE=tempfile.TemporaryDirectory() 
> atexit.register(lambda: TMP_DIR_FILENAME_BASE.cleanup()) 
>
>
> Needless to say, one can interfere with it in a bad way messing around 
> with ~/.sage/ - but so it the case for any file-based process... 
>
>
It seems to me like a pretty common and sensible practice, that /tmp 
directories get cleaned up, especially on systems with lots of activity. I 
would not consider it as "a bad way messing around". But creating a 
temporary directory and then counting on it being there hours later for 
another quick temporary operation may be a bit too optimistic. That 
callback is designed to clean up after Sage quits, which is a good thing to 
do, but it is not related to being able to use the directory. 

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/231e405f-3af0-4a23-808e-0fb5edeb89bcn%40googlegroups.com.

[sage-devel] Re: Temporary files problems

[Andrey Novoseltsev]

I think it must be customizable, and also its use should be checked for 
errors - if it does not exist anymore, it should be recreated... Public 
SageCell servers probably see enough traffic to keep it alive, but people 
with private servers are very likely to run into problems, e.g. when their 
students are sleeping or pub crawling, all at once ;-)

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/53ebd0ef-dd25-4324-b2fd-0b44a478c577n%40googlegroups.com.

[sage-devel] Temporary files problems

[Andrey Novoseltsev]

Hello!

I've run into this while testing upgrades to Ubuntu 22.04 from 20.04, but 
it seems that it comes from how temporary files are handled in Sage 9.7:

TMP_DIR_FILENAME_BASE=tempfile.TemporaryDirectory()

this temporary directory is created and then used ever after. But what if 
it gets deleted? For example, what if someone uses tmpreaper or something 
similar, when old files are removed from /tmp? Sage does not notice it, but 
things break in a weird way, e.g. plots are not displayed and the error 
message is not transparent.

I believe this is exactly what is happening and I would appreciate some 
attention to this matter ;-)

Thank you!
Andrey

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/5ee63ec7-26b2-4c32-a5a7-45fb63a11327n%40googlegroups.com.

Re: [sage-devel] Re: VOTE: move Sage development to Github

[Andrey Novoseltsev]

+1 for Github

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/c259f261-dc79-4d93-8fe2-fad5f977de54n%40googlegroups.com.

[sage-devel] Suppressing R NULL output

[Andrey Novoseltsev]

Could somebody familiar with R interface take a look at
https://trac.sagemath.org/ticket/34378#ticket
please? I imagine it should be easy to fix ;-)

Thank you!
Andrey

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/001acae1-273c-4f82-8ab7-a4d8ff82fb23n%40googlegroups.com.

[sage-devel] Limitations of GAP interface

[Andrey Novoseltsev]

This has been reported at https://github.com/sagemath/sagecell/issues/560 
but as far as I can see is the problem of GAP interface in Sage itself. 
Perhaps somebody familiar with it may want to take a look ;-)

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/4247d756-91f0-49f5-aa4d-20a5e6baf662n%40googlegroups.com.

[sage-devel] Re: jpeg support in pillow

[Andrey Novoseltsev]

"sage -pip install pillow" does not install anything, saying that pillow is 
already there.

"sage -pip install -U pillow" does install a newer version with 
jpeg-support, but a) presumalby won't work if we happen to have the latest 
version already and b) has potential to break some other dependencies

So ideally jpeg would be supported by Sage directly, but if is not going to 
happen I'd rather keep things as is on the SageMathCell side.

On Monday, 18 April 2022 at 12:52:29 UTC-6 Matthias Koeppe wrote:

> Does the binary wheel that would be installed by "sage -pip install 
> pillow" provide jpeg support?
>
> On Monday, April 18, 2022 at 10:41:02 AM UTC-7 novo...@gmail.com wrote:
>
>> sage-cell got a request for jpeg support in pillow, which was explicitly 
>> removed 6 years ago, it seems, but people would like to have:
>> https://groups.google.com/g/sage-cell/c/_kZdSpnEtMs
>>
>> Are there any chances of doing it? My understanding is that it would 
>> require a dependency on libjpeg-dev and presumably it was not desirable at 
>> some point.
>>
>

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/ad1a754f-f722-47bf-9c4d-ab3b7daaf50bn%40googlegroups.com.

[sage-devel] jpeg support in pillow

[Andrey Novoseltsev]

sage-cell got a request for jpeg support in pillow, which was explicitly 
removed 6 years ago, it seems, but people would like to have:
https://groups.google.com/g/sage-cell/c/_kZdSpnEtMs

Are there any chances of doing it? My understanding is that it would 
require a dependency on libjpeg-dev and presumably it was not desirable at 
some point.

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/fea9f145-f7b4-48ec-83f6-70ee303e13b7n%40googlegroups.com.

[sage-devel] Re: multi-polynomial substitution consistency

[Andrey Novoseltsev]

On Sunday, 9 May 2021 at 02:50:06 UTC-6 vdelecroix wrote:

> What is annoying in the above scenario is that it breaks 
> badly the rule "same input parent" -> "same output parent". 
> Here are three distinct output parents for the same substitution 
> command
>

I completely agree with Vincent on this point. If you are going to do 
something further with the output in your code, having potentially 
different objects is very inconvenient. Since partial substitution (which 
is supported by "subs" method) almost certainly has to live in the same 
polynomial ring as the starting polynomial, I strongly favour the result of 
subs to be in this ring whenever possible. Perhaps if there is a full 
substitution and another natural ring to live in, as is the case with ring 
homomorphisms, it can be there, but do not drop to the ring of coefficients 
just because it is possible for a particular case.

On the other hand, call method does not support partial substitution and 
acts as an evaluation where it makes a lot of sense to use the coefficient 
ring by default, but again one can consider natural subcases.

Having these two methods with different intended output would be quite 
clean, it seems to me. 

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/df74f796-f506-437a-bf6e-a1f726794b46n%40googlegroups.com.

[sage-devel] Where should polynomial substitution live?

[Andrey Novoseltsev]

Consider the following:

sage: R. = QQ[] 
: p = x + y 
: print(parent(p.subs(x=1))) 
: print(parent(p.subs(x=1, y=2))) 
: print(parent(p(x=1))) 
: print(parent(p(x=1, y=2))) 
: print(parent(p(1, 2))) 
: print() 
: R. = QQ["a"][] 
: p = x + y 
: print(parent(p.subs(x=1))) 
: print(parent(p.subs(x=1, y=2))) 
: print(parent(p(x=1))) 
: print(parent(p(x=1, y=2))) 
: print(parent(p(1, 2))) 
Multivariate Polynomial Ring in x, y over Rational Field
Multivariate Polynomial Ring in x, y over Rational Field
Multivariate Polynomial Ring in x, y over Rational Field
Multivariate Polynomial Ring in x, y over Rational Field
Rational Field

Multivariate Polynomial Ring in x, y over Univariate Polynomial Ring in a 
over Rational Field
Univariate Polynomial Ring in a over Rational Field
Multivariate Polynomial Ring in x, y over Univariate Polynomial Ring in a 
over Rational Field
Univariate Polynomial Ring in a over Rational Field
Univariate Polynomial Ring in a over Rational Field

It seems to me that when a tuple of values is provided for all variables, 
the result is expected to be the polynomial evaluation and should live in 
the base ring. But if named substitutions are used, the result is going to 
be a polynomial of the same ring, even if it happens to be a constant for 
some reason. If that is not the case, one has to write different code to 
treat evaluations. So I'd say that the first ring here behaves well and the 
second one does not. What do you think?

Thank you!
Andrey

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/70ebd34f-0748-4472-a4db-d538fc896a96n%40googlegroups.com.

[sage-devel] Re: What should Sage import by default?

[Andrey Novoseltsev]

Perhaps it makes sense to keep "math related" imports given that we have a 
bunch of other "math functionality" available without extra imports and 
that's the point of Sage in general. While stuff for dealing with files and 
warnings is special enough to require special commands.

On Wednesday, 2 September 2020 at 00:08:57 UTC-6 Kwankyu Lee wrote:

> +1 to remove all.
>
> On Wednesday, September 2, 2020 at 1:50:07 PM UTC+9 John H Palmieri wrote:
>
>> There is a ticket (https://trac.sagemath.org/ticket/25383) about 
>> removing some Sage functions from the global namespace, which I think is a 
>> good idea. But Sage also imports some Python modules: 
>>
>> - os
>> - sys
>> - operator
>> - math
>> - warnings
>>
>> Should we keep all of these? Remove all? Keep some?
>>
>> -- 
>> John
>>
>>

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/d623980b-2004-47c0-af18-31af87e0dabdn%40googlegroups.com.

[sage-devel] Preparser is seriously broken!

[Andrey Novoseltsev]

It interferes with multiline strings now!!!

┌┐
│ SageMath version 9.2.beta9, Release Date: 2020-08-18   │
│ Using Python 3.7.3. Type "help()" for help.│
└┘
┏┓
┃ Warning: this is a prerelease version, and it may be unstable. ┃
┗┛
sage: html(r""" 
: $2 \leq x$ 
: """) 


Integer(2)  * BackslashOperator() * leq x

sage: html(r"""$2 \leq x$""")   
   
2 \leq x

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/80aeffb4-4608-42e5-86d0-77d15a748f06n%40googlegroups.com.

[sage-devel] Re: cryptominisat does not buld on Ubuntu 20.04

[Andrey Novoseltsev]

I missed the comment indeed, but I remember doing "make cmake-clean"

On Friday, 24 July 2020 11:26:27 UTC-6, Matthias Koeppe wrote:
>
> Andrey, did you see this comment?
>
> On Wednesday, July 15, 2020 at 1:48:02 PM UTC-7, Matthias Koeppe wrote:
>>
>> Are you sure you did "make cmake-clean"? The last build log still seems 
>> to be using a cmake from SAGE_LOCAL
>>
>> On Wednesday, July 15, 2020 at 1:02:20 PM UTC-7, Andrey Novoseltsev wrote:
>>>
>>> Here it is, for some reason much bigger than before, maybe there are 
>>> several versions - are logs overwritten on each attempt? I forgot to make 
>>> cmake-clean first, but the last one was after installing cmake from Ubuntu 
>>> and then doing "make cmake-clean" in sage root.
>>>
>>

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/b4040e09-3fb6-4efc-951b-31188ce3546ao%40googlegroups.com.

[sage-devel] Re: cryptominisat does not buld on Ubuntu 20.04

[Andrey Novoseltsev]

For the record: NOT installing system libboost-dev allowed cryptominisat to 
be installed in Sage. Good enough solution for me, but the reason I was 
installing libboost-dev was ./configure recommendation...

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/5056b33c-bf13-4cc6-85fa-751ad99f7765o%40googlegroups.com.

[sage-devel] Re: cryptominisat does not buld on Ubuntu 20.04

[Andrey Novoseltsev]

On Wednesday, 15 July 2020 14:21:59 UTC-6, Matthias Koeppe wrote:
>
> On Wednesday, July 15, 2020 at 1:02:20 PM UTC-7, Andrey Novoseltsev wrote:
>>
>> Here it is, for some reason much bigger than before, maybe there are 
>> several versions - are logs overwritten on each attempt? 
>>
>
> The package log files are never deleted. A output from a new build is 
> appended.  
>
> Could you also check the contents of local/lib/cmake/ ?
>
>
sc_serv@precell:~$ ls sage/local/lib/cmake/
openblas

That's 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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/d7a9ccad-97b0-4f08-b4a4-ddf3cbb2002ao%40googlegroups.com.

[sage-devel] Re: cryptominisat does not buld on Ubuntu 20.04

[Andrey Novoseltsev]

On Wednesday, 15 July 2020 10:38:46 UTC-6, Matthias Koeppe wrote:
>
> Andrey, actually could you try if installing "cmake" from the system 
> package manager (and doing "make cmake-clean") fixes this problem?
> We ship an ancient version of cmake in sage.
>

Does not seem to make any difference! 

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/3fa32110-6c9f-423d-887f-cef8e0766e27o%40googlegroups.com.

[sage-devel] Re: cryptominisat does not buld on Ubuntu 20.04

[Andrey Novoseltsev]

On Tuesday, 14 July 2020 10:42:49 UTC-6, Matthias Koeppe wrote:
>
> Thanks for spotting this.
>
> Andrey: I would suggest to try whether a cryptominisat upgrade (
> https://trac.sagemath.org/ticket/25374) has better Boost detection.
>
> If not, we should change the boost spkg-configure so that it passes the 
> found boost library location on to cryptominisat.
>
> I've tried following the directions on the ticket in the same installation 
that fails current version and the result is the same:

...
[cryptominisat-5.8.0] /usr/bin/ld: 
CMakeFiles/cryptominisat5-bin.dir/main.cpp.o:(.data.rel.ro._ZTVN5boost15program_options11typed_valueIbcEE[_ZTVN5boost15program_options11typed_valueIbcEE]+0x38):
 
more undefined references to 
`boost::program_options::value_semantic_codecvt_helper::parse(boost::any&,
 
std::vector, 
std::allocator >, std::allocator, std::allocator > > > const&, bool) const' 
follow
[cryptominisat-5.8.0] /usr/bin/ld: 
CMakeFiles/cryptominisat5-bin.dir/main.cpp.o:(.data.rel.ro._ZTVN5boost16exception_detail19error_info_injectorINS_15program_options20invalid_option_valueEEE[_ZTVN5boost16exception_detail19error_info_injectorINS_15program_options20invalid_option_valueEEE]+0x20):
 
undefined reference to 
`boost::program_options::error_with_option_name::what() const'
[cryptominisat-5.8.0] /usr/bin/ld: 
CMakeFiles/cryptominisat5-bin.dir/main.cpp.o:(.data.rel.ro._ZTVN5boost16exception_detail19error_info_injectorINS_15program_options20invalid_option_valueEEE[_ZTVN5boost16exception_detail19error_info_injectorINS_15program_options20invalid_option_valueEEE]+0x30):
 
undefined reference to 
`boost::program_options::error_with_option_name::substitute_placeholders(std::__cxx11::basic_string, std::allocator > const&) const'
[cryptominisat-5.8.0] /usr/bin/ld: 
CMakeFiles/cryptominisat5-bin.dir/main.cpp.o:(.data.rel.ro._ZTVN5boost16exception_detail10clone_implINS0_19error_info_injectorINS_15program_options20invalid_option_valueE[_ZTVN5boost16exception_detail10clone_implINS0_19error_info_injectorINS_15program_options20invalid_option_valueE]+0x28):
 
undefined reference to 
`boost::program_options::error_with_option_name::what() const'
[cryptominisat-5.8.0] /usr/bin/ld: 
CMakeFiles/cryptominisat5-bin.dir/main.cpp.o:(.data.rel.ro._ZTVN5boost16exception_detail10clone_implINS0_19error_info_injectorINS_15program_options20invalid_option_valueE[_ZTVN5boost16exception_detail10clone_implINS0_19error_info_injectorINS_15program_options20invalid_option_valueE]+0x38):
 
undefined reference to 
`boost::program_options::error_with_option_name::substitute_placeholders(std::__cxx11::basic_string, std::allocator > const&) const'
[cryptominisat-5.8.0] /usr/bin/ld: 
CMakeFiles/cryptominisat5-bin.dir/main_exe.cpp.o: in function 
`boost::program_options::variables_map::~variables_map()':
[cryptominisat-5.8.0] 
/usr/include/boost/program_options/variables_map.hpp:146: undefined 
reference to `vtable for boost::program_options::variables_map'
[cryptominisat-5.8.0] collect2: error: ld returned 1 exit status
[cryptominisat-5.8.0] make[4]: *** 
[cmsat5-src/CMakeFiles/cryptominisat5-bin.dir/build.make:133: 
cryptominisat5] Error 1
[cryptominisat-5.8.0] make[4]: Leaving directory 
'/home/sc_serv/sage/local/var/tmp/sage/build/cryptominisat-5.8.0/src'
[cryptominisat-5.8.0] make[3]: *** [CMakeFiles/Makefile2:163: 
cmsat5-src/CMakeFiles/cryptominisat5-bin.dir/all] Error 2
[cryptominisat-5.8.0] make[3]: Leaving directory 
'/home/sc_serv/sage/local/var/tmp/sage/build/cryptominisat-5.8.0/src'
[cryptominisat-5.8.0] make[2]: *** [Makefile:130: all] Error 2
[cryptominisat-5.8.0] make[2]: Leaving directory 
'/home/sc_serv/sage/local/var/tmp/sage/build/cryptominisat-5.8.0/src'

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/0f6e93b7-7f87-41cc-ac96-c8b0a485d5c5o%40googlegroups.com.

[sage-devel] Re: cryptominisat does not buld on Ubuntu 20.04

[Andrey Novoseltsev]

The way I was trying to do it:

- create Ubuntu 20.04 LXC container (host is the same OS on a fresh GCE 
instance)
- update & install programs suggested by ./configure
- build Sage

Everything works apart from cryptominisat and whatever the problem is, it 
is not due to me not updating the system or doing something strange to 
break it ;-)

Andrey

On Monday, 13 July 2020 23:26:06 UTC-6, Zachary Scherr wrote:
>
> I was able to successfully build cryptominisat using an Ubuntu 20.04 
> docker image.  Some weird things I noticed from your log files:
>
> config log says you have boost lib >= 1.66 yet your cryptominisat log file 
> says "Boost 1.46 found".  On the next line it says:
>
> "-- Found Boost components: program_options"
>
> which seems to be related to all of the errors.
>
> I'll let a more experienced person chime in, but I imagine that if you get 
> the latest version of libboost-dev and cmake from apt and let sage find it 
> via ./configure then maybe cyrptominisat will build.
>
> On Monday, July 13, 2020 at 11:26:06 PM UTC-4 Andrey Novoseltsev wrote:
>
>> On Monday, 13 July 2020 21:11:11 UTC-6, Matthias Koeppe wrote:
>>>
>>> Please also send the top-level config.log 
>>>
>>
>> Attached!
>>
>>
>>> Lots of optional and experimental packages are broken. See 
>>> https://trac.sagemath.org/ticket/29900 - which needs help.
>>>
>>
>> Thank you for working on this! Can't commit to more help than supplying 
>> logs...
>>
>>

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/5298591f-0961-4aa1-a887-60b5109b9688o%40googlegroups.com.

[sage-devel] Re: Interacts on wiki need some care

[Andrey Novoseltsev]

To make optional or pip or else packages available on SageMathCell public 
servers one has to let me know what should be installed - assuming that it 
is indeed possible to install these things they will become available ;-)

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/f475ea9c-f04b-431d-b7ae-5ad8023f7a7e%40googlegroups.com.

[sage-devel] Interacts on wiki need some care

[Andrey Novoseltsev]

Hello,

William recently mentioned our collection of interacts 
https://wiki.sagemath.org/interact in 
https://stackoverflow.com/questions/59997930/embed-local-jupyter-notebook-cells-in-web-browser/60089271#60089271

No surprises - many of these interacts need some care to start working 
without displaying warnings under Python 3. If anyone cares about it - 
please do fix them!

Thank you,
Andrey

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/0b4d183f-1c65-4ea9-b9d0-08cb506cca47%40googlegroups.com.

[sage-devel] Re: drop python2 compatibility in 9.1 ?

[Andrey Novoseltsev]

On Monday, 6 January 2020 03:13:02 UTC-7, Frédéric Chapoton wrote:
>
> the wiki promise was written by myself alone, and I think it should be 
> changed. I think there will be very few (or  no) people that will desire to 
> compile sage 9.0 with python 2. And 9.1 even more. People that want to be 
> up to date do not care about py2.
> I therefore strongly vote for the immediate drop of python2, and for 
> removing the promise that 9.1 will be py2-compatible
>

I think that is the most important voice so far. When a single person on 
his own accord made a statement and posted it on a wiki page it should not 
be considered an official and unbreakable promise. When that same person 
wants to pull it back, and for a good reason, why should not we allow this?

We can talk about what would be nice, like running multiple cell servers, 
but let's pay attention to a) what is really happening and b) what are 
people actually willing to do in the future. I am personally not willing to 
work on multiple cell servers - those who really want and have machines 
available can quite easily set up their own server, I have tried hard to 
make it happen a long time ago. Others HAVE to switch to Python 3 sooner or 
later no matter what. Well, that time is now. I've timed the switch as best 
as I could releasing Python 3 version on Christmas and giving those 
preparing for the new term two weeks to adjust. Even with running another 
server some adjustments would be necessary.

Having a deprecation for print statement a year ago would be very useful 
and it was discussed and agreed on - as far as usefulness goes. But nobody 
actually worked on that change, which, I suspect, would be much easier than 
maintaining compatibility with two Pythons now.

I want to reiterate again that I am personally not affected by the decision 
made here in any way. But after years of adjusting SageMathCell to new 
versions of Sage, IPython, and now Python I think I have some idea of 
effort involved for those who do the porting. And I strongly believe that 
efforts of these people are too valuable to be spent on this support.

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/1c27defa-db0d-4e0e-826a-7b82d620c5b5%40googlegroups.com.

[sage-devel] Re: drop python2 compatibility in 9.1 ?

[Andrey Novoseltsev]

On Sunday, 5 January 2020 16:24:38 UTC-7, Nils Bruin wrote:
>
> On Sunday, January 5, 2020 at 2:51:44 PM UTC-8, Nils Bruin wrote:
>>
>> I don't think there's a doubt that we'll drop py2.7 support. This Wiki, 
>> however, has been referred to in many places describing the py2/py3 
>> transition.
>>
>
> One of these places is the FAQ, which is official documentation:
>
>
> http://doc.sagemath.org/html/en/faq/faq-usage.html#can-i-use-sagemath-with-python-3-x
>
> The 9.0 FAQ still mentions:
>
> As of August 2019, most of SageMath works fine with Python 3. However, we 
> still consider Python 3 support to be experimental and no official Python 3 
> release has been made yet.
>
> That would probably be good to adjust too: there is now an official Py3 
> release and I don't think we consider it experimental anymore -- unless 
> sage in its entirety is still considered experimental :-).
>
>
I fail to see the difference between "officialness" of this documentation 
statement that Python 3 support is experimental and the wiki promise that 
9.1 will work with Python 2. If the first one is wrong and it is OK to 
correct it and move forward, why can't the second one also be corrected?

Those who do want to keep that promise intact - can you please explain who 
will be affected and how if we drop Python 2 support? "Regular" users do 
not have a choice - precompiled binaries are with Python 3 only. Those who 
can build from source and add configure parameters may just as well 
checkout 9.0 and compile that.

Note that I am not a particularly active Sage developer anymore - I 
maintain SageMathCell, which is switched to Python 3 already. So it does 
not affect me and my workload in any way whether 9.1 is Python 2 compatible 
or not. But for the sake of those who do work on transition - please don't 
force them to spend their valuable energy on a pretty much pointless 
compatibility with unsupported languages.

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/48277f25-e4e8-4ff5-a353-d27d6c0a618c%40googlegroups.com.

[sage-devel] jmol directory permissions

[Andrey Novoseltsev]

Hello,

I believe jmol directory permissions are screwed up, sage/local/share/jmol 
is 0700 making it unusable by other users.

Looking at the install scripts it is not obvious to me why does this 
happen, so I would appreciate if someone more knowledgeable fixed it ;-)

Thank you!
Andrey

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/48a4c388-f566-4ca0-af46-811568448b28%40googlegroups.com.

Re: [sage-devel] No output from GAP

[Andrey Novoseltsev]

OK, here is me not trying to rely on implicit printing by REPL:

sage: gap.eval("Print({});".format(10^1))
'10'
sage: gap.eval("Print({});".format(10^10))
'100'
sage: gap.eval("Print({});".format(10^100))
''

Is gap pexpect interface still used internally? If not, it seems that 
consistent end-user behaviour is important, while those who want 
optimization should use libgap...

-- 
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] No output from GAP

[Andrey Novoseltsev]

Here is the logic of executing GAP commands via files:

if it has := before "
  write it to file as is
otherwise
  wrap it in Print( )
execute via Read() which has no input on its own

However when multiple commands are fed into gap.eval as in the original 
post, they all are combined into a single line for which the above 
algorithm makes no sense - if there is an assignment there will be no 
output and if there is no assignment multiple commands will be wrapped in a 
single Print resulting in an error inside GAP.

Perhaps it is supposed that there is some logic breaking up multiple 
commands into separate "lines", but it is obviously broken and in general 
it is not straightforward since one has to take into account comments, 
strings, and compound operators like loops.

It seems to me that the best fix for the current situation is to prohibit 
using files completely. Any reasons NOT to do it?

I also played a bit with libgap.eval and while it requires a "single" 
command and complains if I give many, it is happy to accept compound 
statements like loops. Which makes me think that its restriction on having 
a single command only is a bit artificial/non-essential. Does anyone know 
why it is present?

-- 
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] No output from GAP

[Andrey Novoseltsev]

The use case here is: take a snippet of GAP code, execute it, and display 
the result. GAP is one of the languages "supported" by SageMathCell and it 
is done via print gap.eval(...) It is not really for users of Sage who need 
GAP functionality, it is an easy way for GAP users to embed live 
computations into web pages.

-- 
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] No output from GAP

[Andrey Novoseltsev]

On Monday, 29 April 2019 14:20:39 UTC-6, Dima Pasechnik wrote:
>
> would it be easier to switch to libgap? 
>

Do you mean this? 
http://doc.sagemath.org/html/en/reference/libs/sage/libs/gap/libgap.html?highlight=libgap#sage.libs.gap.libgap.Gap.eval
There is however a restriction on no trailing semicolon. For SageMathCell 
it is very likely that the input contains several commands (as it was the 
case for the original example here). What can be done in this case? Just 
splitting on l; is likely to break on comments and strings.


> On Mon, Apr 29, 2019 at 5:54 PM Andrey Novoseltsev  > wrote: 
> > 
> > Reminded me of this old issue with Macaulay2 many years ago: 
> https://trac.sagemath.org/ticket/5467 I wonder if it is something similar 
> 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-...@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] No output from GAP

[Andrey Novoseltsev]

Reminded me of this old issue with Macaulay2 many years ago: 
https://trac.sagemath.org/ticket/5467 I wonder if it is something similar 
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.

Re: [sage-devel] No output from GAP

[Andrey Novoseltsev]

Thank you for the confirmation and debugging, Vincent!

It was pointed out to me that we even have it documented:
http://doc.sagemath.org/html/en/reference/interfaces/sage/interfaces/gap.html#long-input
Is it really that difficult to fix??? Surely there has to be some reason 
and solution for this behaviour...

-- 
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] No output from GAP

[Andrey Novoseltsev]

Hello,

I got a report about GAP problem in SageMathCell, but it looks like the GAP 
interface in Sage itself is having issues. I am getting an empty string 
from the following command:
sage: print(gap.eval("""
: D_4 := SmallGroup(8,3);
: IsAbelian( D_4 );
: List( ConjugacyClasses( D_4 ), Size );
: IsCyclic( D_4 );
: Size( AllSubgroups( D_4 ) );
: Size( Set( D_4, Order ) );
: IsSolvable( D_4 );"""))

All commands work fine in GAP and even in Sage each one seems to be 
working, but certain combinations of them do not, while some do - I can't 
understand the system here...

Does anyone have any ideas on what is going on?

Thank you!
Andrey

-- 
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: Poll: set online=True as the default for threejs viewer

[Andrey Novoseltsev]

https://trac.sagemath.org/ticket/27575 addresses this by fixing an 
apparently already existing "fallback to the online CDN" and claiming that 
this discussion here becomes moot.

But I think what the discussion here tried to address is that it is bad for 
a local install of Sage to silently fetch stuff from Internet, which is 
going to be the result of that fix in some circumstances.

-- 
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: [sage-cell] Re: Is sagecell python3-ready?

[Andrey Novoseltsev]

I actually just checked: if you create linked cells, it is possible to run 
multiple interacts in them, so perhaps it is the solution for your use 
case. Things did get a bit strange when I caused an error in one cell - I 
didn't investigate the details, but likely this should not be the default 
behaviour.

-- 
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: [sage-cell] Re: Is sagecell python3-ready?

[Andrey Novoseltsev]

On Saturday, 23 March 2019 00:25:10 UTC-6, Kwankyu Lee wrote:
>
> I would expect one process runs all interact cells from one client (if 
> this makes sense).   
>
> I have a webpage that contains many interact cells that run on my own 
> sagecell server. The webpage stops working if many of the interact cells go 
> active. 
>

How exactly multiple interacts are supposed to share resources of this 
single process? (What happens actually if you link them together now?) And 
what do you expect to gain? More efficient RAM usage? Note that if a single 
interact crashes, you will crash all of them. And things like CPU time 
limits would count cumulatively, so you may use up your allotment faster 
(which is not necessarily a bad thing, just something to keep in mind). 
Have you determined what exactly goes wrong when your webpage stops working?

-- 
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: [sage-cell] Re: Is sagecell python3-ready?

[Andrey Novoseltsev]

On Thursday, 21 March 2019 22:48:08 UTC-6, Kwankyu Lee wrote:
>
>
>
> On Friday, March 22, 2019 at 12:56:58 AM UTC+9, Andrey Novoseltsev wrote:
>>
>> "languages" for SageMathCell are very different from kernels for Jupyter. 
>> SageMathCell keeps a bunch of preforked kernels (all are the same)
>>
>
> And a new kernel is created for each interact cell. Right? This seems to 
> me a very bad architectural weakness of SageMathCell.  
>

Right. Feel free to fix it ;-) Do you want to have the same process serving 
multiple clients simultaneously? With each client allowed to execute 
arbitrary code, in particular crashing that kernel?

-- 
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: [sage-cell] Re: Is sagecell python3-ready?

[Andrey Novoseltsev]

When I was actively using SageMathCell for teaching, I indeed got quite 
irritated that things that used to work during preparation did not work in 
class because of some recent changes. My eventual solution was to run my 
own server, where no changes were ever unexpected and in general I had 
better stability. I think this is still the way to go for those who want to 
be certain that things will work, although I have a feeling that not very 
many people are doing it. Python version change is likely to be the most 
disruptive one, so it may make sense to do it only during the summer or 
around New Year, so that we don't cause major disruptions in the middle of 
the course. Perhaps UTMOST people may chime in with their predictions on 
how bad the transition would be?

-- 
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: [sage-cell] Re: Is sagecell python3-ready?

[Andrey Novoseltsev]

"languages" for SageMathCell are very different from kernels for Jupyter. 
SageMathCell keeps a bunch of preforked kernels (all are the same), so that 
new computations don't have to wait for startup. "language" just turns 
"code" into something like "print octave.eval(code)", which is then sent to 
the same kernels as Sage. Supporting Python2 and Python3 kernels 
simultaneously would require substantial adjustments to how kernels and 
languages are managed. An alternative I mentioned at sage-cell is to have 
different domain names for servers running different versions. Since we 
currently have 2 servers, it is doable, but we will loose redundancy.

-- 
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] Poll: set online=True as the default for threejs viewer

[Andrey Novoseltsev]

On Friday, 15 February 2019 14:21:45 UTC-7, mmarco wrote:
>
> 2) Patching sage to look for those files in the new addresses
>

There is actually a special function to provide these addresses, here is 
its use in SageMathCell:
https://github.com/sagemath/sagecell/blob/master/backend_cell.py#L176

-- 
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] Poll: set online=True as the default for threejs viewer

[Andrey Novoseltsev]

The bottom line for all these data points, however, is not that "online HAS 
to be default" but rather "all these interfaces didn't figure out (or 
perhaps even didn't bother) yet how to make offline threejs work". And the 
solution is either to fix it case by case or, if it is 
difficult/impossible, make "online" default just for those particular 
interfaces, not in the Sage itself.

As for the vote, I join those who think that silent default "online" is 
unacceptable.

I could agree to some warning showing that "offline" is not working in 
particular environments and explicitly specifying "online" is a workaround.

-- 
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] R.eval behaves differently with the new interface

[Andrey Novoseltsev]

Just want to bring some attention to
https://trac.sagemath.org/ticket/27025#ticket
in case someone knows what to do with it.

There is now 'NULL' string returned in cases when there was no output and 
multiple commands may not generate "automatic" output.

I do get complaints about it from SageMathCell users.

Thank you!
Andrey

-- 
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: What the hell is JSAGE ?

[Andrey Novoseltsev]

>From the foundation page: "People that have used SageMath in their research 
work and have some available money from their grants are also very welcome 
to give a little back to help."

Is it indeed easy to donate money from a grant??? I.e. can it easily be 
justified as a reasonable expense and fit with any spending guidelines? If 
so, perhaps this should be somehow more publicized, as I hope there may be 
a few researches who wound not mind donating some money, especially if 
there are some "leftovers" in a soon expiring grant?

-- 
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] Single instance of R only

[Andrey Novoseltsev]

On Tuesday, 25 December 2018 14:55:02 UTC-7, Timo Kaufmann wrote:
>
> Am Dienstag, 25. Dezember 2018 22:46:22 UTC+1 schrieb Nils Bruin:
>>
>> Perhaps for reference, maxima_lib can only be instantiated once and, 
>> since sage.interfaces.maxima_lib is only referenced at start-up via 
>> lazy_import (which has its own problems), the instantiation happens upon 
>> import of sage.interfaces.maxima_lib:
>>
>> sage: import sage.interfaces.maxima_lib
>> sage: sage.interfaces.maxima_lib.MaximaLib()
>> RuntimeError: Maxima interface in library mode can only be instantiated 
>> once
>> sage: sage.interfaces.maxima_lib.maxima #already instantiated
>> Maxima_lib
>>
>> The pexpect interface, on the other hand, can be instantiated as much as 
>> you want, and separate instances are *not* equal (and they shouldn't be, 
>> because generally they would be in quite different states):
>>
>> sage: M1=sage.interfaces.maxima.Maxima()
>> sage: M2=sage.interfaces.maxima.Maxima()
>> sage: M1 == M2
>> False
>>
>> I would say it would be reasonable to have R_lib follow the same pattern. 
>> If it is deemed that maintaining a pexpect interface to R is not 
>> worthwhile, then that's a separate decision.
>>
>
> Currently the R instance (which is in the global namespace by default) is 
> initialized lazily on first use. The pexpect interface doesn't exist 
> anymore. 
>

Given maxima example, I think that indeed the new interface should be 
renamed into R_lib (or RLib?) with the same behaviour, while R() should be 
a deprecated alias to it explaining what has happened. It is a bit 
unfortunate that the change has already happened and 8.5 has "silent 
replacement", but there are plans to release 8.6 soon, if I understood 
things correctly. To be clear - I do not volunteer to do this, but perhaps 
for Timo it will be easy after all the great work he has already done on 
this interface ;-)

-- 
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] Single instance of R only

[Andrey Novoseltsev]

On Tuesday, 25 December 2018 13:53:07 UTC-7, Timo Kaufmann wrote:
>
> Am Dienstag, 25. Dezember 2018 16:32:10 UTC+1 schrieb Andrey Novoseltsev:
>>
>> It does not affect me personally at all, my use of R is limited to making 
>> it work in SageMathCell. But if there is only one instance, then I think 
>> R() should always return that single instance rather than a new wrapper 
>> each time and perhaps it should be deprecated and removed from the global 
>> namespace so that interaction goes through the precreated r object only.
>>
>
> Not sure how deprecation in sage works. Wouldn't that needlessly break 
> some scripts?
>  
>
Depends on your definition of needlessly - if there is no point anymore to 
call R(), what's the point in keeping doing it? And if the use case before 
was to create two different instances, then it is highly desirable to break 
those scripts to make sure that there are no unexpected and hard to debug 
consequences.

-- 
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] Single instance of R only

[Andrey Novoseltsev]

On Tuesday, 25 December 2018 03:55:13 UTC-7, vdelecroix wrote:
>
> Since there is only one process, there should be a single instance of 
> the R object in Sage. Isn't that a trivial fix? 
>
> Disallowing multiple R instances might be a misfeature. But it is not 
> dramatic (to my mind and to the OP). 
>
> Vincent 
>

It does not affect me personally at all, my use of R is limited to making 
it work in SageMathCell. But if there is only one instance, then I think 
R() should always return that single instance rather than a new wrapper 
each time and perhaps it should be deprecated and removed from the global 
namespace so that interaction goes through the precreated r object only. It 
seems to me that the main use for R() vs. r before would be to create a new 
instance which does not work anymore. (Although I used it to pickup a new 
working directory and alter graphic initialization commands.)

-- 
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] Single instance of R only

[Andrey Novoseltsev]

On Sunday, 23 December 2018 19:23:27 UTC-7, William wrote:
>
> I didn't have anything to do with the change, but for what it is 
> worth, it was definitely not my intention that calling R() twice only 
> creates one interface. 
> When I designed and implemented the interfaces, my intention was that 
> it be possible to create multiple interfaces to multiple copies of R 
> (or to any 
> other pexpect interface). 
>
>  -- William 
>
>
Well, that was also my impression given how Interface class starts:

class Interface(WithEqualityById, ParentWithBase):
"""
Interface interface object.

.. NOTE::

Two interfaces compare equal if and only if they are identical
objects (this is a critical constraint so that caching of
representations of objects in interfaces works
correctly). Otherwise they are never equal.
"""

The change for R() happened since it is no longer a pexpect interface, but 
rpy2 one. I have no idea how exactly it works and if it is possible to use 
it to create multiple instances or namespaces with it, but perhaps 
knowledgeable people should look into 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] Single instance of R only

[Andrey Novoseltsev]

This is what we had:
┌┐
│ SageMath version 8.4, Release Date: 2018-10-17 │
│ Type "notebook()" for the browser-based notebook interface.│
│ Type "help()" for help.│
└┘
sage: R().eval("a = 2"); R().eval("a + 3")
''
"Error: object 'a' not found"


and this is what we have now:
┌┐
│ SageMath version 8.5, Release Date: 2018-12-22 │
│ Using Python 2.7.15. Type "help()" for help.   │
└┘
sage: R().eval("a = 2"); R().eval("a + 3")
'[1] 2'
'[1] 5'
sage: R() == R()
False

is this supposed to be the case now, i.e. that multiple attempts to create 
R interface are linked to the same instance despite being different objects?

-- 
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] Dropping support for old-style .spkgs?

[Andrey Novoseltsev]

On Friday, 10 August 2018 04:23:45 UTC-6, Erik Bray wrote:
>
> On Fri, Aug 10, 2018 at 11:46 AM Dima Pasechnik  > wrote: 
> > 
> > could you offer this package for download, at least temporarily? 
> > Perhaps we could put it on github... 
>
> Where did it come from in the first place?  Is this some pre-computed 
> data?  Was it computed by Sage?  Something else...? 
>

Data originates from http://hep.itp.tuwien.ac.at/~kreuzer/CY/

The whole list can be generated using PALP (a standard package in Sage), 
but when they have done it first, it took about a year on multiple 
computers, so there is sense in just downloading these polytopes. Sage 
package contains precomputed data for Hodge numbers as well, allowing quick 
search for particular cases. Volker knows more about exact details, I was 
just happily using the result.

While we are at it: it would be awesome if installation of this package 
after upgrade was faster. "Installation" just means unpacking files and 
copying them somewhere, but it takes something like 45 minutes. Some time 
ago I asked to use parallel decompressor if it is available in the system, 
which helped a lot, but then it stopped again.

Thank you!
Andrey

-- 
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] Dropping support for old-style .spkgs?

[Andrey Novoseltsev]

On Wednesday, 8 August 2018 17:19:13 UTC-6, Travis Scrimshaw wrote:
>
> IMO, it would be good to have here a current list of old-style spkgs so we 
> can be explicit about what things we are dropping support for.
>

And can we please not omit "huge" packages like polytopes_db_4d-1.0.spkg - 
I do use it and would like to keep it available, but I have no idea where 
it is located and whether downloading it still works - I use a local copy 
for many years (it is 8.7G). 

-- 
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] How much do we support the casual user

[Andrey Novoseltsev]

On Tuesday, 27 March 2018 18:03:48 UTC-6, saad khalid wrote:
>
> Why not assume by default that when someone enters a floating point 
> number, they intend it as a member of QQ, at least in this case. 
>

It may be an interesting option to have similar to "automatic_names". It 
may also be nice to be able to do computations over QQ, but display results 
as floats of some not very high precision. When I was teaching the simplex 
method, it was a bit tricky to choose between exact computations with scary 
looking (for students ;-)) denominators where it is hard to compare two 
numbers in your head, or inexact computations where you have to feel when a 
non-zero is actually zero. But I definitely would not want to have either 
conversion a silent default.

-- 
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] How much do we support the casual user

[Andrey Novoseltsev]

On Tuesday, 27 March 2018 13:57:12 UTC-6, Michael Orlitzky wrote:
>
> But will MATLAB tell you that two equal matrices have different ranks? I 
> know sage will do it. 
>

A few years back it could give you completely different eigenvalues for a 
matrix and its transposition. There are presumably many cases which are 
"easy to fix" one by one, but it is tricky to take care of all of them at 
once systematically. 

-- 
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: ~/.sage/R created on the second run only

[Andrey Novoseltsev]

On Sunday, 4 March 2018 19:12:14 UTC-7, Andrey Novoseltsev wrote:
>
> On the first run of Sage, it is supposed to create some folders in .sage 
> What I observe is that R directory is created only on the second run (and 
> Sage fails to start if by then .sage is made immutable). Looking at 
> https://github.com/sagemath/sage/blob/master/src/bin/sage-env I fail to 
> see how it can happen.
>

Looking at it from the beginning, here is what happens:
- sage-env sets a variable for  .sage, but does not care about its 
(non)existence
- if .sage does not exist, it will be created by some other scripts later on
- sage-env creates files/directories to make R happier, but only if .sage 
exists, which is not the case on the first run

Personally I find this logic a bit convoluted and unexpected - why not 
create all pieces of the required tree in a single place or at least in 
hierarchical order? 

-- 
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] ~/.sage/R created on the second run only

[Andrey Novoseltsev]

Hello,

On the first run of Sage, it is supposed to create some folders in .sage 
What I observe is that R directory is created only on the second run (and 
Sage fails to start if by then .sage is made immutable). Looking at 
https://github.com/sagemath/sage/blob/master/src/bin/sage-env I fail to see 
how it can happen. Any hints/fixes?

Thank you!
Andrey

.sage after the first and second runs:

sc_work@precell:~$ /home/sc_serv/sage/sage 
┌┐
│ SageMath version 8.1, Release Date: 2017-12-07 │
│ Type "notebook()" for the browser-based notebook interface.│
│ Type "help()" for help.│
└┘
Setting permissions of DOT_SAGE directory so only you can read and write it.
sage:   

  

Exiting Sage (CPU time 0m0.06s, Wall time 0m1.34s).
sc_work@precell:~$ ls -la .sage/
total 0
drwx-- 1 sc_work sagecell  80 Mar  5 02:00 .
drwxr-xr-x 1 sc_work sagecell 110 Mar  5 02:00 ..
drwxr-xr-x 1 sc_work sagecell  94 Mar  5 02:00 cache
drwxr-xr-x 1 sc_work sagecell   0 Mar  5 02:00 db
drwxr-xr-x 1 sc_work sagecell  74 Mar  5 02:00 ipython-5.0.0
drwxr-xr-x 1 sc_work sagecell   0 Mar  5 02:00 matplotlib-1.5.1
drwxr-xr-x 1 sc_work sagecell  14 Mar  5 02:00 temp
sc_work@precell:~$ /home/sc_serv/sage/sage 
┌┐
│ SageMath version 8.1, Release Date: 2017-12-07 │
│ Type "notebook()" for the browser-based notebook interface.│
│ Type "help()" for help.│
└┘
sage:   

  

Exiting Sage (CPU time 0m0.04s, Wall time 0m1.14s).
sc_work@precell:~$ ls -la .sage/
total 0
drwx-- 1 sc_work sagecell  82 Mar  5 02:00 .
drwxr-xr-x 1 sc_work sagecell 110 Mar  5 02:00 ..
drwxr-xr-x 1 sc_work sagecell  94 Mar  5 02:00 cache
drwxr-xr-x 1 sc_work sagecell   0 Mar  5 02:00 db
drwxr-xr-x 1 sc_work sagecell  74 Mar  5 02:00 ipython-5.0.0
drwxr-xr-x 1 sc_work sagecell   0 Mar  5 02:00 matplotlib-1.5.1
drwxr-xr-x 1 sc_work sagecell  26 Mar  5 02:00 R
drwxr-xr-x 1 sc_work sagecell  14 Mar  5 02:00 temp
sc_work@precell:~$

-- 
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] Orphaned kernels

[Andrey Novoseltsev]

Hello,

I've started using Jupyter notebooks, as this seems to be the bright 
future, and I am not impressed by handling restarted or crashed kernels: I 
have restarted a worksheet quite a few times today as I was changing 
modules it used and discovered now that there are ~30 kernels running after 
I shut down 2 worksheets and quit the notebook completely. With 16GB RAM it 
was not completely horrible yet, but on a weak laptop it likely would be. 
So to me there are 2 serious problems:
1) kernels are not killed/restarted properly leading to hanging processes
2) when the main notebook process quits there is no clean up of all started 
processes

I don't recall having such problems with SageNB, so presumably these 
problems are solvable ;-)

Andrey

-- 
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] trac not showing branch changes

[Andrey Novoseltsev]

I've seen it happen before a few times and now Ursula Whitcher reports it 
while reviewing https://trac.sagemath.org/ticket/22391 : if you click on 
the branch name it just shows that 3 files are deleted, which is scary and 
useless. Would be nice to know what is causing this issue and how to fix it!

Perhaps trac has to be rebooted/cleaned up or something since the next in 
the chain ticket https://trac.sagemath.org/ticket/22524 shows internal 
sever error when clicking on the branch.

-- 
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] How to configure MathJax for Jupyter notebook?

[Andrey Novoseltsev]

Specifically I want to enable color.js extension as was done here 
https://github.com/sagemath/sagenb/blob/master/sagenb/data/sage/js/mathjax_sage.js
 
so that https://trac.sagemath.org/ticket/23730 is resolved. But I have no 
idea which files have to be adjusted.

-- 
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: Plotting fans

[Andrey Novoseltsev]

On Friday, 1 September 2017 03:16:45 UTC-6, Carlo Verschoor wrote:
>
> Hello, I had this problem with plotting fans in Sage 8.0 which still works 
> in Sage 7.6. The problem breaks down to the following piece of code:
>
> cones = [Cone([(0,1),(1,0)]),Cone([(-2,-1),(-1,-3)]),Cone([(0,1),(-2,-1
> )]),Cone([(-1,-3),(1,0)])] 
> F = Fan(cones) 
> F.plot()
>
> I get an "unable to simplify to float approximation" error. So looking 
> at geometry/toric_plotter.py and plot/disk.py, one of the calculations it 
> could boil down to is: 
>
> float(arctan2(RDF(-3),RDF(-1)))
>
> Here the same error occurs. Due to the RDF ring, the function arctan2 now 
> returns a good value with a small imaginary part such that it cannot be 
> floated. 
>
> Hopefully, this issue can be resolved in some way.
>
> Carlo Verschoor.
>

Thank you for reporting and diagnosing, this is now 
https://trac.sagemath.org/ticket/23776 which I am going to fix over weekend 
unless someone does it faster!

-- 
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] Old-style packages

[Andrey Novoseltsev]

There is also "huge" package polytopes_db_4d which I do care about, but it 
is hard to find in listings, and given its 8GB size I don't download again 
and again but rather install locally. 

-- 
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] trac notifications

[Andrey Novoseltsev]

Hello,

I've checked my trac preferences and I have "Notify" for all events 
available. Although I cannot see *"Notify: Any ticket changes" *which is 
mentioned in the text around settings. I do not get notifications about 
closed tickets like this one https://trac.sagemath.org/ticket/21066 (I 
think I am getting all others, but maybe not and I just don't know) Is it 
possible that something is off with my trac settings that I cannot see and 
modify?

Thank you!
Andrey

-- 
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: Poll for issue G5 a specific guideline for writing docstrings

[Andrey Novoseltsev]

On Wednesday, 17 May 2017 15:05:55 UTC-6, Eric Gourgoulhon wrote:
>
> +1 (a list can have a single item anyway)
>
>
+1

I actually like how OUTPUT block is formatted in the same way as INPUT in 
HTML when a hyphen is used.

On a related note, I do not like allowing :param: etc directives since they 
are harder to read in plain text and look quite different from the list 
format when processed.

-- 
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: Poll for issue G4 a specific guideline for writing docstrings

[Andrey Novoseltsev]

+1

-- 
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] Poll for issue G3 a specific guideline for writing docstrings

[Andrey Novoseltsev]

-1

I prefer (3) or (4) or "Check" instead of "Test" and have no problems with 
mixing them.

-- 
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: Poll for issue G2 a specific guideline for writing docstrings

[Andrey Novoseltsev]

X

I think ``self`` should be allowed in a longer description and allowing it 
there but not in a one-liner is a bit confusing. So I'd rather not say 
anything about it and let people use their preferences.

-- 
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: Poll for issue G1 a specific guideline for writing docstrings

[Andrey Novoseltsev]

-1

-- 
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: Rant about plotting documentation

[Andrey Novoseltsev]

On Monday, 20 March 2017 08:09:31 UTC-6, kcrisman wrote:
>
>
>
> On Sunday, March 19, 2017 at 2:41:48 PM UTC-4, Andrey Novoseltsev wrote:
>>
>> I wanted to check how to make new threejs plotting code to use CDN. show? 
>> and plot? don't mention viewer options and their parameters. So, I go to 
>> the reference manual
>> http://doc.sagemath.org/html/en/reference/
>> and close to the top I see 3D Graphics
>> http://doc.sagemath.org/html/en/reference/plot3d/index.html 
>> <http://www.google.com/url?q=http%3A%2F%2Fdoc.sagemath.org%2Fhtml%2Fen%2Freference%2Fplot3d%2Findex.html=D=1=AFQjCNFhas5EMXd8qNRBQt0IiWkNxUM-3g>
>> which starts with introduction
>>
>> http://doc.sagemath.org/html/en/reference/plot3d/sage/plot/plot3d/examples.html
>> consisting of a single contrived example without any explanations!!! (Or 
>> output for that matter.)
>>
>> Next there are "Plotting functions"
>>
>> http://doc.sagemath.org/html/en/reference/plot3d/sage/plot/plot3d/plot3d.html
>> which look pretty at first glance because of colorful plots, but then I 
>> notice that there are also no opening words!
>>
>>
> A big job, to be sure; I didn't even know examples.py existed in that 
> directory.  Open a ticket and put "beginner" on it.  The problem is that 
> for much of our documentation, we are in the habit of making the 
> tab-completion where interesting documentation happens - e.g. 
> http://doc.sagemath.org/html/en/reference/plot3d/sage/plot/plot3d/plot3d.html#sage.plot.plot3d.plot3d.plot3d
>  
> for "plot3d".  But that isn't necessarily how someone "reads" the doc.
>

The ticket is https://trac.sagemath.org/ticket/22685 waiting for authors! 

-- 
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] Let R pick up system/user configuration files

[Andrey Novoseltsev]

I would like to tweak R interface a little: 
https://trac.sagemath.org/ticket/22682

Can someone who uses R please check it out and report if there are any 
problems?

There are 2 changes:
1) Allow picking up configuration files. This should not affect those who 
don't have these files, obviously, while for those who do I hope it will do 
more good then harm, in particular it will allow configuring graphic 
preferences.
2) Do not force PNG device in the interface, which overrides anything 
coming from configuration files. This potentially may cause problems, but 
for me things seem either to work fine, or be broken even before any 
changes.

While I don't use R, my motivation is to converge Sage master with the 
branch used by SageMathCell - this is one of 3 remaining differences.

Thank you!
Andrey

-- 
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: Default display for equations in notebook

[Andrey Novoseltsev]

"implemented on the Sage side" as opposed to in notebooks? I would very 
much be in favour of this so that the difference between code behaviour in 
different environments is as small as possible!

-- 
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] Rant about plotting documentation

[Andrey Novoseltsev]

I wanted to check how to make new threejs plotting code to use CDN. show? 
and plot? don't mention viewer options and their parameters. So, I go to 
the reference manual
http://doc.sagemath.org/html/en/reference/
and close to the top I see 3D Graphics
http://doc.sagemath.org/html/en/reference/plot3d/index.html
which starts with introduction
http://doc.sagemath.org/html/en/reference/plot3d/sage/plot/plot3d/examples.html
consisting of a single contrived example without any explanations!!! (Or 
output for that matter.)

Next there are "Plotting functions"
http://doc.sagemath.org/html/en/reference/plot3d/sage/plot/plot3d/plot3d.html
which look pretty at first glance because of colorful plots, but then I 
notice that there are also no opening words!

Since I know very well what I am looking for, I now click on "Three.js 
JavaScript WebGL Renderer" (and realize that I want a newer version from 
develop ;-), but how can a new user find out about different output options 
and their parameters???

Very much hope that someone will care enough to improve the situation...

-- 
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] Why is patchbot testing closed tickets?

[Andrey Novoseltsev]

This one
https://trac.sagemath.org/ticket/20940
was closed 5 months ago, yet as can be seen from
https://patchbot.sagemath.org/ticket/20940/
it was tested again and again and again like 70 times after the few ones 
that were useful. That's a lot of wasted computing power for other tickets 
and just electricity... I don't know how common is this situation, but it 
definitely makes me less eager to run a patchbot!

Thank you,
Andrey

-- 
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: threejs as standard package

[Andrey Novoseltsev]

Trying to get back to threejs:

- it is apparently possible to use HTTPS for rawgit.com and it has a CDN 
which we actually must use to avoid being blocked
- "more usual CDNs" provide threejs, but not extra files like OrbitControl
- it seems that the recommended way of using threejs is to download and 
serve your own copy

So to me "the right way" seems to be:

- have threejs package in Sage as originally proposed, stripping as much as 
we want (I guess it is perfectly fine to keep just the minified version and 
OrbitControl, adding more stuff later on if it becomes necessary)
- create scenes that can be downloaded/embedded without any explicit code 
loading threejs
- have clear instructions in documentation that one needs to load it and 
have an example, e.g. with https://cdn.rawgit.com

-- 
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] phc installation problems

[Andrey Novoseltsev]

Hello,

I am getting the following error when trying to install phc optional 
package from a script:

...
Error: package 'phc' not found
Assuming it is an old-style package... (this is deprecated: use -p instead 
of -i to install old-style packages)

Found package phc in /home/sc_serv/sage/upstream/phc-2.3.65.p0.spkg
phc-2.3.65.p0

Extracting package /home/sc_serv/sage/upstream/phc-2.3.65.p0.spkg
-rw-r--r-- 1 sc_serv sagecell 21614221 Oct 24 05:41 
/home/sc_serv/sage/upstream/phc-2.3.65.p0.spkg
Finished extraction
/home/sc_serv/sage/build/bin/sage-spkg: line 628: cd: src: No such file or 
directory
No patch files found in ../patches
Error: There is no spkg-install script, no setup.py, and no configure
script, so I do not know how to install 
/home/sc_serv/sage/upstream/phc-2.3.65.p0.spkg.
...

When I try the same command from a terminal, it proceeds fine, because 
apparently only the first one fails. On a freshly built Sage-7.5.rc1 I get:

...
novoselt@sage:~/sage$ sage -i -y phc
make -j8 build/make/Makefile
make[1]: warning: -jN forced in submake: disabling jobserver mode.
make[1]: Entering directory '/home/novoselt/sage'
make[1]: 'build/make/Makefile' is up to date.
make[1]: Leaving directory '/home/novoselt/sage'
build/bin/sage-logger \
"cd build/make && ./install 'all-toolchain'" logs/install.log
Nothing to (re)build / all up-to-date.

Error: package 'phc' not found
Assuming it is an old-style package... (this is deprecated: use -p instead 
of -i to install old-style packages)

Attempting to download package phc
>>> Checking online list of optional packages.
>>> Found phc-2.3.65.p0
>>> Trying to download 
http://mirrors.mit.edu/sage//spkg/optional/phc-2.3.65.p0.spkg
[..]
phc-2.3.65.p0

Extracting package /home/novoselt/sage/upstream/phc-2.3.65.p0.spkg
-rw-r--r-- 1 novoselt novoselt 21614221 Jan  1 13:47 
/home/novoselt/sage/upstream/phc-2.3.65.p0.spkg
Finished extraction
/home/novoselt/sage/build/bin/sage-spkg: line 628: cd: src: No such file or 
directory
No patch files found in ../patches
Error: There is no spkg-install script, no setup.py, and no configure
script, so I do not know how to install 
/home/novoselt/sage/upstream/phc-2.3.65.p0.spkg.
novoselt@sage:~/sage$ sage -i -y phc
make -j8 build/make/Makefile
make[1]: warning: -jN forced in submake: disabling jobserver mode.
make[1]: Entering directory '/home/novoselt/sage'
make[1]: 'build/make/Makefile' is up to date.
make[1]: Leaving directory '/home/novoselt/sage'
build/bin/sage-logger \
"cd build/make && ./install 'all-toolchain'" logs/install.log
Nothing to (re)build / all up-to-date.

Error: package 'phc' not found
Assuming it is an old-style package... (this is deprecated: use -p instead 
of -i to install old-style packages)

Found package phc in /home/novoselt/sage/upstream/phc-2.3.65.p0.spkg
phc-2.3.65.p0

Extracting package /home/novoselt/sage/upstream/phc-2.3.65.p0.spkg
-rw-r--r-- 1 novoselt novoselt 21614221 Jan  1 13:47 
/home/novoselt/sage/upstream/phc-2.3.65.p0.spkg
Finished extraction
/home/novoselt/sage/build/bin/sage-spkg: line 628: cd: src: No such file or 
directory
No patch files found in ../patches

Host system:
Linux sage 3.16.0-4-amd64 #1 SMP Debian 3.16.36-1+deb8u1 (2016-09-03) 
x86_64 GNU/Linux

C compiler: gcc
C compiler version:
Using built-in specs.
COLLECT_GCC=/usr/bin/gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/4.9/lto-wrapper
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Debian 4.9.2-10' 
--with-bugurl=file:///usr/share/doc/gcc-4.9/README.Bugs 
--enable-languages=c,c++,java,go,d,fortran,objc,obj-c++ --prefix=/usr 
--program-suffix=-4.9 --enable-shared --enable-linker-build-id 
--libexecdir=/usr/lib --without-included-gettext --enable-threads=posix 
--with-gxx-include-dir=/usr/include/c++/4.9 --libdir=/usr/lib --enable-nls 
--with-sysroot=/ --enable-clocale=gnu --enable-libstdcxx-debug 
--enable-libstdcxx-time=yes --enable-gnu-unique-object 
--disable-vtable-verify --enable-plugin --with-system-zlib 
--disable-browser-plugin --enable-java-awt=gtk --enable-gtk-cairo 
--with-java-home=/usr/lib/jvm/java-1.5.0-gcj-4.9-amd64/jre 
--enable-java-home 
--with-jvm-root-dir=/usr/lib/jvm/java-1.5.0-gcj-4.9-amd64 
--with-jvm-jar-dir=/usr/lib/jvm-exports/java-1.5.0-gcj-4.9-amd64 
--with-arch-directory=amd64 --with-ecj-jar=/usr/share/java/eclipse-ecj.jar 
--enable-objc-gc --enable-multiarch --with-arch-32=i586 --with-abi=m64 
--with-multilib-list=m32,m64,mx32 --enable-multilib --with-tune=generic 
--enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu 
--target=x86_64-linux-gnu
Thread model: posix
gcc version 4.9.2 (Debian 

[sage-devel] Re: threejs as standard package

[Andrey Novoseltsev]

On Wednesday, 28 December 2016 18:12:36 UTC-7, Paul Masson wrote:
>
>
>
> On Wednesday, December 28, 2016 at 3:47:53 PM UTC-8, Andrey Novoseltsev 
> wrote:
>>
>> On Tuesday, 27 December 2016 15:34:51 UTC-7, Paul Masson wrote:
>>>
>>> I'm very much in favor of making Three.js a standard package, with the 
>>> following caveats:
>>>
>>> 1) As of r80 the Three.js library has been reorganized to use ES6 
>>> Modules. There is little point in asking people to download the entire 
>>> library (even with examples excluded as in the current optional package) 
>>> unless we're going to make Node.js a standard package as well.
>>>
>>> 2) The template only needs two files to function: the main build file, 
>>> minified or not, and OrbitControls.js for allowing user interaction. As a 
>>> standard package I'm in favor of Three.js only including the parts of the 
>>> library that will actually be used. If another file is needed in the 
>>> future, it can be added then.
>>>
>>
>> If nobody objects and you can make a package taking care of these points 
>> I should be able to review it!
>>
>
> Should I rewrite the current optional package to download only the minimum 
> needed files, or leave it alone and create a new threejs_min package?
>

I am not sure what are uses of the current package beyond SageMathCell. If 
someone is aware of other cases, would be nice to know, otherwise I think 
it is fine to trim it as much as necessary to keep your new code 
operational. (For the record, the current size is 14M, so it would be nice 
if the standard package was a little smaller, although the long term plan 
is probably to make jmol optional and that will cut 37M.)

 
>
>>  
>>
>>>
>>> 3) As part of the review process of #12402, Andrey asked 
>>> <https://trac.sagemath.org/ticket/12402#comment:37> me explicitly about 
>>> the ease of embedding Three.js output in web pages. While I certainly want 
>>> to add a local copy of the required Three.js files for offline use, I'd 
>>> like to keep the output as portable as possible, and the simplest way to 
>>> achieve that is with a CDN link in the file. The template can be modified 
>>> to check for a local copy and fall back to the CDN or vice versa, but the 
>>> HTTPS issue would still need to be fixed on the server.
>>>   Alternately, a flag can be added to the viewer to specify that 
>>> generated output will be for online use and the template modified at 
>>> runtime to use the CDN rather than the local copy. That would of course 
>>> require an extra input from the end user, but may be preferable to the 
>>> majority.
>>>
>>
>> Well, if someone is going to embed plots into their webpage, presumably 
>> that someone is capable of following simple instructions like "add this 
>> script tag to the head". Note also that most of the time people will not 
>> save the output, so while it is important to support saving/embedding, 
>> proper showing is the first priority and at the moment SageNB does not work 
>> over HTTPS. Presumably there would be no problems if scripts were served 
>> from local installation.
>>
>
> I'm not understanding what the problem is with SageNB: the Three.js viewer 
> already works just fine in the legacy notebook. What issue is there with 
> downloading an external script securely?
>

It is not working: I have SageNB running over HTTPS and using threejs shows 
just blank area and messages in console about mixed content blocked. My 
understanding is that if you are loading a page over HTTPS it should not 
requests components over HTTP or browsers will complain or just show 
nothing. There is no problem in the opposite case, so perhaps you can just 
always use secure CDN.

 
>
>>
>> On Monday, December 26, 2016 at 7:16:55 PM UTC-8, Andrey Novoseltsev 
>> wrote:
>>>
>>> Hello,
>>>
>>> How about making threejs a standard package?
>>>
>>> It was optional for a while, used in SageMathCell to power its own 
>>> version of threejs viewer. https://trac.sagemath.org/ticket/12402 has 
>>> added threejs as possible output for a bunch of backends and ideally it 
>>> will become standard for all interfaces. One of the problems now - the 
>>> template loads scripts from the Internet, which means no offline use and 
>>> causes issues with HTTPS anyway.
>>>
>>> Thank you,
>>> Andrey
>>>
>>

-- 
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: threejs as standard package

[Andrey Novoseltsev]

On Tuesday, 27 December 2016 15:34:51 UTC-7, Paul Masson wrote:
>
> I'm very much in favor of making Three.js a standard package, with the 
> following caveats:
>
> 1) As of r80 the Three.js library has been reorganized to use ES6 Modules. 
> There is little point in asking people to download the entire library (even 
> with examples excluded as in the current optional package) unless we're 
> going to make Node.js a standard package as well.
>
> 2) The template only needs two files to function: the main build file, 
> minified or not, and OrbitControls.js for allowing user interaction. As a 
> standard package I'm in favor of Three.js only including the parts of the 
> library that will actually be used. If another file is needed in the 
> future, it can be added then.
>

If nobody objects and you can make a package taking care of these points I 
should be able to review it!
 

>
> 3) As part of the review process of #12402, Andrey asked 
> <https://trac.sagemath.org/ticket/12402#comment:37> me explicitly about 
> the ease of embedding Three.js output in web pages. While I certainly want 
> to add a local copy of the required Three.js files for offline use, I'd 
> like to keep the output as portable as possible, and the simplest way to 
> achieve that is with a CDN link in the file. The template can be modified 
> to check for a local copy and fall back to the CDN or vice versa, but the 
> HTTPS issue would still need to be fixed on the server.
>   Alternately, a flag can be added to the viewer to specify that generated 
> output will be for online use and the template modified at runtime to use 
> the CDN rather than the local copy. That would of course require an extra 
> input from the end user, but may be preferable to the majority.
>

Well, if someone is going to embed plots into their webpage, presumably 
that someone is capable of following simple instructions like "add this 
script tag to the head". Note also that most of the time people will not 
save the output, so while it is important to support saving/embedding, 
proper showing is the first priority and at the moment SageNB does not work 
over HTTPS. Presumably there would be no problems if scripts were served 
from local installation.

On Monday, December 26, 2016 at 7:16:55 PM UTC-8, Andrey Novoseltsev wrote:
>
> Hello,
>
> How about making threejs a standard package?
>
> It was optional for a while, used in SageMathCell to power its own version 
> of threejs viewer. https://trac.sagemath.org/ticket/12402 has added 
> threejs as possible output for a bunch of backends and ideally it will 
> become standard for all interfaces. One of the problems now - the template 
> loads scripts from the Internet, which means no offline use and causes 
> issues with HTTPS anyway.
>
> Thank you,
> Andrey
>

-- 
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] threejs as standard package

[Andrey Novoseltsev]

Hello,

How about making threejs a standard package?

It was optional for a while, used in SageMathCell to power its own version 
of threejs viewer. https://trac.sagemath.org/ticket/12402 has added threejs 
as possible output for a bunch of backends and ideally it will become 
standard for all interfaces. One of the problems now - the template loads 
scripts from the Internet, which means no offline use and causes issues 
with HTTPS anyway.

Thank you,
Andrey

-- 
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: Troubles communicating with Sage 7.4

[Andrey Novoseltsev]

On Sunday, 23 October 2016 14:23:10 UTC-6, Andrey Novoseltsev wrote:
>
> That's a stale ticket that would allow one to turn of the prompt, but the 
> real problem is "communicating with Sage"  - I am trying to provide an 
> answer to the question by sending a newline through the pipe, but it is 
> ignored.
>

Apparently this is due to
...
read -p "Are you sure you want to continue [Y/n]? " answer < /dev/tty > 
/dev/tty 2>&1 
...
In my humble but strong opinion command line applications should respect 
standard redirection/pipes...

-- 
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: Troubles communicating with Sage 7.4

[Andrey Novoseltsev]

Yet another possibly related problem: when installing qhull from a script 
it errors out with
...
cd lib && ln -f -s libqhull_r.so.7.2.0 libqhull_r.so


===
== If user_eg fails to link on MinGW or Cygwin, use
==   "make SO=dll" and copy lib/libqhull_r.dll to bin/
== Otherwise if user_eg fails to link, switch to -lqhullstatic_r
===

gcc -o bin/user_eg src/user_eg/user_eg_r.o -O3 -ansi -Isrc -fpic -Wall 
-Wcast-qual -Wextra -Wwrite-strings -Wshadow  -Llib -lqhull_r -lm
bin/qconvex
QH6075 qhull input error: short input file.  Did not find dimension and 
number of points

While executing:  | qconvex
Options selected for Qhull 2015.2 2016/01/18:
  run-id 1314731036
Makefile:323: recipe for target 'qconvex-prompt' failed
make[2]: *** [qconvex-prompt] Error 1
make[2]: Leaving directory 
'/home/sc_serv/sage/local/var/tmp/sage/build/qhull-2015-src-7.2.0/src'
Error building qhull
...

If I try to install qhull after that in an interactive terminal everything 
is OK. The line "While executing:  | qconvex" suggests that perhaps there 
are again issues with pipes.

-- 
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: Cannot run sage in postinst

[Andrey Novoseltsev]

On Sunday, 23 October 2016 03:23:12 UTC-6, Jan Groenewald wrote:
>
> Hi
>
> I'm preparing the PPA for sage 7.4. In the postinst script I run sage once 
> as root to set hardcoded paths (which may have been replaced by 
> replocate-once.py??).
>
> It fails to run under apt or dpkg --configure with 
>
> /usr/lib/sagemath/local/lib/python2.7/site-packages/prompt_toolkit/terminal/vt100_input.pyc
>  
> in __init__(self= .terminal.vt100_input.raw_mode object>, fileno=0)
> --> 409 self.attrs_before = termios.tcgetattr(fileno)
> error: (25, 'Inappropriate ioctl for device')
>

Looks like the same problem as in 
https://groups.google.com/d/topic/sage-devel/5GXnV4tlMXw/discussion with a 
fix for some situations already created. 


> But I can run sage in a terminal. Is it safe to remove 'running sage once 
> as root to set hardcoded paths' from the postinst?
>
> Regards,
> Jan
>
> -- 
>   .~. 
>   /V\ Jan Groenewald
>  /( )\www.aims.ac.za
>  ^^-^^ 
>

-- 
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: Troubles communicating with Sage 7.4

[Andrey Novoseltsev]

On Sunday, 23 October 2016 01:18:58 UTC-6, Matthias Koeppe wrote:
>
>
>
> On Saturday, October 22, 2016 at 10:52:20 PM UTC-7, Andrey Novoseltsev 
> wrote:
>>
>>  I have further problems (now with 7.5beta0 + #21743), sending new line 
>> during installation of an experimental package (cryptominisat) is 
>> ignored here:
>>
>> https://github.com/sagemath/sagecell/blob/master/contrib/vm/container_manager.py#L461
>> The terminal got stuck on the [Y/n] question and then continued when I 
>> pressed Enter myself. Any clue or solution for this one?
>>
>
> https://trac.sagemath.org/ticket/21082 
>

That's a stale ticket that would allow one to turn of the prompt, but the 
real problem is "communicating with Sage"  - I am trying to provide an 
answer to the question by sending a newline through the pipe, but it is 
ignored.

-- 
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: Troubles communicating with Sage 7.4

[Andrey Novoseltsev]

On Saturday, 22 October 2016 13:49:57 UTC-6, Volker Braun wrote:
>
> This is now https://trac.sagemath.org/ticket/21743
>

Thanks a lot Volker!

That fixed the crash, but I have further problems (now with 7.5beta0 + 
#21743), sending new line during installation of an experimental package 
(cryptominisat) 
is ignored here:
https://github.com/sagemath/sagecell/blob/master/contrib/vm/container_manager.py#L461
The terminal got stuck on the [Y/n] question and then continued when I 
pressed Enter myself. Any clue or solution for this one?


>
> On Saturday, October 22, 2016 at 5:01:06 AM UTC+2, Andrey Novoseltsev 
> wrote:
>>
>> Hello,
>>
>> I need help figuring out what's going on when I am preparing Sage for 
>> SageMathCell.
>>
>> When the script is at
>>
>> https://github.com/sagemath/sagecell/blob/master/contrib/vm/container_manager.py#L445
>> trying to pipe some commands into a freshly compiled Sage it crashes. The 
>> crash report is attached and the problem seems to be with a wrong or absent 
>> terminal (given that it is done by a script in screen).
>>
>> After the build there is "Sage starts OK" message and if I try to connect 
>> via interactive terminal and execute commands by hand everything is fine. 
>> Exactly the same setup was also going without crashes for 7.3, so it is 
>> some recent change/upgrade.
>>
>> Thank you!
>> Andrey
>>
>

-- 
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] Troubles communicating with Sage 7.4

[Andrey Novoseltsev]

Hello,

I need help figuring out what's going on when I am preparing Sage for 
SageMathCell.

When the script is at
https://github.com/sagemath/sagecell/blob/master/contrib/vm/container_manager.py#L445
trying to pipe some commands into a freshly compiled Sage it crashes. The 
crash report is attached and the problem seems to be with a wrong or absent 
terminal (given that it is done by a script in screen).

After the build there is "Sage starts OK" message and if I try to connect 
via interactive terminal and execute commands by hand everything is fine. 
Exactly the same setup was also going without crashes for 7.3, so it is 
some recent change/upgrade.

Thank you!
Andrey

-- 
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.
***

IPython post-mortem report

{'commit_hash': u'1ba246d',
 'commit_source': 'installation',
 'default_encoding': 'UTF-8',
 'ipython_path': '/home/sc_serv/sage/local/lib/python2.7/site-packages/IPython',
 'ipython_version': '5.0.0',
 'os_name': 'posix',
 'platform': 'Linux-4.4.0-36-generic-x86_64-with-debian-stretch-sid',
 'sys_executable': '/home/sc_serv/sage/local/bin/python',
 'sys_platform': 'linux2',
 'sys_version': '2.7.10 (default, Oct 20 2016, 05:24:29) \n[GCC 5.4.0 
20160609]'}

***



***

Crash traceback:

---
---
errorPython 2.7.10: /home/sc_serv/sage/local/bin/python
   Thu Oct 20 06:21:36 2016
A problem occurred executing Python code.  Here is the sequence of function
calls leading up to the error, with the most recent (innermost) call last.
/home/sc_serv/sage/src/bin/sage-ipython in ()
  1 #!/usr/bin/env python
  2 # -*- coding: utf-8 -*-
  3 """
  4 Sage IPython startup script.
  5 """
  6 
  7 from sage.repl.interpreter import SageTerminalApp
  8 
  9 app = SageTerminalApp.instance()
 10 app.initialize()
---> 11 app.start()
global app.start = >

/home/sc_serv/sage/local/lib/python2.7/site-packages/IPython/terminal/ipapp.py 
in start(self=)
333 if self.log_level <= logging.INFO: print()
334 
335 def _pylab_changed(self, name, old, new):
336 """Replace --pylab='inline' with --pylab='auto'"""
337 if new == 'inline':
338 warnings.warn("'inline' not available as pylab backend, "
339   "using 'auto' instead.")
340 self.pylab = 'auto'
341 
342 def start(self):
343 if self.subapp is not None:
344 return self.subapp.start()
345 # perform any prexec steps:
346 if self.interact:
347 self.log.debug("Starting IPython's mainloop...")
--> 348 self.shell.mainloop()
self.shell.mainloop = >
349 else:
350 self.log.debug("IPython not interactive...")
351 
352 def load_default_config(ipython_dir=None):
353 """Load the default config file from the default ipython_dir.
354 
355 This is useful for embedded shells.
356 """
357 if ipython_dir is None:
358 ipython_dir = get_ipython_dir()
359 
360 profile_dir = os.path.join(ipython_dir, 'profile_default')
361 
362 config = Config()
363 for cf in Application._load_config_files("ipython_config", 
path=profile_dir):

/home/sc_serv/sage/local/lib/python2.7/site-packages/IPython/terminal/interactiveshell.py
 in mainloop(self=, 
display_banner=)
387 if (not self.confirm_exit) \
388 or self.ask_yes_no('Do you really want to exit 
([y]/n)?','y','n'):
389 self.ask_exit()
390 
391 else:
392 if code:
393 self.run_cell(code, store_history=True)
394 
395 def mainloop(self, display_banner=DISPLAY_BANNER_DEPRECATED):
396 # An extra layer of protection in case someone mashing Ctrl-C 
breaks
397 # out of our internal code.
398 if display_banner is not DISPLAY_BANNER_DEPRECATED:
399 warn('mainloop `display_banner` argument is deprecated 
since IPython 5.0. Call `show_banner()` if needed.', DeprecationWarning, 
stacklevel=2)

[sage-devel] DeprecationWarning stacklevel issues

[Andrey Novoseltsev]

Consider the following code

def g():
function("f", x)
html(04)

g()

which in SageMathCell gives the following warning messages (in Sage itself 
and SageMathCloud everything is even more useless without using an extra 
module due to 
https://groups.google.com/d/topic/sage-devel/-KiMrG--5x8/discussion ):

sagemathcell.py:5: DeprecationWarning: Calling function('f',x) is deprecated. 
Use function('f')(x) instead.
See http://trac.sagemath.org/17447 for details.
  g()
sagemathcell.py:5: DeprecationWarning: use 0o as octal prefix instead of 0
If you do not want this number to be interpreted as octal, remove the leading 
zeros.
See http://trac.sagemath.org/17413 for details.
  g()
sagemathcell.py:3: DeprecationWarning:  html(...) will change soon to return 
HTML instead of printing it. Instead use pretty_print(html(...)) for strings or 
just pretty_print(...) for math. 
See http://trac.sagemath.org/18292 for details.
  html(04)


The problem is - the first two messages go too far in the call stack and 
complain about a wrong piece of code, which is very confusing in longer 
examples. It may have something to do with Cython involved in dealing with 
symbolic functions and numbers.

The question is - what is the proper way to fix it??? Using different 
stacklevels on a case-by-case basis is too error-prone, so hopefully 
something automatic can be done in the deprecation function itself.

Thank you!
Andrey

-- 
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-cell] Re: [sage-devel] Sage cells in interact wiki

[Andrey Novoseltsev]

On Monday, 19 September 2016 18:59:09 UTC-6, Jason Grout wrote:
>
> Some ideas:
>
> Is the MoinMoin sage cell extension enabled? 
> https://github.com/sagemath/sagecell/blob/master/contrib/moinmoin/sagecell.py.
>  If 
> it is enabled, was MoinMoin updated and the extension no longer works?
>
> Thanks,
>
> Jason
>

Thanks for the pointer, Jason! Cells are created now but it would be great 
to go over interacts to fix deprecated and broken things or report new bugs.


>
> On Mon, Sep 19, 2016 at 8:48 PM William Stein <wst...@gmail.com 
> > wrote:
>
>> On Mon, Sep 19, 2016 at 1:46 PM, Andrey Novoseltsev <novo...@gmail.com 
>> > wrote:
>>
>>> On Monday, 19 September 2016 10:53:27 UTC-6, William wrote:
>>>>
>>>> This seems to be an absolutely *massive* issue, which I'm sure was 
>>>> caused by some API change by Andrey.  These wiki pages have worked fine 
>>>> for 
>>>> about 7 years, so it's annoying that they are all suddenly broken. 
>>>>
>>>> > I guess it was meant to use the Sage cell server to make the 
>>>> examples actually work but it doesn't do that anymore.
>>>>
>>>> They did actually work (and well) until very, very recently.
>>>>
>>>
>>> Do you have a better idea of how recently? Week/month/several months? 
>>>
>>
>> Hmm.  This could also coincide with when Mike Hansen moved the wiki to 
>> GCE.
>>  
>>
>>>
>>> Looking at the page source, I do not understand how embedded_sagecell.js 
>>> is supposed to load and there are no requests for it reported by browser. 
>>> Console complains about $, so something is off with jQuery, I guess. For 
>>> the record there is no need to load/use it with sagecell.makeSagecell at 
>>> all. And as "official API" that I am trying hard not to break (or at least 
>>> fix when I do) I consider 
>>> https://github.com/sagemath/sagecell/blob/master/doc/embedding.rst
>>>
>>> So - it does not work indeed, but I have no idea why and I hope that I 
>>> am not responsible ;-)
>>>
>>
>> I now think that you are not responsible at all.  Thanks for the hints 
>> above!
>>  
>>
>>>
>>> Thank you!
>>> Andrey
>>>
>>>
>>>>  -- William
>>>>
>>>> On Mon, Sep 19, 2016 at 6:04 AM, Jeroen Demeyer <jdem...@cage.ugent.be> 
>>>> wrote:
>>>>
>>>>> The "interact" wiki page
>>>>> https://wiki.sagemath.org/interact
>>>>> is seriously broken: I guess it was meant to use the Sage cell server 
>>>>> to make the examples actually work but it doesn't do that anymore. What 
>>>>> is 
>>>>> worse: also the source code is not shown, making the examples almost 
>>>>> totally useless...
>>>>>
>>>>> Well, the Sage source code is still in the source of the wiki page, 
>>>>> but that's not very practical. However, there seem to be some 
>>>>> backslash-escaping going on, making the examples not 100% copy-pastable.
>>>>>
>>>>> Some context: I am working on an implementation of the SageNB 
>>>>> interacts in Jupyter at https://trac.sagemath.org/ticket/21267 and 
>>>>> that page is a valuable source of examples for me.
>>>>>
>>>>> -- 
>>>>> 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.
>>>>>
>>>>
>>>>
>>>>
>>>> -- 
>>>> William (http://wstein.org)
>>>>
>>> -- 
>>> You received this message because you are subscribed to the Google 
>>> Groups "sage-cell" group.
>>> To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to sage-cell+...@googlegroups.com .
>>> To view this discussion on the web visit 
>>> https://groups.google.com/d/msgid/sage-cell/87a9ae7b-5ea6-4a7f-b435-5eced2efe633%40googlegroups.com
>>>  
>>> <ht

Re: [sage-cell] Re: [sage-devel] Sage cells in interact wiki

[Andrey Novoseltsev]

On Monday, 19 September 2016 18:59:08 UTC-6, Jason Grout wrote:
>
> Some ideas:
>
> Is the MoinMoin sage cell extension enabled? 
> https://github.com/sagemath/sagecell/blob/master/contrib/moinmoin/sagecell.py.
>  If 
> it is enabled, was MoinMoin updated and the extension no longer works?
>

While I think the old version should work if it is enabled (no clue about 
this stuff), it is not necessary to load jQuery, we do not promote aleph 
name, and HTTPS is better. I've made some changes here and hope that they 
will work: 
https://github.com/sagemath/sagecell/commit/b3f3985c6286f745d92e677acf3a17314c7087ce


> Thanks,
>
> Jason
>
>

-- 
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 cells in interact wiki

[Andrey Novoseltsev]

On Monday, 19 September 2016 10:53:27 UTC-6, William wrote:
>
> This seems to be an absolutely *massive* issue, which I'm sure was caused 
> by some API change by Andrey.  These wiki pages have worked fine for about 
> 7 years, so it's annoying that they are all suddenly broken. 
>
> > I guess it was meant to use the Sage cell server to make the examples 
> actually work but it doesn't do that anymore.
>
> They did actually work (and well) until very, very recently.
>

Do you have a better idea of how recently? Week/month/several months? 

Looking at the page source, I do not understand how embedded_sagecell.js is 
supposed to load and there are no requests for it reported by browser. 
Console complains about $, so something is off with jQuery, I guess. For 
the record there is no need to load/use it with sagecell.makeSagecell at 
all. And as "official API" that I am trying hard not to break (or at least 
fix when I do) I consider 
https://github.com/sagemath/sagecell/blob/master/doc/embedding.rst

So - it does not work indeed, but I have no idea why and I hope that I am 
not responsible ;-)

Thank you!
Andrey


>  -- William
>
> On Mon, Sep 19, 2016 at 6:04 AM, Jeroen Demeyer  > wrote:
>
>> The "interact" wiki page
>> https://wiki.sagemath.org/interact
>> is seriously broken: I guess it was meant to use the Sage cell server to 
>> make the examples actually work but it doesn't do that anymore. What is 
>> worse: also the source code is not shown, making the examples almost 
>> totally useless...
>>
>> Well, the Sage source code is still in the source of the wiki page, but 
>> that's not very practical. However, there seem to be some 
>> backslash-escaping going on, making the examples not 100% copy-pastable.
>>
>> Some context: I am working on an implementation of the SageNB interacts 
>> in Jupyter at https://trac.sagemath.org/ticket/21267 and that page is a 
>> valuable source of examples for me.
>>
>> -- 
>> 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.
>>
>
>
>
> -- 
> William (http://wstein.org)
>

-- 
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: Trac notifications again

[Andrey Novoseltsev]

On Friday, 12 August 2016 02:35:23 UTC-6, Erik Bray wrote:
>
> On Thu, Aug 11, 2016 at 11:35 PM, Andrey Novoseltsev <novo...@gmail.com 
> > wrote: 
> > On Wednesday, 27 July 2016 15:35:29 UTC-6, leif wrote: 
> >> 
> >> Orthogonal to that, but also annoying is that we do no longer get 
> >> notifications when a ticket gets closed. 
> >> 
> >> Is this a bug or a feature? 
> >> 
> >> 
> >> -leif 
> >> 
> > 
> > Ping? I definitely like getting notifications that a ticket is closed, 
> > hopefully it can e enabled again. 
>
> I'm not sure what the issue is here--I get notifications when tickets 
> are closed.  Does anyone else? 
>
> You might need to play around with your notification settings on 
> https://trac.sagemath.org/prefs/notification


I didn't even know that these settings exist and I never felt a need to 
adjust them - when there is a change on a ticket I am involved with, I'd 
like to get an email. My problem ticket was #19430  - I noticed that it is 
merged by reading 7.4.beta0 release note. I'm a regular user on trac.


>
> Are you an admin on the site?  I wonder if somehow people without the 
> TRAC_ADMIN permission aren't getting ticket closed notifications.  I 
> don't know why that would be the case but it's the only possible 
> difference other than personal notification settings that I can think 
> of at this time. 
>
> Best, 
> Erik 
>

-- 
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: Trac notifications again

[Andrey Novoseltsev]

On Wednesday, 27 July 2016 15:35:29 UTC-6, leif wrote:
>
> Orthogonal to that, but also annoying is that we do no longer get 
> notifications when a ticket gets closed. 
>
> Is this a bug or a feature? 
>
>
> -leif 
>
>
Ping? I definitely like getting notifications that a ticket is closed, 
hopefully it can e enabled again. 

-- 
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: Trac attachment limit

[Andrey Novoseltsev]

On Tuesday, 26 July 2016 20:13:56 UTC-6, William wrote:
>
> On Tue, Jul 26, 2016 at 1:04 PM, Paul Masson  > wrote: 
> > What about a "packages" repository on GitHub just to store files? 
>
> I was also going to suggest that.  A drawback is we would have to add 
> all trac users to the Github repo, so hundreds and hundreds of people, 
> many probably not even on github right now. So the overhead of a 
> common repo that everyone uses would be prohibitive. 
>
> Of course, people could just put their own packages in a repo and post 
> a link to that.  I assume this was already obvious to Andrey though, 
> and he considers it to be too much of a pain. 
>
>  -- William 
>
>  
Actually, it was not (in my defence I didn't give it much sort). What I 
used to do was copy it to sage.math.washington.edu and post a link (which 
was conveniently wget-able). It was obvious and easy to do. Uploading a 
file to trac would be even easier and more obvious and would not require 
any extra setup/thinking/knowledge on my part. It also would be so obvious, 
that there would be no need to even document it somehow. Any other 
service/solution needs to be explained in the documentation guide and kept 
up-to-date when something changes.

So personally I really really prefer being able to upload bigger files to 
trac directly from the ticket page. What's the point of backing up 
attachments there forever anyway? With mercurial attachments were patches. 
Now I think they tend to be some screenshots or log files or something 
which is not quite as crucial. With disk snapshots there is a fallback and 
a more resilient backup can be concerned only with what people have typed, 
i.e. the code and discussions themselves. Perhaps it is also possible just 
keep the current 2MB limits for picking attachments that have to be backed 
up in case we do care about screenshots.

Thank you,
Andrey

-- 
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: Trac attachment limit

[Andrey Novoseltsev]

On Monday, 25 July 2016 08:53:48 UTC-6, leif wrote:
>
> In the past we were told to not attach *any* binary files because they 
> would end up in some (VCS?) repo where they would remain forever even 
> after deletion on trac. 
>
> I couldn't find any reference to that in our current documentation 
> though.  (I'm pretty sure there used to be a note on that in the Sage 
> trac guidelines, or at least on the main page of the Sage trac wiki, and 
> at some point we had spkg-upload.googlecode.com for exactly that 
> purpose, at least for those without a sage.math cluster account, when 
> everybody's home folder was accessible via HTTP.) 
>
> Have not heard of it previously and it recommends to email to Minh who I 
think has moved on since then. With UW cluster gone/going, what is then the 
recommended way to post packages like 36MB jmol from 
https://trac.sagemath.org/ticket/19430 ?

I've tried to post a link to SageMathCloud - while it was possible, it 
required installing keys into a project and then copying it over SSH since 
the file is too large for the web interface. The resulting link has to be 
handled by a human, since it is not a direct link. I posted an alternative 
to Dropbox which also seem to require extra clicks and interaction. I 
imagine this would be common for many other services. Creating my own site 
just for the sake of posting a package (which I am doing once a year or 
less) is a bit of an overkill.

-- 
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] Trac attachment limit

[Andrey Novoseltsev]

Is there any deep reason to 2MB attachment limit on Trac? Can it be 
increased to, say, 100MB? Would be convenient to be able to upload new 
packages there rather than figuring out some place to post them.

Thank you,
Andrey

-- 
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] Dedicated group for SageMathCell

[Andrey Novoseltsev]

The group https://groups.google.com/forum/#!forum/sage-cell will now be 
used for important announcements regarding SageMathCell, issue reporting 
and any other relevant discussions. If you actively use SageMathCell and 
especially if you maintain web pages with embedded cells, please subscribe 
to it and detect/report regressions before your readers!

Thank you,
Andrey

P.S. I have already mentioned this group before, but it was rightly pointed 
out to me that the end of a long thread is not the best place for 
announcement.

-- 
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: Deprecation warnings are no longer printed?!

[Andrey Novoseltsev]

In 7.3.beta4 I get

sage:   sage: f1(x) = -1
sage:   sage: f2(x) = 2
sage:   sage: f = Piecewise([[(0,pi/2),f1],[(pi/2,pi),f2]])
/home/novoselt/sage/src/bin/sage-ipython:1: DeprecationWarning: use 
lower-case piecewise instead
See http://trac.sagemath.org/14801 for details.
  #!/usr/bin/env python

so warnings do work. I am also unclear which example fails for you. And if 
something is indeed wrong at 20667, please fix it, but NOT by calling 
resetwarnings any time after initialization!

On Wednesday, 29 June 2016 02:42:09 UTC-6, Johan S. R. Nielsen wrote:
>
> sage: import warnings 
> sage: warnings.filterwarnings('always',".*", DeprecationWarning) 
> sage: deprecation(3,"m") 
> Hi sage-devel, 
>
> On my Sage installation, as well as on David Lucas' (having different 
> Linux distros), the following *does not* print the expected deprecation 
> warning in the Sage shell or IPython: 
>
> In fact, I can't get any function to trigger the usual deprecation 
> warnings! Needless to say, this used to work rather recently, though I 
> didn't do a binary search in the commits yet. 
>
> I can do the following to get deprecation warnings back: 
>
>  
>
> I'm suspecting #20667 which fiddled with the way warnings are handled, 
> in particular modifying warnings.filters in src/sage/all.py. 
>
> Can anyone else confirm? 
>
> 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 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: What is the best time for Trac downtime?

[Andrey Novoseltsev]

On Sunday, 19 June 2016 20:19:50 UTC-6, William wrote:
>
> On Sun, Jun 19, 2016 at 4:09 AM, mmarco  
> wrote: 
> > Did you recently enable cloudflare cache services? 
> > 
> > I am using Tor to circunvent some pesky firewall, and when trying to 
> push 
> > some changes, I get some git_trac.trac_error.TracConnectionError: 
> Forbidden 
> > error. 
> > 
> > Would it be possible to disable cloudflare? Or at least, to use the 
> option 
> > to not blacklist Tor exit nodes, please? 
>
> As far as I can tell I have now disabled cloudflare for everything 
> trac related.  Does this help?   Cloudflare is doing nothing but DNS. 
> trac.sagemath.org is currently 128.208.160.253 
>
>  -- William 
>
>
I could not do "git fetch trac" (timeout, network unreachable) but since I 
saw somewhere that it may work 1 out of 10 times I kept trying while doing 
other stuff. No luck. Tried again after receiving your message and it 
worked right away. That's from our department wired network at the 
University of Alberta, highly unlikely to be blacklisted somehow.

Andrey

> 
> > El jueves, 16 de junio de 2016, 19:45:55 (UTC+2), Erik Bray escribió: 
> >> 
> >> Hi all, 
> >> 
> >> We're nearly ready to switch trac.sagemath.org over to its new server. 
> >> This will require roughly an hour for me to take down and dump/load 
> >> the database, as well as the Trac attachments (there are a lot!) and 
> >> to take down git.sagemath.org and transfer the git repository, then 
> >> bring everything back up again. 
> >> 
> >> Will also need to coordinate with William on transferring the DNS. 
> >> 
> >> I was thinking sometime over the weekend, but then a lot of Sage 
> >> developers are doing this on their spare time, which often comes on 
> >> weekends.  Sage-devel is an international community so there's 
> >> probably no one best time. 
> >> 
> >> Any thoughts on this? 
> >> 
> >> Thanks, 
> >> Erik 
> > 
> > -- 
> > 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. 
>
>
>
> -- 
> William (http://wstein.org) 
>

-- 
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 apps for iOS and Android

[Andrey Novoseltsev]

On Monday, 13 June 2016 14:58:54 UTC-6, Paul Masson wrote:
>
> If the error occurs in Mobile Safari, then it's not an issure with the app 
> but with the browser in general, since the app is basically an iOS 
> UIWebView.
>
> The only major change made to the web server was adding require.js, right? 
> There is an open issue in their GitHub repository about timeouts on iOS,
>
> https://github.com/requirejs/requirejs/issues/1234
>
> and here's a link to the documentation for waitSeconds:
>
> http://requirejs.org/docs/api.html#config-waitSeconds
>
> I don't know if you want to set that to zero, but you might try changing 
> it from the default.
>
> My own iOS devices are all updated to the latest version and work fine. 
> The problem appears to stem from loading the web page in older versions.
>
> Thanks for the pointer, I'll look into this, although I am skeptical that 
this is the issue - embedded_sagecell.js is the only JS script loaded at 
the moment and it includes RequireJS. So presumably when the library and 
its configuration start working the script is already in the browser. I 
guess MathJax may be loaded separately from CDN, but presumably its failure 
should not stop the rest of the stuff from working.

-- 
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 apps for iOS and Android

[Andrey Novoseltsev]

Android app and iOS one are still having issues, below is a detailed report 
from one of the users. I cannot test/debug these, so any help would be 
greatly appreciated or they will remain broken...

Thank you,
Andrey



I'm still having issues on the App... I also wrote to
the App developer Ivan Andrus about the same topic and as far as I know he
spent some time trying to debug in his simulator. He also mentioned that
since he's compiling with a newer XCode he's unsure if it's the same issue,

He recommended to switch to another Cell Server, actually the one you are
owner/maintainer, but as I told before it's still just printing/plotting
nothing at the output...

Regarding your inquiries, here are the details...

- *iOS version:* somehow old... 6.16.6 version (can't upgrade anymore in my
device), but according to the Sage Math App web page, this shouldn't be a
problem, since the latest version available provides the following details:
Updated: Jan 31, 2013

   - Version: 2.1.1
   - Size: 0.8 MB
   - Language: English
   - ...
   - Compatibility: Requires iOS 4.3 or later. Compatible with iPhone,
   iPad, and iPod touch.


*- device:* it's an iPod touch I think gen 4 or 5, I have the device since
2011...

-* web page/code trying to run:* any code from the Sage math web site, or
Matplotlib or Python code or the one already available by default in the
app, any of them used to work before,

I also noticed that if there's data or plots generated previously that
worked, they can stay in there, until I'm typing/modifying again... and
then by clicking the 'Evaluate' button... If I do that, that previous data
is no longer available but also nothing comes out on screen...

*- results: *
it just prints very quickly status messages like  *"Loading, please
wait..."* or "*There is no output yet" *it also prints at the bottom a
message like *Loading [ math Jax / extensions]... * or something like
that.. and after this... nothing comes out from the calculation or plot
statements,


I also noticed that if I try the sage Cell Server directly from the web
page through Safari and type anything and then click 'evaluate', after it
appears a disclaimer message, and clicking 'Ok' nothing else comes out from
the web page as well... so I'm wondering if this issue is just happening on
relatively old iOS based devices,


-- 
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: Coming SageMathCell upgrade - please test!

[Andrey Novoseltsev]

This is due to the bug fixed by http://trac.sagemath.org/ticket/20797

Also, I'll try to keep further discussion of SageMathCell at 
https://groups.google.com/forum/#!forum/sage-cell

On Monday, 6 June 2016 02:14:25 UTC-6, Peter Luschny wrote:
>
> ---
> ImportError   Traceback (most recent call last)
>  in ()
> > 1 load('http://')
>   2 load('http://')
>   3 
>
> /home/sc_serv/sage/src/sage/structure/sage_object.pyx in 
> sage.structure.sage_object.load 
> (/home/sc_serv/sage/src/build/cythonized/sage/structure/sage_object.c:11408)()
>1016 
>1017 if sage.repl.load.is_loadable_filename(filename):
> -> 1018 sage.repl.load.load(filename, globals())
>1019 return
>1020 
>
> /home/sc_serv/sage/local/lib/python2.7/site-packages/sage/repl/load.pyc in 
> load(filename, globals, attach)
> 252 raise NotImplementedError("you can't attach a URL")
> 253 from sage.misc.remote_file import get_remote_file
> --> 254 filename = get_remote_file(filename, verbose=False)
> 255 
> 256 from sage.repl.attach import load_attach_path
>
> /home/sc_serv/sage/local/lib/python2.7/site-packages/sage/misc/remote_file.pyc
>  
> in get_remote_file(filename, verbose)
>  34 
>  35 # import compatible with py2 and py3
> ---> 36 from six.moves.urllib.request import urlretrieve
>  37 
>  38 global cur
>
> ImportError: No module named moves.urllib.request
>
>

-- 
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: Weird ImportError with six

[Andrey Novoseltsev]

Thanks a lot for explanation, I was completely lost! The suggested fix is 
at http://trac.sagemath.org/ticket/20797

On Tuesday, 31 May 2016 01:26:35 UTC-6, Volker Braun wrote:
>
> great minds think alike ;-)
>
>
> On Tuesday, May 31, 2016 at 8:26:54 AM UTC+2, Frédéric Chapoton wrote:
>>
>> It's because these is a "six" module in sage/misc.
>>
>> I met this problem in #20694
>>
>> Solution: use from __future__ import absolute import
>>
>> Frederic
>>
>> Le mardi 31 mai 2016 07:55:14 UTC+2, Andrey Novoseltsev a écrit :
>>>
>>> Hello,
>>>
>>> Can anybody explain (even better - fix ;-)) what is going on with this 
>>> error with 7.2 and 7.3.beta0:
>>>
>>> sage: from sage.misc.remote_file import get_remote_file
>>> sage: get_remote_file('
>>> http://193.146.36.205:8080/GgbSageDirect/DiscoverGGB/DiscoverGGBAux/2015_11_24_fromGgbXMLString2Cons.sage
>>> ')
>>> Attempting to load remote file: 
>>> http://193.146.36.205:8080/GgbSageDirect/DiscoverGGB/DiscoverGGBAux/2015_11_24_fromGgbXMLString2Cons.sage
>>>
>>> ---
>>> ImportError   Traceback (most recent call 
>>> last)
>>>  in ()
>>> > 1 get_remote_file('
>>> http://193.146.36.205:8080/GgbSageDirect/DiscoverGGB/DiscoverGGBAux/2015_11_24_fromGgbXMLString2Cons.sage
>>> ')
>>>
>>> /home/novoselt/sage/local/lib/python2.7/site-packages/sage/misc/remote_file.pyc
>>>  
>>> in get_remote_file(filename, verbose)
>>>  34
>>>  35 # import compatible with py2 and py3
>>> ---> 36 from six.moves.urllib.request import urlretrieve
>>>  37
>>>  38 global cur
>>>
>>> ImportError: No module named moves.urllib.request
>>>
>>> Given that I can repeat this line directly without any issues:
>>>
>>> sage:  from six.moves.urllib.request import urlretrieve
>>> 
>>>
>>> Thank you!
>>> Andrey
>>>
>>>

-- 
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: [sage-notebook] Coming SageMathCell upgrade - please test!

[Andrey Novoseltsev]

On Thursday, 2 June 2016 16:39:53 UTC-6, paulmasson wrote:
>
> Andrey, any idea when my embedded interacts will start working again? 
> Thanks.
>

Should be working now, please test. There was a true check (undefined === 
undefined), I think I'd prefer AttributeError:
https://github.com/sagemath/sagecell/commit/94616895636418000c45478e080eb1f2ca7a50fc

-- 
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: [sage-notebook] Coming SageMathCell upgrade - please test!

[Andrey Novoseltsev]

On Thursday, 2 June 2016 16:39:53 UTC-6, paulmasson wrote:
>
> Andrey, any idea when my embedded interacts will start working again? 
> Thanks.
>

Hopefully by Monday. I still don't know what exactly is happening, but I 
didn't work much on it since last weekend was short on time in general and 
I was looking into CodeMirror problems. 

-- 
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] Weird ImportError with six

[Andrey Novoseltsev]

Hello,

Can anybody explain (even better - fix ;-)) what is going on with this 
error with 7.2 and 7.3.beta0:

sage: from sage.misc.remote_file import get_remote_file
sage: 
get_remote_file('http://193.146.36.205:8080/GgbSageDirect/DiscoverGGB/DiscoverGGBAux/2015_11_24_fromGgbXMLString2Cons.sage')
Attempting to load remote file: 
http://193.146.36.205:8080/GgbSageDirect/DiscoverGGB/DiscoverGGBAux/2015_11_24_fromGgbXMLString2Cons.sage
---
ImportError   Traceback (most recent call last)
 in ()
> 1 
get_remote_file('http://193.146.36.205:8080/GgbSageDirect/DiscoverGGB/DiscoverGGBAux/2015_11_24_fromGgbXMLString2Cons.sage')

/home/novoselt/sage/local/lib/python2.7/site-packages/sage/misc/remote_file.pyc 
in get_remote_file(filename, verbose)
 34
 35 # import compatible with py2 and py3
---> 36 from six.moves.urllib.request import urlretrieve
 37
 38 global cur

ImportError: No module named moves.urllib.request

Given that I can repeat this line directly without any issues:

sage:  from six.moves.urllib.request import urlretrieve


Thank you!
Andrey

-- 
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: [sage-notebook] Coming SageMathCell upgrade - please test!

[Andrey Novoseltsev]

On Thursday, 26 May 2016 10:34:10 UTC-6, kcrisman wrote:
>
>
> Please report any new (or old) errors that you notice - I will try to 
>> fix them tomorrow (Saturday) afternoon (MST) and on Sunday/Monday. If 
>> something is horribly wrong and I can't resolve it by Monday evening, 
>> I'll pull back to the old version. 
>>
>>
>
> http://ask.sagemath.org/question/33555/problem-with-encoding-german-characters/
>  
> could be related, or so the poster seems to imply - though probably not bad 
> enough to go back before 7.2. 
>

This should be fixed now - I had to encode user code when writing to a file 
for correct warning messages.

-- 
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: Sage apps for iOS and Android

[Andrey Novoseltsev]

On Friday, 27 May 2016 13:44:29 UTC-6, paulmasson wrote:
>
> If it helps to know, the Sage app is working fine for me on both my iPhone 
> and iPad. Some cells don't display nicely until the "Evaluate" button is 
> tapped, but otherwise everything appears to function.
>
> Perhaps the reported error had to do with changes being made last week on 
> the server. Not seeing any issues now.
>

Interesting, the only thing completely broken that got fixed last weekend 
was not displaying files when embedding script was included multiple 
times...

Related to "don't display nicely" - it seems to happen on desktops as well 
with custom styling, like here
http://users.humboldt.edu/flashman/MD/section-2.1LF.html subsection LF.COMP 
Example LF.COMP.1
if anyone has/gets any insight into it, please let me know. The only guess 
I have is that CodeMirror may be fully configured after some sizing already 
took place.
 

>
> On Thursday, May 26, 2016 at 8:46:15 AM UTC-7, Andrey Novoseltsev wrote:
>>
>> I got a report that iOS app using SageMathCell is broken and I imagine 
>> the same is true for Android. Can someone look into updating them? This 
>> commit may be helpful in figuring out what to do:
>>
>> https://github.com/sagemath/sagecell/commit/9878eadc0b6f9959388549e21d8f7945e94f00b2
>>
>> Thank you!
>> Andrey
>>
>

-- 
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.