[sage-devel] Re: Coming SageMathCell upgrade - please test!

2016-04-17 Thread mmarco 
Forgot to mention, I am using chromium in Linux.

El lunes, 18 de abril de 2016, 8:34:54 (UTC+2), mmarco escribió:
>
> Same Here.
>
> El lunes, 18 de abril de 2016, 4:49:50 (UTC+2), Kwankyu Lee escribió:
>>
>> The webpage from the Madrid server opens but contains no Sage cell. My 
>> web browser is Chrome on Mac. Accessing from Asia might be a cause.
>>
>

-- 
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!

2016-04-17 Thread mmarco 
Same Here.

El lunes, 18 de abril de 2016, 4:49:50 (UTC+2), Kwankyu Lee escribió:
>
> The webpage from the Madrid server opens but contains no Sage cell. My web 
> browser is Chrome on Mac. Accessing from Asia might be a cause.
>

-- 
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: Calculus bug in the latest 7.2.beta: sin(pi*x) returns 0 whenever x > 0

2016-04-17 Thread Ralf Stephan 
On Sunday, April 17, 2016 at 9:05:46 PM UTC+2, Volker Braun wrote:
>
> It seems that the assumption (wrongly) thinks that x is integer:
>

http://trac.sagemath.org/ticket/20456

Fixed in newest Pynac.

-- 
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!

2016-04-17 Thread Kwankyu Lee 
The webpage from the Madrid server opens but contains no Sage cell. My web 
browser is Chrome on Mac. Accessing from Asia might be a cause.

-- 
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] Coming SageMathCell upgrade - please test!

2016-04-17 Thread Andrey Novoseltsev 
Hello and sorry for cross-posting, I'm trying to maximize visibility.

After quite a few changes I have a version of SageMathCell running Sage-7.1 
and using Jupyter etc. shipped with Sage, rather than a custom old branch 
of IPython.

User-visible changes should be minimal, but there are some for script 
interaction that will probably require changes to Android app, WebWorks, 
and what else.

There may be also new bugs introduced in the process. I hope that most of 
the stuff is done correctly and I will do some more 
testing/fixing/documenting over the next month or so, but it would be nice 
if other people tried it with their favourite code, interacts, web-pages, 
etc. To help with this I've pushed all changed to master, so those who have 
their own installations should be using the new version, and I have also 
set up a test instance on our Madrid server:
http://cosmos.mat.uam.es:/
for comparison you can also use old version at
http://cosmos.mat.uam.es/

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: Installing PyX breaks pip

2016-04-17 Thread Dima Pasechnik 


On Sunday, April 17, 2016 at 11:50:37 PM UTC+1, Andrey Novoseltsev wrote:
>
> We have an optional package for PyX-0.10 (recent versions are 0.14). After 
> installing it pip stops working with the message like
>
> Traceback (most recent call last):
>   File "/home/sc_serv/sage/local/bin/pip", line 5, in 
> from pkg_resources import load_entry_point
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 2912, in 
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 2898, in _call_aside
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 2925, in _initialize_master_working_set
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 633, 
> in _build_master
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 626, 
> in __init__
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 682, 
> in add_entry
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 1956, in find_on_path
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 2343, in from_location
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 2660, in _reload_version
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 2308, in _version_from_file
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 2476, in _get_metadata
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 1847, in get_metadata_lines
>   File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 
> 1842, in get_metadata
>   File "/home/sc_serv/sage/local/lib/python/codecs.py", line 302, in decode
> (result, consumed) = self._buffer_decode(data, self.errors, final)
> UnicodeDecodeError: 'utf8' codec can't decode byte 0xf6 in position 165: 
> invalid start byte
>
> which seems to be due to author's name with non-Latin characters.
>
> Installing PyX through pip does not work for some reason, 
>

it's because 0.14 is python3 only. The following works for me:

$ sage --pip install  --allow-external pyx pyx==0.12.1


 

> but I think it should be removed from optional packages as it breaks pip 
> which is way more important.
>

as you can do the above, there is no point in keeping 0.10

 

-- 
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] Installing PyX breaks pip

2016-04-17 Thread Andrey Novoseltsev 
We have an optional package for PyX-0.10 (recent versions are 0.14). After 
installing it pip stops working with the message like

Traceback (most recent call last):
  File "/home/sc_serv/sage/local/bin/pip", line 5, in 
from pkg_resources import load_entry_point
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 2912, 
in 
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 2898, 
in _call_aside
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 2925, 
in _initialize_master_working_set
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 633, 
in _build_master
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 626, 
in __init__
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 682, 
in add_entry
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 1956, 
in find_on_path
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 2343, 
in from_location
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 2660, 
in _reload_version
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 2308, 
in _version_from_file
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 2476, 
in _get_metadata
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 1847, 
in get_metadata_lines
  File "build/bdist.linux-x86_64/egg/pkg_resources/__init__.py", line 1842, 
in get_metadata
  File "/home/sc_serv/sage/local/lib/python/codecs.py", line 302, in decode
(result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf8' codec can't decode byte 0xf6 in position 165: 
invalid start byte

which seems to be due to author's name with non-Latin characters.

Installing PyX through pip does not work for some reason, but I think it 
should be removed from optional packages as it breaks pip which is way more 
important.

-- 
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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread Dima Pasechnik 


On Sunday, April 17, 2016 at 9:51:03 PM UTC+1, mmarco wrote:
>
> Would it be possible to ship such a pip repo (with only the packages 
> needed by sage) in the sage source tarball? Or maybe have two such 
> tarballs, one with it (so it would allow to compile Sage offline, and 
> source review before compiling), and another that would just rely on 
> downloading packages from pypi?
>

It looks like it would make more sense to provide a cache of pypi packages 
needed for Sage building.
It might either be a tarball, or just a list of files to download from a 
pypi mirror.
The less duplication, the better...


>
> El domingo, 17 de abril de 2016, 17:38:23 (UTC+2), William escribió:
>>
>> On Sun, Apr 17, 2016 at 8:27 AM, Dima Pasechnik  
>> wrote: 
>> > 
>> > 
>> > On Sunday, April 17, 2016 at 4:20:42 PM UTC+1, Dima Pasechnik wrote: 
>> >> 
>> >> 
>> >> 
>> >> On Sunday, April 17, 2016 at 4:15:12 PM UTC+1, William wrote: 
>> >>> 
>> >>> Dima: 
>> >>> > sorry, William, I don't see what you mean. 
>> >>> [...] 
>> >>> > 2) Putting data on your computer does not equal to installing crud 
>> on 
>> >>> > your 
>> >>> > computer, although people wearing tinfoil hats 
>> >>> > might disagree. 
>> >>> 
>> >>> They do disagree.  And they are right. 
>> >> 
>> >> 
>> >> Why would they trust somebody putting hundreds of megabytes of 3rd 
>> party 
>> >> sources into Sage? Is it because we mislead them into thinking that we 
>> did 
>> >> some kind of source review? 
>>
>> You are right.However, such a tinfoil hat person would rather 
>> (themselves) inspect the packages that are just needed for Sage, 
>> rather than all of Pypi. 
>>
>> > they actually would be less paranoid about having the 3rd party from 
>> its 
>> > original source rather than re-packaged, and this is precisely a part 
>> of my 
>> > proposal. And, by the way, 
>> > http://doc.devpi.net/latest/quickstart-pypimirror.html 
>> > is precisely a way to create a mini-repo containing only things needed 
>> by 
>> > Sage. 
>>
>> Perfect -- I agree that is the right approach. 
>>
>>
>> -- 
>> 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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread mmarco 
Would it be possible to ship such a pip repo (with only the packages needed 
by sage) in the sage source tarball? Or maybe have two such tarballs, one 
with it (so it would allow to compile Sage offline, and source review 
before compiling), and another that would just rely on downloading packages 
from pypi?


El domingo, 17 de abril de 2016, 17:38:23 (UTC+2), William escribió:
>
> On Sun, Apr 17, 2016 at 8:27 AM, Dima Pasechnik  > wrote: 
> > 
> > 
> > On Sunday, April 17, 2016 at 4:20:42 PM UTC+1, Dima Pasechnik wrote: 
> >> 
> >> 
> >> 
> >> On Sunday, April 17, 2016 at 4:15:12 PM UTC+1, William wrote: 
> >>> 
> >>> Dima: 
> >>> > sorry, William, I don't see what you mean. 
> >>> [...] 
> >>> > 2) Putting data on your computer does not equal to installing crud 
> on 
> >>> > your 
> >>> > computer, although people wearing tinfoil hats 
> >>> > might disagree. 
> >>> 
> >>> They do disagree.  And they are right. 
> >> 
> >> 
> >> Why would they trust somebody putting hundreds of megabytes of 3rd 
> party 
> >> sources into Sage? Is it because we mislead them into thinking that we 
> did 
> >> some kind of source review? 
>
> You are right.However, such a tinfoil hat person would rather 
> (themselves) inspect the packages that are just needed for Sage, 
> rather than all of Pypi. 
>
> > they actually would be less paranoid about having the 3rd party from its 
> > original source rather than re-packaged, and this is precisely a part of 
> my 
> > proposal. And, by the way, 
> > http://doc.devpi.net/latest/quickstart-pypimirror.html 
> > is precisely a way to create a mini-repo containing only things needed 
> by 
> > Sage. 
>
> Perfect -- I agree that is the right approach. 
>
>
> -- 
> 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.

[sage-devel] Re: Calculus bug in the latest 7.2.beta: sin(pi*x) returns 0 whenever x > 0

2016-04-17 Thread Volker Braun 
It seems that the assumption (wrongly) thinks that x is integer:

sage: assume(x>0)
sage: sin(pi*x)
0
sage: cos(pi*x)
(-1)^x



On Sunday, April 17, 2016 at 5:58:07 PM UTC+2, Eric Gourgoulhon wrote:
>
> Hi,
>
> In Sage 7.2.beta4:
>
> sage: assume(x>0)
> sage: sin(pi*x)
> 0
>
> As far as I can tell, this bug has been introduced in Sage 7.2.beta0. 
> Maybe by ticket #20134  (upgrade 
> to pynac-0.6.4) ?
>
> Best wishes,
>
> Eric.
>
>

-- 
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] Calculus bug in the latest 7.2.beta: sin(pi*x) returns 0 whenever x > 0

2016-04-17 Thread Ralf Stephan 
Many thanks for reporting but who will open the ticket?

-- 
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] Calculus bug in the latest 7.2.beta: sin(pi*x) returns 0 whenever x > 0

2016-04-17 Thread Eric Gourgoulhon 
Hi,

In Sage 7.2.beta4:

sage: assume(x>0)
sage: sin(pi*x)
0

As far as I can tell, this bug has been introduced in Sage 7.2.beta0. Maybe 
by ticket #20134  (upgrade to 
pynac-0.6.4) ?

Best wishes,

Eric.

-- 
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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread William Stein 
On Sun, Apr 17, 2016 at 8:27 AM, Dima Pasechnik  wrote:
>
>
> On Sunday, April 17, 2016 at 4:20:42 PM UTC+1, Dima Pasechnik wrote:
>>
>>
>>
>> On Sunday, April 17, 2016 at 4:15:12 PM UTC+1, William wrote:
>>>
>>> Dima:
>>> > sorry, William, I don't see what you mean.
>>> [...]
>>> > 2) Putting data on your computer does not equal to installing crud on
>>> > your
>>> > computer, although people wearing tinfoil hats
>>> > might disagree.
>>>
>>> They do disagree.  And they are right.
>>
>>
>> Why would they trust somebody putting hundreds of megabytes of 3rd party
>> sources into Sage? Is it because we mislead them into thinking that we did
>> some kind of source review?

You are right.However, such a tinfoil hat person would rather
(themselves) inspect the packages that are just needed for Sage,
rather than all of Pypi.

> they actually would be less paranoid about having the 3rd party from its
> original source rather than re-packaged, and this is precisely a part of my
> proposal. And, by the way,
> http://doc.devpi.net/latest/quickstart-pypimirror.html
> is precisely a way to create a mini-repo containing only things needed by
> Sage.

Perfect -- I agree that is the right approach.


-- 
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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread Dima Pasechnik 


On Sunday, April 17, 2016 at 4:20:42 PM UTC+1, Dima Pasechnik wrote:
>
>
>
> On Sunday, April 17, 2016 at 4:15:12 PM UTC+1, William wrote:
>>
>> Dima: 
>> > sorry, William, I don't see what you mean. 
>> [...] 
>> > 2) Putting data on your computer does not equal to installing crud on 
>> your 
>> > computer, although people wearing tinfoil hats 
>> > might disagree. 
>>
>> They do disagree.  And they are right.  
>
>
> Why would they trust somebody putting hundreds of megabytes of 3rd party 
> sources into Sage? Is it because we mislead them into thinking that we did 
> some kind of source review?
>
 
they actually would be less paranoid about having the 3rd party from its 
original source rather than re-packaged, and this is precisely a part of my 
proposal. And, by the 
way, http://doc.devpi.net/latest/quickstart-pypimirror.html
is precisely a way to create a mini-repo containing only things needed by 
Sage.

Anyhow, it seems that what I propose is not evolutional enough to merit 
a normal discussion. :-(


>
>
>  
>
>>
>> -- 
>> 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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread Dima Pasechnik 


On Sunday, April 17, 2016 at 4:15:12 PM UTC+1, William wrote:
>
> Dima: 
> > sorry, William, I don't see what you mean. 
> [...] 
> > 2) Putting data on your computer does not equal to installing crud on 
> your 
> > computer, although people wearing tinfoil hats 
> > might disagree. 
>
> They do disagree.  And they are right.  


Why would they trust somebody putting hundreds of megabytes of 3rd party 
sources into Sage? Is it because we mislead them into thinking that we did 
some kind of source review?



 

>
> -- 
> 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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread William Stein 
Dima:
> sorry, William, I don't see what you mean.
[...]
> 2) Putting data on your computer does not equal to installing crud on your
> computer, although people wearing tinfoil hats
> might disagree.

They do disagree.  And they are right.

-- 
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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread Dima Pasechnik 


On Sunday, April 17, 2016 at 3:43:15 PM UTC+1, William wrote:
>
>
>
> On Sunday, April 17, 2016, Dima Pasechnik > 
> wrote:
>
>>
>>
>> On Sunday, April 17, 2016 at 2:14:09 AM UTC+1, kcrisman wrote:
>>>
>>>  
>>>

> (1) Providing a source tarball from which one can build Sage without 
> internet access is a basic requirement from a company that provides 
> some funding for Sage development over the years.   We don't 
> contractually have to do it, but I think it is a good to be able to do 
> so. (2) It is also helpful when we have Sage days workshops at places 
> with crappy internet (frequent), since at least everybody can build 
> Sage locally from a single tarball. 
>

 PS. in my eyes, my proposal would bring Sage closer to "normal 
 open-source development
 practices" that you were arguing for recently. No normal project I know 
 creates
 megamegatarball of everything; reducing these would be a step in the 
 right direction.

>
>
>>> Remind me again how this helps with (2)?  But more seriously, is there a 
>>> way to have the "normal" distribution be pip or whatever but have a 
>>> scriptable way to create a tarball for everything that is just not the 
>>> default one people are pointed to?  I assume yes. 
>>>
>> here is one I found
>> http://doc.devpi.net/latest/quickstart-pypimirror.html 
>>
>> it will create a pypi cache that can be served locally.
>>
>> Then, I saw a link saying that a complete pypi mirror is "just" 120GB.
>> Even if this is 5 times too optimistic, it would still fit on a portable 
>> 1TB or so disk you can take with you.
>>
>
> This fails to solve the problem of a security-paranoid person with an air 
> gapped computer, since such a person would be careful about what data they 
> put on their computer.  They would prefer to just put the 
> sage-related packages, not all pypi.
>
 
sorry, William, I don't see what you mean.

1) We talk about a workshop at a remote location situation, bad internet 
access etc.

2) Putting data on your computer does not equal to installing crud on your 
computer, although people wearing tinfoil hats
might disagree. And I never said: "copy the whole of pypi to evreyone's 
machine".
I said, basically, have a server on a local net with a mirror of pypi (or 
its part). 

3) if Sage must be approved for use in psychiatric wards, it's a different 
story all together. :-)


 
>
>>
>> Perhaps a single tarball was important to have 10 years ago, nowadays 
>> it's really not something so important...
>>
>>
>> -- 
>> 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.
>>
>
>
> -- 
> Sent from my massive iPhone 6 plus.
>

-- 
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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread William Stein 
On Sunday, April 17, 2016, Dima Pasechnik  wrote:

>
>
> On Sunday, April 17, 2016 at 2:14:09 AM UTC+1, kcrisman wrote:
>>
>>
>>
>>>
 (1) Providing a source tarball from which one can build Sage without
 internet access is a basic requirement from a company that provides
 some funding for Sage development over the years.   We don't
 contractually have to do it, but I think it is a good to be able to do
 so. (2) It is also helpful when we have Sage days workshops at places
 with crappy internet (frequent), since at least everybody can build
 Sage locally from a single tarball.

>>>
>>> PS. in my eyes, my proposal would bring Sage closer to "normal
>>> open-source development
>>> practices" that you were arguing for recently. No normal project I know
>>> creates
>>> megamegatarball of everything; reducing these would be a step in the
>>> right direction.
>>>


>> Remind me again how this helps with (2)?  But more seriously, is there a
>> way to have the "normal" distribution be pip or whatever but have a
>> scriptable way to create a tarball for everything that is just not the
>> default one people are pointed to?  I assume yes.
>>
> here is one I found
> http://doc.devpi.net/latest/quickstart-pypimirror.html
>
> it will create a pypi cache that can be served locally.
>
> Then, I saw a link saying that a complete pypi mirror is "just" 120GB.
> Even if this is 5 times too optimistic, it would still fit on a portable
> 1TB or so disk you can take with you.
>

This fails to solve the problem of a security-paranoid person with an air
gapped computer, since such a person would be careful about what data they
put on their computer.  They would prefer to just put the
sage-related packages, not all pypi.


>
> Perhaps a single tarball was important to have 10 years ago, nowadays it's
> really not something so important...
>
>
> --
> 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.
>


-- 
Sent from my massive iPhone 6 plus.

-- 
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] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread Isuru Fernando 
On Sun, Apr 17, 2016 at 7:15 PM, Dima Pasechnik  wrote:

>
>
> On Sunday, April 17, 2016 at 2:14:09 AM UTC+1, kcrisman wrote:
>>
>>
>>
>>>
 (1) Providing a source tarball from which one can build Sage without
 internet access is a basic requirement from a company that provides
 some funding for Sage development over the years.   We don't
 contractually have to do it, but I think it is a good to be able to do
 so. (2) It is also helpful when we have Sage days workshops at places
 with crappy internet (frequent), since at least everybody can build
 Sage locally from a single tarball.

>>>
>>> PS. in my eyes, my proposal would bring Sage closer to "normal
>>> open-source development
>>> practices" that you were arguing for recently. No normal project I know
>>> creates
>>> megamegatarball of everything; reducing these would be a step in the
>>> right direction.
>>>


>> Remind me again how this helps with (2)?  But more seriously, is there a
>> way to have the "normal" distribution be pip or whatever but have a
>> scriptable way to create a tarball for everything that is just not the
>> default one people are pointed to?  I assume yes.
>>
> here is one I found
> http://doc.devpi.net/latest/quickstart-pypimirror.html
>
> it will create a pypi cache that can be served locally.
>
> Then, I saw a link saying that a complete pypi mirror is "just" 120GB.
> Even if this is 5 times too optimistic, it would still fit on a portable
> 1TB or so disk you can take with you.
>

Or you can just download the packages to SAGE_ROOT/upstream from pip and
create a tarball for everything and use the downloaded packages to install
using pip without an internet connection

./sage -pip install --download $SAGE_ROOT/upstream sympy  # Download sympy
and dependencies to $SAGE_ROOT/upstream
./sage -pip install --no-index --find-links=$SAGE_ROOT/upstream --upgrade
sympy  # Use the downloaded packages


>
> Perhaps a single tarball was important to have 10 years ago, nowadays it's
> really not something so important...
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Proposal: install all pip-installable components of Sage via pip, directly

2016-04-17 Thread Dima Pasechnik 


On Sunday, April 17, 2016 at 2:14:09 AM UTC+1, kcrisman wrote:
>
>  
>
>>
>>> (1) Providing a source tarball from which one can build Sage without 
>>> internet access is a basic requirement from a company that provides 
>>> some funding for Sage development over the years.   We don't 
>>> contractually have to do it, but I think it is a good to be able to do 
>>> so. (2) It is also helpful when we have Sage days workshops at places 
>>> with crappy internet (frequent), since at least everybody can build 
>>> Sage locally from a single tarball. 
>>>
>>
>> PS. in my eyes, my proposal would bring Sage closer to "normal 
>> open-source development
>> practices" that you were arguing for recently. No normal project I know 
>> creates
>> megamegatarball of everything; reducing these would be a step in the 
>> right direction.
>>
>>>
>>>
> Remind me again how this helps with (2)?  But more seriously, is there a 
> way to have the "normal" distribution be pip or whatever but have a 
> scriptable way to create a tarball for everything that is just not the 
> default one people are pointed to?  I assume yes. 
>
here is one I found
http://doc.devpi.net/latest/quickstart-pypimirror.html 

it will create a pypi cache that can be served locally.

Then, I saw a link saying that a complete pypi mirror is "just" 120GB.
Even if this is 5 times too optimistic, it would still fit on a portable 
1TB or so disk you can take with you.

Perhaps a single tarball was important to have 10 years ago, nowadays it's 
really not something so important...


-- 
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: Does any repository at github.com/sagemath accept Github pull requests?

2016-04-17 Thread Harald Schilly 


On Saturday, April 16, 2016 at 11:54:11 PM UTC+2, paulmasson wrote:
>
> Are all of these repositories used merely for hosting files?
>
> Does any active development take place purely on Github, i.e., outside of 
> the Trac system?
>
>
Yes, website and publications. 

-- h
 

-- 
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: Proposal: move SageNB back to Sage

2016-04-17 Thread Jeroen Demeyer 

On 2016-04-17 03:27, kcrisman wrote:

Last night I was thinking about another issue - the eventual proposed
complete removal of sagenb from Sage


Just a small addition: even if we switch to Jupyter by default, that 
doesn't mean that we will drop SageNB immediately. We should still ship 
it as non-default notebook for a few years.


--
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.