from:"Sergey Kirpichev"

[sympy] Re: Code of Conduct

2016-02-02 Thread Sergey Kirpichev


On Monday, February 1, 2016 at 12:10:57 AM UTC+3, Aaron Meurer wrote:
>
> We have been discussing for a while about adding a code of conduct to 
> SymPy, so that it is very clear to everyone what kind of conduct we expect 
> from those who participate in the community.
>

Great news!

I'm just curious, who from developers of Sympy are ready to "enforce the 
Code of Conduct" or fill "complains to be reviewed".  Chris Smith, perhaps, 
or Matthew Rocklin?

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/09c077bd-0da4-41b1-80bc-b9c9490ff92c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: hyperexpand returns incorrect result for negative coefficients

2016-01-31 Thread Sergey Kirpichev

On Sun, Jan 31, 2016 at 8:06 PM, Ondřej Čertík  wrote:
> Regarding the legal question that arose in this thread, I have asked here:
> http://programmers.stackexchange.com/questions/308909/how-to-manage-copyright-notices-from-contributors-to-a-bsd-licensed-project

Ondrej, I'm sorry, but it's a wrong question, do you understand?

You can consider me as a contributor - in case, like with _submitted_
patch above.  And here you have a clear permission in the very beginning
of this conversation: please do, whatever you want.  Is this still not clear?!

But in general, my repo - is not a reservoir of public domain code for
inclusion eventually in sympy, that can be just cherry-pick'ed or
copy-pasted.  Unless all this wasn't submitted to you by author - I'm not
your contributor and you have simple BSD-terms, if you really want to
use that, it's all.

PS:
> I have asked here

Please respect intelligence of you readers and don't misinform people with
ridiculous claims that I require from you changing Copyright notice just
for a few-line patch.

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20160131182619.GA5199%40darkstar.order.hcn-strela.ru.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: hyperexpand returns incorrect result for negative coefficients

2016-01-29 Thread Sergey Kirpichev

FYI https://github.com/skirpichev/omg/pull/204

On Thursday, January 28, 2016 at 4:12:18 AM UTC+3, Ondřej Čertík wrote:
>
> GitHub is down, so I am reporting here for now: 
>
>
> It returns 0 for the following function, even though we can verify the 
> (correct) answer using .n(), that it is nonzero: 
>
> In [123]: hyper((-6, -7, -5), (-6, -6), 1) 
> Out[123]: 
>  ┌─  ⎛-6, -7, -5 │  ⎞ 
>  ├─  ⎜   │ 1⎟ 
> 3╵ 2 ⎝  -6, -6   │  ⎠ 
>
> In [124]: hyperexpand(hyper((-6, -7, -5), (-6, -6), 1)) 
> Out[124]: 0 
>
> In [125]: hyper((-6, -7, -5), (-6, -6), 1).n() 
> Out[125]: -0.167 
>
>
> In some other cases it works: 
>
> In [128]: hyperexpand(hyper((-5, -3, -4), (-6, -6), 1)) 
> Out[128]: 1/30 
>
> In [129]: hyper((-5, -3, -4), (-6, -6), 1).n() 
> Out[129]: 0.0333 
>
>
> Ondrej 
>

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/1f0b5434-c5a6-458c-a906-41a8a5676541%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Wrong results for limits with sign() function

2015-12-14 Thread Sergey Kirpichev

On Sunday, February 15, 2015 at 3:55:28 PM UTC+3, Arnaud Usciati wrote:
>
> If x = symbols('x', real=True).
>
> I found wrong results for limits with sign() function.
> limit(sign(x), x, 0, '+') = 1, and limit(sign(x), x, 0, '-') = -1 ---> OK
>
> But :
> - limit(sign(ln(x)), x, 1, '+') = 1 (OK) but limit(sign(ln(x)), x, 1, '-') 
> = 0 instead of -1 ;
> - limit(sign(sin(x)), x, 0, '+') = 0 and limit(sign(sin(x)), x, 0, '-') = 
> 0, instead of 1 and -1 ;
> - limit(sign(tan(x)), x, 0, '+') = 0 and limit(sign(tan(x)), x, 0, '-') = 
> 0, instead of 1 and -1 ;
> - limit(sign(cos(x)), x, pi/2, '+') = 0 and limit(sign(cos(x)), x, pi/2, 
> '-') = 0, instead of 1 and -1 ;
> Etc..
>

Patch: https://github.com/skirpichev/omg/pull/154
(Should work for sympy with trivial replacement .is_extended_real -> 
.is_real)

Please note that
> limit(sign(cos(x)), x, pi/2, '+') = 0 and limit(sign(cos(x)), x, pi/2, 
'-') = 0, instead of 1 and -1 ;

is incorrect.  Should be -1 and 1 instead, c.f. Mathematica output:

In[3]:= Limit[Sign[Cos[x]], x->Pi/2, Direction->-1] (* default, equivalent 
to dir="+" *)
Out[3]= 
-1
In[4]:= Limit[Sign[Cos[x]], x->Pi/2, Direction->+1]
Out[4]= 
1

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/7be3c90d-fd7d-46f3-abed-1e8c40a25aa7%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Issue : limit(abs(log(x)),x,0,'+')

2015-12-14 Thread Sergey Kirpichev

On Thursday, February 12, 2015 at 10:26:32 PM UTC+3, Arnaud Usciati wrote:
>
> Maybe abs() function have to be reconsidered, because I found other issues 
> with abs...
>

All examples above are solved with patch from
 https://github.com/skirpichev/omg/pull/82
(on top of the https://github.com/skirpichev/omg/pull/154)
That should work for sympy with trivial replacements (is_extended_real -> 
is_real).

limit(abs(ln(x))-ln(x), x, 2) or limit(abs(ln(x))-ln(x), x, 2, '+') return 
> 0 (OK), but limit(abs(ln(x))-ln(x), x, 2, '-') runs endless and I need to 
> interrupt it to stop calculating.. Same way with functions : 
> abs(cos(x))-cos(x) or abs(sin(x))-sin(x) or abs(sqrt(x))-sqrt(x) for x = 
> symbols('x', real=True)
>

The problem is that nseries constantly emits an Order term and lseries 
can't yield anything at all.  But for real,
positive _and finite_ x < 1: abs(ln(2-x))-ln(2-x) == 0.

But if I try : limit(abs(ln(x))-ln(x), x, a, '-') with a = symbols('a', 
> real=True), it returns (sign(log(a)) - 1)*log(a). Idem for the direction 
> '+' and without direction.
>

This is an example of lucky answer, due to a non-strict approach in the 
calculate_series routine to
the zero-decision problem.  I.e. if sympy can't simplify term to zero, it 
thinks term is nonzero.

Wrong idea in general, but in this particular case it happens to be right.

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at https://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/394800bf-d5b2-42d1-826a-aac73f9f3409%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: a bug in definite integration in sympy-0.7.6.1

2015-11-30 Thread Sergey Kirpichev

This bug is in polylog eval method.  It returns wrong type (builtin "int", 
not sympy's Integer).

Fix is very simple:
https://github.com/skirpichev/omg/pull/137

On Monday, November 30, 2015 at 10:26:24 PM UTC+3, Andrey Grozin wrote:
>
> grozin@elrond ~ $ python
> Python 3.4.3 (default, Nov  8 2015, 11:52:08) 
> [GCC 4.9.3] on linux
> Type "help", "copyright", "credits" or "license" for more information.
> >>> from sympy import *
> >>> x=Symbol('x')
> >>> integrate(log(x)/(1-x),(x,0,1))
> Traceback (most recent call last):
>   File "", line 1, in 
>   File "/usr/lib/python3.4/site-packages/sympy/utilities/decorator.py", 
> line 35, in threaded_func
> return func(expr, *args, **kwargs)
>   File "/usr/lib/python3.4/site-packages/sympy/integrals/integrals.py", 
> line 1232, in integrate
> risch=risch, manual=manual)
>   File "/usr/lib/python3.4/site-packages/sympy/integrals/integrals.py", 
> line 532, in doit
> function = antideriv._eval_interval(x, a, b)
>   File "/usr/lib/python3.4/site-packages/sympy/core/expr.py", line 745, in 
> _eval_interval
> if B.has(S.NaN) or B.has(S.Infinity) or B.has(S.NegativeInfinity):
> AttributeError: 'int' object has no attribute 'has'
> >>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/acf7b217-5493-466c-a650-3355ef72ac98%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Quitting

2015-02-02 Thread Sergey Kirpichev

On Sunday, February 1, 2015 at 1:19:35 AM UTC+3, Joachim Durchholz wrote:

 Am 30.01.2015 um 17:35 schrieb Ondřej Čertík: 
  The door is always open if you want to come back. 

 I'll bring a revolving door :-D 

 Some aspects I was made aware of in private mails have made that 
 quitting decision far less certain than I thought just yesterday.


Oh, SymPy is safe now! :)

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/3c786230-6c24-4821-a008-a0fb4c2a3fc3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Quitting

2015-01-30 Thread Sergey Kirpichev



On Friday, January 30, 2015 at 6:37:11 AM UTC+3, Joachim Durchholz wrote:

 B) Argue to the arbiters into sanctioning Sergey more so that he stops 
 his passive aggression. 


They can't do more, I'm not part of this project.  But Aaron, maybe you can
remove me from whitelist for this maillist too?  (I don't think this will 
add too much
work for moderators...)
 

 C) Put up the him or me choice again.


No such choice at all is possible, I'm not involved into sympy.  The reason
I responded here - I was subscribed on this conversation before. 

Stop this insanity, please.  You are no fun anymore.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/d17babae-6672-40eb-95c8-9e0164c5c415%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] debian package (to be orphaned)

2015-01-29 Thread Sergey Kirpichev

On Thu, Jan 29, 2015 at 11:33 AM, Joachim Durchholz j...@durchholz.org wrote:
 Please clarify where I went wrong.

I did.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20150129112650.GA28421%40darkstar.order.hcn-strela.ru.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] debian package (to be orphaned)

2015-01-28 Thread Sergey Kirpichev

On Thursday, January 29, 2015 at 12:34:08 AM UTC+3, Joachim Durchholz wrote:

 Policy is that if it's a function that users would expect, it should be 
 depends


No, policy is not about this.  Don't misguide people, please.  Read the 
documentation.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/1fd56239-51aa-4064-85ef-115f189700c7%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] debian package (to be orphaned)

2015-01-28 Thread Sergey Kirpichev

On Wednesday, January 28, 2015 at 10:48:25 PM UTC+3, Aaron Meurer wrote:

 What is the Debian policy?


https://www.debian.org/doc/debian-policy/ch-relationships.html#s-binarydeps
 

 If a single function uses something is it a recommended dependency?


It depends on how important this function is for your users. 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/02248ac2-14d9-458d-812f-ab6cdfe87abb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] debian package (to be orphaned)

2015-01-20 Thread Sergey Kirpichev

Hello,

I'll stop maintain the debian sympy package too. Is there anyone interested in 
supporting this?

Package now is in the debian-science repo:
http://anonscm.debian.org/cgit/debian-science/packages/sympy.git/
In the NEW queue there is an upload, that fixes all reported debian bugs:
https://ftp-master.debian.org/new/sympy_0.7.5-4.html
So, I think, it's in a good shape.  And it should be trivial to update this to 
a new upstream version.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20150120122148.GB22117%40darkstar.order.hcn-strela.ru.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Wow. (was: Add pep8 config, use pep8 in travis (#8538))

2015-01-20 Thread Sergey Kirpichev

Hello,

I feel it would be a bad idea to quit silently. Probably, I should
explain this action for others (without fair warings and so on). This
thread - probably the best place, due to context.

I got several private messages about the tone in pull requests and so on.
Actually, besides the current example, also these:
https://github.com/sympy/sympy/issues/7971
https://github.com/sympy/sympy/pull/7356
(Perhaps, there are more, but I'm not informed of.)

While I'm ok when someone point on my flaws, errors and so on (sometimes
people
just have bad days, it happens) and I'll try to correct my behaviour
accordingly,
I think this should be done with several principles:
1) assume good faith
2) be fair, i.e. analyze the whole conversation, not
part of it or just single commentary (all above cases)
3) be helpful. For example, if you want to improve
something - just do it. You see bad answer - post good one; you
see unconstructive conversation - help people to reach a consensus.
4) please remember that all you assertions about
the tone, kindness and so on - are more or less subjective.

I feel that these principles are violated here too often. Ondrej also said
me, that
we are loosing people due to the harsh atmosphere I created. If it's a
real issue (i.e.
there are such people), I would like to apologize to them personally and
I'll happy
to announce that I'm not a treat for the sympy community anymore.
Perhaps, it's my bad and my communication skills are too low for the sympy.

On another hand, if this is a potential loss - please remember, you are
loosing
people for a number of different reasons. For example, if the review is
very
polite, but will end with committing of the wrong and non-professional code
-
this is also bad. People will be disappointed. Also, should we welcome
to comment on any issue or email thread without adding some requirements
(see below what I assume from reviewers, for example)? Longer
conversation - harder for other reviewers. Being friendly, welcoming
and polite - not a real requirement here (in first - it's too subjective),
rather a sane suggestion. But if you prefer such requirements instead...
Well, perhaps, you guys should wipe out quote talk is cheap, show me the
code
from the instructions for newcomers. From the history of the thread in
subject and persons involved, it seems now clear that the sympy community
prefer talks. I don't
think this is a good direction for the project as a whole. For example, I
would like to assume, that every PR reviewer will at least read the workflow
instruction, PR content, relevant documentation and will actually run the
code.
It seems, this is not accepted as a sane assumption by the sympy community.
Then, I think that my involvement in the project was waste of my time and I
should stop one.

Just being welcoming and polite actually looks as a very secondary
goal. Primary one should be - create and save an alive, professional and
productive
community around the project. But for you it's not, for example you lost a
real and very
productive contributor (well, at least one I'm aware off) just due to the
license
change (only potential gain; I'm not sure you got any profit to the present
time) - is this ok? (I don't think so and authors_skip variable in the
bin/mailmap_update.py - seems for me as a shame for the project.)

Sorry for long message, but I think there is something to
think about. Esp. that most members of the sympy group are
inactive now (I hope, this not my bad only) and that the sympy is
just as close to enter serious CAS market as it was
in 2008 (when you lost Kirill).

On Friday, January 2, 2015 at 10:37:40 AM UTC+3, Joachim Durchholz wrote:
Fair warning: If this kind of behaviour is considered acceptable by the
project, I'm out of here. Everybody's chance to rid the SymPy project of
that stubborn, nagging Toolforger me ;-)

Thanks to Joachim, who showed me the door. Really thanks, no
bad feelings from my side at all.

Thanks to all, thanks to all who helped me with patches, esp. to Chris and
Aaron.

--
You received this message because you are subscribed to the Google Groups
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/42947884-38ea-4012-aa75-dce287207ed9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] debian package (to be orphaned)

2015-01-20 Thread Sergey Kirpichev

On Tuesday, January 20, 2015 at 6:51:48 PM UTC+3, Ondřej Čertík wrote:

 Thanks Sergey for maintaining it. I created the initial Debian 
 package, and now I am a Debian Developer, so I can upload it myself. 


Good news.

I think, you are member of the debian science team, so there no problems to 
access the repo.
But it's also possible now to use single repo (i.e. github's) to keep both 
master branch
and stuff for debian's work (debian, upstream, pristine-tar branches):
https://wiki.debian.org/PackagingWithGit?highlight=%28\bCategoryGit\b%29#Using_the_upstream_repo

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/2ee94840-574b-4e06-ab8a-58ec75683d93%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: summation of function

2015-01-14 Thread Sergey Kirpichev



On Wednesday, January 14, 2015 at 12:56:40 PM UTC+3, Gaurav Dhingra wrote:

 Hi
 Is this a bug ?


Sure, it's.
https://github.com/sympy/sympy/issues/8822

Sympy's bugtracker is here:
https://github.com/sympy/sympy/issues/

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/2ddbb956-3c46-4b4c-95e9-325d89310749%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Unable to solve 2xcos(2x) + sin(2x)

2015-01-13 Thread Sergey Kirpichev

On Wednesday, January 14, 2015 at 1:14:01 AM UTC+3, Amit Saha wrote:

 Thanks for any insights into this and whether there is a workaround. 


What do you expect instead and why?  Not all algebraic equations
have analytic solutions.  Mathematica can't do this eq as well.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/2fb262a2-cabd-4cd7-b579-b205b721c8ec%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: limit() and assumptions

2015-01-09 Thread Sergey Kirpichev

On Friday, January 9, 2015 at 4:46:53 PM UTC+3, Amit Saha wrote:

 NotImplementedError: Result depends on the sign of sign(log(x)) 

 What does it mean? Can a more user friendly error be given? 


What's unfriendly here?  Very clear error message, IMHO.
 

 I understand that mathematically, if |x|  1, then the above limit 
 should be 0. How can I tell sympy that the absolute value of |x|  1? 


You can't (to be fair, in old assumptions, which we use in the
core).  But you can do something like this:
In [4]: limit((1 + x)**n/n, n, S.Infinity)
Out[4]: ∞

In [5]: limit((1/(1 + x))**n/n, n, S.Infinity)
Out[5]: 0

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/fd43e60d-c927-4735-abab-89834efc64be%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: How about removing unused imports?

2015-01-07 Thread Sergey Kirpichev

On Wednesday, January 7, 2015 1:15:48 PM UTC+3, Joachim Durchholz wrote:

 Hey all, 

 I'd like to work on cutting down the number of unused imports; a sample 
 pull request is on https://github.com/sympy/sympy/pull/8761 . 


If we can first add a test for this in the code quality module - I'm +1.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/f03816cf-5399-43c3-80d7-6dda11e8aca3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: How about removing unused imports?

2015-01-07 Thread Sergey Kirpichev

On Wed, Jan 7, 2015 at 6:31 PM, Joachim Durchholz j...@durchholz.org wrote:
 I'm not sure that that is possible.

That's bad news.

 With any other order, the next person to git pull will get a nasty surprise:
 all their bin/test runs will fail with those newly tested-for errors, so
 they can't rely on that I can make the PR as soon as all tests pass rule
 anymore.

You can break other people PR's in a zillons of ways.  For example,
their pr can rely on removed import.

To be sure that tests pass - they will have to merge with master
first.  That's the rule.

 However, I'd also like to know if such fixes are acceptable if it turns out
 we can't make it an automated test.

Sure, they are.  But at least, please try to minimize number of such PR's.
(I.e., do this for all modules once)

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20150107160856.GA2172%40darkstar.order.hcn-strela.ru.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] new idea

2015-01-07 Thread Sergey Kirpichev

On Wednesday, January 7, 2015 6:13:54 PM UTC+3, Joachim Durchholz wrote:

 Adding a GUI? 
 That's a really large project. 


In some sense, we have GUI.  It is the Ipython Notebook.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/bdcf2a6c-005c-4683-b7ba-b0617b31d2ec%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: transcendental equation solving

2015-01-07 Thread Sergey Kirpichev

On Wednesday, January 7, 2015 10:48:38 AM UTC+3, Harsh Gupta wrote:

 * There is a clear code level and interface level separation between 
 solvers for equations in complex domain and equations in real domain.


To be fair - this is a very subjective sentence.

The clear advantage is new output format, but even here we still have 
unresolved issues.
For example: how to represent solutions for equations with parameters,
like a*x**2+b*x+c.  Perhaps, that fits in Build the set infrastructure 
item.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/547ea289-483b-4476-b005-f541bfd775bf%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: mpmath missing, installation instructions fail

2015-01-07 Thread Sergey Kirpichev

Hello,

On Tuesday, January 6, 2015 11:38:29 PM UTC+3, Joachim Durchholz wrote:

I have to say that my fears have fully become reality: The process of
installing mpmath, while simple, isn't described in a way that just
works, and still invites subtle error.

Lets hope, I can understand this from the following verbose text...

As a developer, I expect to be able to download the git repo and start
working. Possibly after reading a README file.

Can you just literally follow instructions in that file? What's wrong?
I don't see your command history...

* It sends me to http://docs.sympy.org/dev/install.html - erh, a detour.

No. README has all required instructions.

I haven't fully read the readme

Then please do read Installation section before posting such things.

OK, that's just a detour because I'm really after mpmath, not html

We have README in the docs directory. Perhaps, this stuff should
be elsewhere.

How this related to mpmath at all?

However, this would add to the problems a new dev would have to
overcome;

New developers are redirected to
https://github.com/sympy/sympy/wiki/Development-workflow

Everything doc-related is documented here, section Install other software.

maybe it's a good idea to list rsvg-convert among the required
installs in the documentation building docs.

It listed.

The same goes for convert (from imagemagick).

Same.

Indeed, we can improve docs documentation, but that's different issue.

It says I should just start python and run a few import statements.
... aaand I'm getting this:

$ python [...]

ImportError: No module named mpmath.libmp

Did you install mpmath as described in the README.rst?

If no, what do you expect?

The following error occurred while trying to add or remove files in the
installation directory:

[Errno 13] Permission denied:
'/usr/local/lib/python2.7/dist-packages/test-easy-install-15567.write-test'

There's a long explanation and a link... but it does not explain what to
do with a permission denied situation.

As you don't explain what you do before - it's hard to guess what's was
wrong.
Perhaps, first time you did sudo python install ..

Still. I'm not after almost 100 lines of writeup, with several problems
along the way.

IS THIS ALL?

The one problem I see - it's just the specific user, who can't even read
the README.rst. Let's wait for real bugreport before listing something as
a problem.

Exactly one import error - all mpmath's problem you have. The reason
for one is simple: please do install mpmath first. That's the first
sentence of the
installation instructions in the README.

Sergey said it's easy and you should do your reading.

Clearly, you don't. That's a problem.

for a change that never had a clear motive (except making
life easier for packagers

Please read relevant discussions first.

We should have done this with some other dependency.
pep8.py comes to mind

We don't have hard dependences on anything else, including
pep8. Right now, we don't use pep8 at all.

--
You received this message because you are subscribed to the Google Groups
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/5ccad9df-f8ef-4a55-80fd-5e31f65549d7%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: transcendental equation solving

2015-01-06 Thread Sergey Kirpichev

On Tuesday, January 6, 2015 7:16:42 PM UTC+3, Luv Agarwal wrote:

 In GSoC last year there was a project on improvising solvers. I wanted to 
 know how much of that is implemented (the ideas page of this year is not 
 updated yet) as I am interested in working on solvers.


See  https://github.com/sympy/sympy/pull/7523 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/e0a71966-c410-4c3f-ab74-4450687dab1c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: What's the consensus about multiples-of-four indentation?

2015-01-02 Thread Sergey Kirpichev

On Friday, January 2, 2015 11:39:09 AM UTC+3, Joachim Durchholz wrote:

2a) Should the rule be applied to continuation lines?
Typical case as found by me in the codebase:
a = self.fubar_the_worg(A, x) \
+ make_me_happy(x, y, z)

No errors here.

And this other category of continuation lines:
klaatu = self.barada_nikto(x, y, z,
a, b, c)
klaatu = self.barada_nikto(x, y, z,
a, b, c)
klaatu = self.barada_nikto(
x, y, z, a, b, c)

FYI: The only error here is the second example.

2b) Should the rule be applied to comment lines?
Typical situations in our code base:
blah() # Here we need to blah()
# to avoid frobnication
blub(a, x, d, w + x + y + z)
# w2 + x + y + z

E116. Please note, that this error is not enabled in my
PR (and also E261 on the first line, that's
disabled too).

If pep8.py can be made to do the tests in the way we want them done,
should we use it?

Sure. SymPy is about symbolic computer algebra, stylistic validators
out of the subject.

4) How do we want to handle external dependencies in general?
E.g. pep8.py is currently exhibiting a bug that makes it unsuitable for
our purposes (this is likely to get fixed in the near future).

Just like we handle them now (or do this other projects, whatever). I.e. -
we should
specify version requirements.

4a) Should we simply put a copy of the external dependency into SymPy?

We already have discussion about this while unbundling mpmath. The end
of it, I think - that this is bad practice in general (though we can have
reasons to do this).

4c) Installing dependencies through pip is not a hassle if SymPy itself
is installable through pip. Right now, SymPy is not prepared for that.

What's wrong with this?

$ pip install sympy
Downloading/unpacking sympy
Downloading sympy-0.7.6.tar.gz (6.4Mb): 6.4Mb downloaded
Running setup.py egg_info for package sympy

Installing collected packages: sympy
Running setup.py install for sympy
changing mode of build/scripts-2.7/isympy from 640 to 755

changing mode of /home/sk/.virtualenvs/sympy1/bin/isympy to 755
Successfully installed sympy
Cleaning up...
$ isympy --version
0.7.6

Please do trivial checks for your claims.

4d) Should the code that needs the dependency (sympy for mpmath,
test_code_quality for pep8.py) run pip install?

Not in the case of test_code_quality.py. test_*.py - for testing,
not for installation of dependencies.

--
You received this message because you are subscribed to the Google Groups
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/147a8d4b-44b8-4dc5-9497-87017b2113b5%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Unbundle mpmath

2015-01-02 Thread Sergey Kirpichev

FYI: now #2192 was merged.

Please be aware about new installation instructions for SymPy:
https://github.com/sympy/sympy#installation
https://github.com/sympy/sympy/wiki/Development-workflow#install-mpmath

On Friday, June 28, 2013 5:42:07 PM UTC+4, Sergey Kirpichev wrote:

 There is a long term issue [1] to make mpmath an external dependency.  It 
 was suggested by
 Aaron (see discussion in PR [2]), that we should discuss this on the 
 mailing list.

 Perhaps, the main reason to prefer bundled mpmath copy is a
 mystical complex installation.  Not clear enough.  For example, I don't
 think that additional steps for package installation from mpmath manual 
 does matter.  For non-root users
 there shouldn't be any problems too, at least for Python 2.6+ (per-user 
 site-package directory).

 [1] http://code.google.com/p/sympy/issues/detail?id=2482
 [2] https://github.com/sympy/sympy/pull/2192 
 https://www.google.com/url?q=https%3A%2F%2Fgithub.com%2Fsympy%2Fsympy%2Fpull%2F2192sa=Dsntz=1usg=AFQjCNGaQLh2_zejpGZGtkqYXkB6EDT-DA



-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/6a2481a1-7831-46a5-acce-01f9f2ad3b73%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Wow. (was: Add pep8 config, use pep8 in travis (#8538))

2015-01-02 Thread Sergey Kirpichev

On Friday, January 2, 2015 10:37:40 AM UTC+3, Joachim Durchholz wrote:

 Wow. Locked out without fair warning. No community consensus either. 


Yes.  I would like to see constructive discussion in that thread.  When
(and if) consensus will be reached among collaborators, we can unlock the
thread again.  After all, this PR is not so important. 

Am 01.01.2015 um 20:00 schrieb Sergey B Kirpichev: 
  On Thu, Jan 01, 2015 at 09:33:20AM -0800, Joachim Durchholz wrote: 
  You can't claim that you're impartial after all. 
  
  Why not? 

 Conflict of interest. Duh. 


I don't have other interests here, except for making SymPy better.  And we 
have
a clear policy on how to do this:
https://github.com/sympy/sympy/wiki/Development-workflow#requirements-for-inclusion

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/ffec5b8b-ee6e-4cfc-a6b5-23c4f2834c61%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Want to contribute to SymPy

2014-12-30 Thread Sergey Kirpichev

https://github.com/sympy/sympy/wiki/introduction-to-contributing

On Tuesday, December 30, 2014 8:28:02 PM UTC+3, Karan Mangla wrote:

 Hi 
 I am Karan.I am a student of Computer Science and Engineering at 
 IIIT-Hyderabad(India). I am a Mathematics and Science enthusiast . 
 I am new to SymPy but have worked previously with python on a couple of 
 projects . I want to contribute to sympy. 
 Please could anyone help me in getting started.

 Thanks .


-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/733243fb-f881-4be0-aced-2004a7d561b1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: using Interval method

2014-12-23 Thread Sergey Kirpichev

On Tuesday, December 23, 2014 9:30:28 AM UTC+3, Gaurav Dhingra wrote:

  from sympy.sets.sets import Interval 
  Interva(0, 1) 
 [0, 1] 


Pretty-printing is not an exact representation of objects.

But in this case, [0,1] returned by repr() - I think this is
a bug:
In [2]: print(str(Interval(0, 1)))
[0, 1]
In [3]: print(repr(Interval(0, 1)))
[0, 1]
In [4]: print(pretty(Interval(0, 1)))
[0, 1]
In [5]: print(latex(Interval(0, 1)))
\left[0, 1\right]
 

  isinstance([0, 1], Interval) 
 False #returns False 

 should not it return True' 


[0, 1] is a python list.  Please read https://docs.python.org/3/tutorial 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/73f820a4-edba-4347-a5e2-09dc4a8a5b34%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] A modest proposal for typed expressions

2014-12-21 Thread Sergey Kirpichev

On Sunday, December 21, 2014 7:39:37 AM UTC+3, Richard Fateman wrote:

 It seems to me this can cause problems if, inside some scope, you need
 to create or manipulate SymPy symbols  for local
 usage.  (It can be done by gensym in Lisp, so maybe you do this?)


You can use Dummy: 
x1 = Dummy('x')
x2 = Dummy('x')

x1 and x2 - different objects.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/523d11f2-6442-49e9-a2d6-2fa623048cf3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] A modest proposal for typed expressions

2014-12-19 Thread Sergey Kirpichev

On Friday, December 19, 2014 2:47:44 AM UTC+3, Richard Fateman wrote:

 Please forgive me if this is off track -- I've never used assumptions in 
 sympy, and
 don't know much about sympy implementation.


Actually, we have two:
sympy/core/assumptions.py (old one)
and
sympy/assumptions/ (new)

Lets consider old assumptions first.
 

 Can any program variable in python have assumptions?
 Or are declared vars what can have assumptions?


Python variables can't have assumptions.  Values, objects, in this case 
Symbol's - can.

 v = Symbol('r', real=True)
 v.is_real
True
 v.assumptions0
{'commutative': True, 'complex': True, 'hermitian': True, 'imaginary': 
False, 'real': True}

Objects are immutable and every Symbol is created with some
set of assumptions which determine certain properties of symbolic objects
and can have 3 possible values: True, False, None.

We can deduce assumptions about expressions, like this:
 (v**2).is_nonnegative
True

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/27ea714b-da27-4d03-9ea6-081ecc52f39f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Help to choose a first task

2014-12-18 Thread Sergey Kirpichev

Hello,

https://github.com/sympy/sympy/wiki/introduction-to-contributing

On Friday, December 19, 2014 3:13:56 AM UTC+3, Willian Moreira wrote:

 Hello,

 I am Willian, a CS student from Brazil. I want to start contributing to 
 sympy developing. I got the code from the git repo, and now I am trying to 
 find some bug or new feature that is suitable for a newcomer. Can you 
 please indicate me something?

 Thanks in advance
 Willian Moreira



-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/a1d59128-d0d6-4c30-baed-82fdb54550f5%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: How to run Sympy Test on a module or patch.

2014-12-15 Thread Sergey Kirpichev

https://github.com/sympy/sympy/wiki/Development-workflow#be-sure-that-all-tests-of-sympy-pass

see also file bin/test in the sympy's sources:
./bin/test --help

On Monday, December 15, 2014 2:16:02 PM UTC+3, Adrián Ashley wrote:

 Hello Everyone!

 I am new to unit testing. It would be great if anyone could tell me how to 
 run sympy tests, so as to test a patch.

 Thanks in advance!


-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/a608d0b3-3c23-4740-927b-4daf58e179ee%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Wish to contribute to SymPy

2014-12-14 Thread Sergey Kirpichev

https://github.com/sympy/sympy/wiki/introduction-to-contributing

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/d05c3376-7a01-4fbb-bbac-cf3a909879d8%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Error in symbolic integration

2014-12-10 Thread Sergey Kirpichev

On Wednesday, December 10, 2014 3:28:44 PM UTC+3, ali...@gmail.com wrote:

 When integrating with sympy something goes wrong - maybe a bug:
 x = Symbol('x')
 s = (cos(x)**2 / (1 + sin(x)**2))
 integrate(s, (x, 0, pi))

 the answer from sympy is -pi 


From which version of the sympy?  I see unevaluated integral in the master.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/30e646de-5704-4428-861f-df5487171af4%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Factor options

2014-12-08 Thread Sergey Kirpichev

In [6]: factor(x**4 + 1, extension=[I, sqrt(2)])
Out[6]: 
⎛  ___ ___  ⎞ ⎛  ___ ___  ⎞ ⎛  ___ ___  ⎞ ⎛  
___  
⎜╲╱ 2╲╱ 2 ⋅ⅈ⎟ ⎜╲╱ 2╲╱ 2 ⋅ⅈ⎟ ⎜╲╱ 2╲╱ 2 ⋅ⅈ⎟ ⎜╲╱ 
2   
⎜x - ─ - ───⎟⋅⎜x - ─ + ───⎟⋅⎜x + ─ - ───⎟⋅⎜x + 
─ +
⎝  22   ⎠ ⎝  22   ⎠ ⎝  22   ⎠ ⎝  
2

   ___  ⎞
 ╲╱ 2 ⋅ⅈ⎟
 ───⎟
2   ⎠

On Monday, December 8, 2014 9:03:25 PM UTC+3, Paul Royik wrote:

 This is good method, but it doesn't work for x^4+1


-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/24d952bd-bde3-4860-8840-12337fec4f62%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: GitHub multiple CI services

2014-12-08 Thread Sergey Kirpichev

On Monday, December 8, 2014 11:41:00 PM UTC+3, Aaron Meurer wrote:

 It also 
 means we can take a second look at 
 https://github.com/sympy/sympy/pull/1960 
 https://www.google.com/url?q=https%3A%2F%2Fgithub.com%2Fsympy%2Fsympy%2Fpull%2F1960sa=Dsntz=1usg=AFQjCNENjYJXohmibXzOcZHXI6cqSlNeAg.
  



Probably, using py.test could be a separate issue.

+1 to add some coveralls support.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/5962fe5b-007a-471d-8d00-7aa2538fe0ba%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] SymPy 0.7.6 is released

2014-11-25 Thread Sergey Kirpichev

On Tuesday, November 25, 2014 2:53:05 AM UTC+3, François wrote:

 Thank you. I had finally worked it out from sympy's github repo. I 
 guess 0.18 + 1 commit not yet in mpmath.


Yes there are some commits in bundled mpmath's tree which exists
only in newer mpmath versions. 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/a508bf24-d1d1-410b-aa87-ece068ea703f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Python 2.6 support

2014-11-24 Thread Sergey Kirpichev

On Sunday, November 23, 2014 10:22:23 PM UTC+3, Ondřej Čertík wrote:

 Btw, IPython also doesn't support Python 2.6, and I bet IPython is 
 used by more people than SymPy, so if it is good for IPython, I think 
 it will be good for us as well. 


IPython doesn't support 3.2 too.  Should we drop everything, except 2.7 and 
= 3.3?

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/9384524f-9033-43d4-83d9-43d4a52ac0a1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Issue #7335: FiniteSet not accepting iterables

2014-11-22 Thread Sergey Kirpichev

On Sunday, November 23, 2014 5:21:42 AM UTC+3, Amit Saha wrote:

 Hi, 

 I only am realizing that I cannot do FiniteSet[1, 2, 3, 4]) any more. 
 I did find it very useful to create a set from the elements of a set. 
 Anyway, it's probably way late for that. 

So, what is the recommended way now to create a FiniteSet from an 
 existing list or any other iterable? 


FiniteSet(*[1,2,3,4]) or FiniteSet(*set([1,2,3]))


 PS: The release note for 0.7.6 does not contain an entry for this change. 


Indeed.  Fixed.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/5b20cfb2-eeec-4a02-960b-6e68bac6953d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Release notes in the wiki vs with the source

2014-11-21 Thread Sergey Kirpichev

See this thread:
https://groups.google.com/d/msg/sympy/qkP3WAuWwbc/k3mi2NDea9MJ

In my view, we should adopt some tagging for merge commits (e.g. with pr
headers). This
can help us with automatic generation of the release notes. Something like
the scipy project.

Take for example our last release notes or scipy's
doc/release/0.15.0-notes.rst. Both have:
1) new features
2) bugfixes
3) deprecations
4) backward-incompatible changes
We should divide our pr's to such or similar categories.

Also, list of closed issues in the given release - great feature for next
changelog.

Automatic generation of the release notes will be trivial if we adopt the
above
mentioned tagging and this syntax for closing issues:
https://help.github.com/articles/closing-issues-via-commit-messages/

Sure, it will require a manual intervention from the release manager, but
others can help.

We can add old release notes retroactively (at least, for the
code history in our git tree).

On Friday, November 21, 2014 7:12:14 PM UTC+3, Jason Moore wrote:

We keep track of our release notes on the wiki. To me, it seems like these
should be in a file included with the source code so it always travels with
the git repo and our source distributions. It would also be nice if every
pull request with significant changes had to include a change to the
changelog file. A check to see if the changelog file was modified in a PR
could be done on with travis too.

Are there reasons that the wiki is preferred?

Jason
moorepants.info
+01 530-601-9791

--
You received this message because you are subscribed to the Google Groups
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/d9d4490b-a339-484a-bbfb-16c24d3e46e1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] SymPy 0.7.6 release candidate 2

2014-11-20 Thread Sergey Kirpichev

On Thursday, November 20, 2014 12:09:46 AM UTC+3, Aaron Meurer wrote:

 Thanks for everyone for testing this. I haven't heard of any issues so I 
 will do the final release tomorrow. If anyone else was planning on testing 
 this, please do so now.



Are we sure that our benchmarks work:
 https://github.com/sympy/sympy/pull/8361#issuecomment-62064249
?

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/3df62b34-e4ee-48b1-89e6-e190f2a7f74d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: SymPy plans for the future: 1.0 on the horizon

2014-11-20 Thread Sergey Kirpichev

On Friday, November 21, 2014 12:49:40 AM UTC+3, Aaron Meurer wrote:

A few I'd like to propose:

New solve is another important stuff for 1.0, see
https://github.com/sympy/sympy/pull/7523

- mpmath should be made a dependency of SymPy (it should no longer be
bundled). Let's wait until the end of the year to actually make this
change, in case a bugfix 0.7.6.1 or 0.7.7 release is necessary it will
be easier.

https://github.com/sympy/sympy/pull/2192#issuecomment-63891482

- This is not a necessity for the release, but the performance issue
has come up several times. In my view, the most important thing we can
do here is set up a benchmarking system, so that we can really
understand why SymPy is slow.

It's important, but not as important for 1.0 as API stability.

We should be hesitant to adding more required dependencies. I think
that unbundling mpmath is a good thing, but don't want SymPy to go
from zero dependencies to ten dependencies. Let's start with mpmath
and see how that goes.

I don't know why you don't like dependencies so much. It's easy
to install any deps without root privileges with virtualenv pip:
https://groups.google.com/d/msg/sympy/BCOn8oOyUV4/I9zJV7k03zgJ

At least, there is no practical issues to do so for development
dependencies (e.g. py.test instead of home-grown testing framework).

--
You received this message because you are subscribed to the Google Groups
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/b2132735-8adf-48a9-bacf-99699fc70d29%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: new to Sympy

2014-11-06 Thread Sergey Kirpichev


https://github.com/sympy/sympy/wiki/introduction-to-contributing

On Thursday, November 6, 2014 3:57:01 PM UTC+3, Ankush wrote:

 Hi everyone,
   I am new to sympy and really interested in this topic. I 
 want to contribute in this community. Can anybody guide me how to proceed 
 further.

   Thanks,
   Ankush


-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/6a330c7b-4b91-42d6-bebe-ef28914b54d3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] A goof-up in symbolic math software

2014-10-31 Thread Sergey Kirpichev


On Friday, October 31, 2014 1:49:23 AM UTC+3, Richard Fateman wrote:

 If they knew anything about debugging and SymPy, which is not so probable.


The point is - it's ultimately improbable for closed-source program.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/03eec8f7-c952-41c4-bbc9-a3b963d2ef15%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: sympy and IVP

2014-10-15 Thread Sergey Kirpichev



On Thursday, October 16, 2014 12:08:30 AM UTC+4, ali...@gmail.com wrote:

 I have a problem in solving IVP's with sympy [...] Do you have any hints?


We don't support initial conditions yet (issue 4720).  Please read 
documentation.

However, in this case you can find a generic solution for this ODE with 
dsolve.  Then, just
substitute initial values and find unknown coefficients with solve.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/8c6f8052--466b-b3dc-31de9249f28a%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Why 1/3 == S(1)/3 is True ?

2014-10-03 Thread Sergey Kirpichev



On Thursday, October 2, 2014 5:02:10 PM UTC+4, Christophe Bal wrote:

 For me the fact that  1/3 == S(1)/3  has value  True  sounds like a bug 
 because  1/3  is a float, and  S(1)/3  a rational.


Looks as a bug, please report one on https://github.com/sympy/sympy/issues 
(through please search first for similar issues).

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/96c64d5e-b2f4-4f43-9b17-99ff5b5a9322%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: git rebase question

2014-10-01 Thread Sergey Kirpichev

On Thursday, October 2, 2014 12:46:25 AM UTC+4, Colin Macdonald wrote:

 [replying to maillist instead of issue tracker]

 Would this latter case include cleaning up a (sordid) history that has 
 approaches
 tried but latter reverted?


It's ok, but only as a last resort, after the review was 1) done and 2) is 
positive.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/99f3496e-db67-4878-9e93-3701ebda4836%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Find a issue about the doc

2014-09-29 Thread Sergey Kirpichev


On Monday, September 29, 2014 12:18:33 PM UTC+4, Garvin Li wrote:

 When I clone the project,and try tomake html.There is something 
 wrong happened.


Yes.  You forgot to read the docs first )

 

 I think many computer may not have the rsvg-convert
 why not add a judgement about it ,just like if rsvg-convert  


Because it's not an optional requirement.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/ff9e1faf-0e92-47c6-a670-6db321f381cb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: elliptic_e._eval_conjugate for one arg?

2014-09-28 Thread Sergey Kirpichev



On Sunday, September 28, 2014 5:15:06 AM UTC+4, Chris Smith wrote:

 elliptic_e can take 1 or two args but the _eval_conjugate assumes 2 (and 
 fails in a branch I am working on). The following mod works without 
 returning anything for the 1 arg case, but can someone tell me if I should 
 be uncommenting the 2 lines that I added for the 1 arg case?


Should be:
def _eval_conjugate(self):
if len(self.args) == 2:
z, m = self.args
if (m.is_real and (m - 1).is_positive) is False:
return self.func(z.conjugate(), m.conjugate())
else:
z = self.args[0]
if (z.is_real and (z - 1).is_positive) is False:
return self.func(z.conjugate())

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/2d116ae9-e7b9-4aea-b8ec-cdfb4ab727d2%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: elliptic_e._eval_conjugate for one arg?

2014-09-28 Thread Sergey Kirpichev



On Sunday, September 28, 2014 2:18:42 PM UTC+4, Kalevi Suominen wrote:


 The one-argument form of elliptic-e is useless (in is present form) 
 because the modulus has no default value.


Oh, no.  It's just the Legendre complete elliptic integral of the second 
kind.  Please read the docs. 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/38575ccf-f48a-4b93-9b69-4977044c015b%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Metaclass and comparison ops

2014-09-05 Thread Sergey Kirpichev

I see that some strange things happens for classes, that define comparison 
ops AND use metaclasses.  Sometimes, the interpreter do flip of operands 
and executes e.g for a  b first b.__gt__(a), not a.__lt__(b).

And example:
https://github.com/sympy/sympy/issues/7951
Simple tests:
https://github.com/sympy/sympy/issues/7951#issuecomment-54514684

Any thoughts?

BTW, why we need the metaclass stuff for old assumptions?  Performance?

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/ef2afa6e-3838-4730-a49d-38ce1499d0e3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] [RFC] what test to add for an issue

2014-09-03 Thread Sergey Kirpichev

On Wednesday, September 3, 2014 6:44:06 AM UTC+4, Aaron Meurer wrote:

 I think such situations are OK, as long as you understand why the test 
 failed before and verified that the new test really tests the same 
 thing. 


1. We can make mistakes here.
2. Probably, we want to be sure, that we never break again the
fixed example (incl. by different reason).  How we can be sure about
this, if we don't preserve the original test?

 I noticed that the slow test was taking about 2 minutes.

Can't we exclude slow tests with some cli switch?

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/49cad101-0f4a-4f99-b300-7b2549250f59%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: A plan for the assumptions

2014-08-30 Thread Sergey Kirpichev

On Sunday, August 24, 2014 9:46:40 PM UTC+4, Aaron Meurer wrote:

Some of you may have seen Sergey started a pull request to remove uses of
the new assumptions in the code, because it's confusing to users who set
the old assumptions and it doesn't work.

BTW, if we decide to use this approach - I think that change is a good
candidate for the next release.

So, please help with review.

You make a good point. However, if you are going to spend time on this, it
would be better spent improving the new assumptions (and let's finally
merge my branch as a start to this), or merging the two. I still maintain
that this pull request is a step backwards.

We must offer for users a consistent interface. But if some parts use
foo.is_real assumption, and some ask(Q.real(foo)) - this is not the case.
Second, I don't think there are regressions in the functionality of sympy.

With these changes, the new assumptions aren't used at all, meaning they
are tested much less, and more importantly, there is no more motivation to
improve them.

We still have refine* helpers, we have some code in the unify module. And
we have our tests for the assumptions module. Also, it seems we have bad
test coverage for the code, changed in my pr.

The old assumptions syntax isn't going away, so why don't we focus on
making it so that the new assumptions can read the old assumptions.

Yes, but this variant is much more complex then my PR. Basically, we must
first make our assumptions identical in both systems.

The first step though will be to go through all the assumptions and make
sure they are identical in all systems. This is a good opportunity to clean
up some of the contradictory or unexpected assumptions, so it should
involve some community input. So I would do it on the mailing list.

I think, first we should cleanup and document (see
https://github.com/sympy/sympy/pull/7936) old assumptions. Then freeze old
assumptions. No new facts, no new features - only bugfixes. First step.

- [ ] Go through all the assumptions facts and make a note of what is
different, and also what is confusing. A lot of this is already written
down in various places (https://github.com/sympy/sympy/issues/5976,
https://github.com/sympy/sympy/pull/2538, probably some of the issues
mentioned in https://github.com/sympy/sympy/issues/6730,
https://github.com/sympy/sympy/pull/2508/files#diff-e0f95401c86d07d90903363122990de0R172,

https://github.com/sympy/sympy/wiki/assumptions_handlers), but we should
gather them all down in one place. I think the main thing will be how we
handle infinity.

We must do all this in the old assumptions first (my first step above).

- [ ] Discuss with the community on the mailing list how we should handle
the differences, until we come to a consensus.
- [ ] Make the new assumptions call the old assumptions first. Also make
the handlers access the old assumptions, so that everything sees
`Symbol('x', real=True)` as real. This will be the hard part, and lead to
lots of test failures. It might be worth trying this first, to really see
what facts are different in the two systems.

Second step. We can make the new assumptions call the old assumptions on
symbols (i.e. ask(Q.real(x)) returns True for x = Symbol('x', real=True)).
And we can start add new facts in a way, fully compatible with the old
system.

Only at this stage we can start experimenting with new assumptions in the
sympy codebase (i.e. not only in sympy.assumptions).

I don't think there's a big issue with this, because already
`ask(Q.real(x))` and `x.is_real` are different. My proposal is to have the
first call the second but not the other way around.

That's impossible now. Simple example: x.is_real = Q.extended_real(x), not
Q.real.

--
You received this message because you are subscribed to the Google Groups
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/cac2dc56-5021-43ab-ba97-83b3b4699f81%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: A plan for the assumptions

2014-08-27 Thread Sergey Kirpichev

 Or I don't understand your problem [about the problem with calling for
 arg in the new assumptions handler]

 The problem is that arg is written using old assumptions, so unless you
 rewrite arg to use the new assumptions and pass assumptions along to it,

Yes.  In particular, if we add a compatibility layer (is_foo predicates
call the new assumptions).  Nothing new here, of course.

 x has no direct assumption so False will be returned from foo and thus 2
 will result from the ask. So unless we either 1) assign assumptions to
 symbols (but that will not be sufficient to handle things like Q.positive(x
 + 2)

Or add new global assumptions in the Symbol constructor (see my last PR).  Of
course, this is sufficient to handle Q.positive(x + 2):

In [1]: p = Symbol('p', positive=True)

In [2]: ask(Q.positive(p + 2))
Out[2]: True

In [3]: ask(Q.positive(p - 2))

In [4]:

 or 2) start passing assumptions to all functions which have a
 dependence on assumptions (so something like def foo(e) - def foo(e,
 assumptions={})) I don't know how we are going to get the
 old-assumptions-API-based functions to start working with the new
 assumptions. What am I missing? How will foo above need to be modified so
 it works?

To alter the global_assumptions context we have context managers.

An example from my branch:

In [10]: r = Symbol('r')

In [11]: ask(Q.real(r))

In [12]: with assuming(Q.real(r)):
...: print(ask(Q.real(r)))
...:
True

In [13]: ask(Q.real(r))

In [14]:


But, optionally, we can provide also assumptions kwarg to override
the current assumptions context.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20140827102157.GA3938%40debian.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: SymPy 0.7.6 release

2014-08-27 Thread Sergey Kirpichev


On Wednesday, August 27, 2014 8:10:31 AM UTC+4, Aaron Meurer wrote:

 Now that GSoC is over, let's start thinking about doing a 0.7.6 release. 
 Are there any major pull requests from GSoC that are still to be merged?


Avichal Dayal's pull requests aren't merged yet:
https://github.com/sympy/sympy/pulls/avichaldayal

Probably, mrvasympt pr could be merged in this release, but there is still 
a problem to investigate.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/7510dda3-ce9f-4b91-a673-23fe1f085f6d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: A plan for the assumptions

2014-08-26 Thread Sergey Kirpichev

On Tuesday, August 26, 2014 9:32:29 AM UTC+4, Chris Smith wrote:

 I know that the new system can handle is_true expressions and can handle 
 assumptions involving Or (while the old assumptions only allow And-based 
 assumptions: var('x', integer=True, positive=True).


Not just Or-ed, but the new system allow to specify arbitrary assumptions 
on expressions.  Something like Q.positive(log(x)).
 

 One thing I noted today was that the old system can make a calculation 
 based on arg(expression) which depends on the old assumptions. I don't know 
 how that could be done in the new system.


I think, exactly as in the old one.  No difference.  But, you will have to 
use different syntax to query assumptions while computing arg(expr).  
ask(Q.positive(foo)) instead of foo.is_positive and so on.

Or I don't understand your problem.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/0e6fe317-00fb-4a61-a52f-8627b4324ec3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: A plan for the assumptions

2014-08-25 Thread Sergey Kirpichev

On Sunday, August 24, 2014 9:46:40 PM UTC+4, Aaron Meurer wrote:

It's a much more difficult task then the simple pr in question. But I did
some attempts
in the past, see my branches assumptions-merge-new-old-DRAFT and
assumptions-merge-new-old-DRAFT-REBASED (recent rebase).

With these changes, the new assumptions aren't used at all, meaning they
are tested much less, and more importantly, there is no more motivation to
improve them.

I don't think they are much tested now. For example, I have seen only one
test failure after my modifications in the sets module.

I fail to see reasons to keep such confusing code in the master. We should
use one interface for the assumptions. In fact, this can simplify the
mirgation too.

The old assumptions syntax isn't going away, so why don't we focus on
making it so that the new assumptions can read the old assumptions.

This is what @mrocklin and I attempted to do last summer (
https://github.com/sympy/sympy/pull/2210). It didn't work, because we
tried to do a two-way merge of the assumptions, i.e., so that both
`x.is_integer` and `ask(Q.Integer(x))` were always identical (the logic I
used was https://github.com/sympy/sympy/pull/2210/files#r5463892, which
required both assumptions to be the same).

Maybe an easier approach would be along the lines of what I have done at
#2508. Have ask call a bunch of things, in order, until it gets an answer.
In my branch, `ask` calls the current new assumptions, and then my satask.
We could make it so that ask first calls the old assumptions, then the new
assumptions. Furthermore, the new assumptions should gain trivial handlers
for Symbols to so that `ask(Q.real(Symbol('x', real=True)))` is True.

So I what I would recommend doing:

- [ ] Merge my branch #2508. We can maybe wait until after the release to
do this so it is minimally disruptive, but this is important, as it will be
much easier to add new facts to the new assumptions with satask (and I also
want people to start thinking about how to improve the speed of it).
- [ ] Go through all the assumptions facts and make a note of what is
different, and also what is confusing. A lot of this is already written
down in various places (https://github.com/sympy/sympy/issues/5976,
https://github.com/sympy/sympy/pull/2538, probably some of the issues
mentioned in https://github.com/sympy/sympy/issues/6730,
https://github.com/sympy/sympy/pull/2508/files#diff-e0f95401c86d07d90903363122990de0R172,

https://github.com/sympy/sympy/wiki/assumptions_handlers), but we should
gather them all down in one place. I think the main thing will be how we
handle infinity.
- [ ] Discuss with the community on the mailing list how we should handle
the differences, until we come to a consensus.
- [ ] Make the new assumptions call the old assumptions first. Also make
the handlers access the old assumptions, so that everything sees
`Symbol('x', real=True)` as real. This will be the hard part, and lead to
lots of test failures. It might be worth trying this first, to really see
what facts are different in the two systems.

Note that absent from my plan is making `x.is_real` call the new
assumptions. This is what I tried at #2210, and it failed, because it
requires everything to work perfectly, and there are so many places that
the assumptions are used that it's too much work to get it all in one go.
We should get the above working first before we consider this. I don't
think there's a big issue with this, because already `ask(Q.real(x))` and
`x.is_real` are different. My proposal is to have the first call the second
but not the other way around.

Aaron Meurer

--
You received this message because you are subscribed to the Google Groups
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/d19cead7-0bcc-4377-a502-5f9780f49e41%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Benefit of XFAIL over SKIP?

2014-08-24 Thread Sergey Kirpichev

On Sat, Aug 23, 2014 at 11:36 PM, Aaron Meurer asmeu...@gmail.com wrote:
 Maybe it would help to finish implementing 
 https://github.com/sympy/sympy/issues/7044. The biggest issue with XFAIL 
 is that there will be some silly error in the way the test is written so 
 that it never XPASSes, even when it should.

We can add some documentation for developers (how to write
XFAIL'ed tests).  The simplest case of such errors - when we can't
be sure in the exact form of the answer.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20140824134858.GA3684%40debian.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Benefit of XFAIL over SKIP?

2014-08-07 Thread Sergey Kirpichev


On Thursday, August 7, 2014 7:52:26 PM UTC+4, Ondřej Čertík wrote:

 On Thu, Aug 7, 2014 at 9:49 AM, Harsh Gupta wrote: 
  What is the is benefit of all these XFAIL'ed tests? 
  
  A passed XFAIL'ed test is reported. Just as a seemly unrelated change 
  can make a test fail, a seemly unrelated change can also make the test 
  pass and it important to be aware both the cases. 

 Though, in practice, I don't know if anyone is using this feature.


I use :)  Well, sometimes.

There may be be a better way.  E.g. we can add these tests as specially 
formatted
comments for related issues.  And we can automate testing of the master for
such tests with some bot...  AFAIR, something like this was suggested by 
Harsh.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/fe56f692-548e-4f0c-990e-5b3a143f3c1c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: summation over tuple of symbols

2014-08-04 Thread Sergey Kirpichev

Hmm.  I have no idea of what you want, but...

In [32]: A = IndexedBase('A')

In [33]: n=Symbol('n', integer=True)

In [34]: Sum(A[n],(n,0,2))
Out[34]: 
  2   
 ___  
 ╲
  ╲   A[n]
  ╱   
 ╱
 ‾‾‾  
n = 0 

In [35]: _.doit()
Out[35]: A[0] + A[1] + A[2]

On Monday, August 4, 2014 9:30:38 AM UTC+4, Nova Munch wrote:

 hello.

 i have a tuple of symbols, a where a is (a_0, a_1, a_2, a_3, a_4)

 i want to call:
  summation(a[i], (i, 0, 4))

 but python returns a TypeError that the indices must be integers, not 
 Symbol. 

 this is more of a python constraint instead of Sympy. but is there some 
 workaround to it?

 thanks folks!



-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/46c8adbf-4070-44ac-b0c8-720612b48a4e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: unexpected behavior with partial derivatives of order greater than one

2014-07-29 Thread Sergey Kirpichev

Looks as a bug for me.

Our bugtracker:
https://github.com/sympy/sympy/issues

On Tuesday, July 29, 2014 1:29:59 AM UTC+4, J. S. Oishi wrote:

 Hi all,

 I'm using sympy 0.7.4.1, and I've come across an issue I don't quite 
 understand. If I do

 from sympy import *
 x,z,t = symbols('x, z, t')

 psi = Function('psi')(x,z)

 j1 = Derivative(psi,x) - Derivative(psi,z)

 j1.subs(Derivative(psi,x), t)

 I get what I expect, t - d psi/dx. However, if I do the same thing with a 
 second derivative with respect to z, I get the following:

 j2 = Derivative(psi,x) - Derivative(psi,z,2)

 j2.subs(Derivative(psi,x), t)

 it returns t - d^2 t/dz^2. I expect it to return t - d^2 psi/dz^2. Can 
 anyone point me to anything I might be doing wrong? I've tried defining t 
 as a Function, rather than as a symbol (t = Function('t')(x,z)), but that 
 does the exact same thing. I've also tried writing the second derivative in 
 j2 as Derivative(psi,z,z), but that also does the same thing (which I would 
 expect).

 I've found this behavior both on my computer with Python 3.3.2 and on the 
 sympy live shell on sympy.org, which is currently SymPy 0.7.4.1 (Python 
 2.7.5).

 Any advice you could provide would be very helpful.


 thanks,


 Jeff


-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/be250a32-5198-4f24-b538-3763f3efad1f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Draft for a new pattern matcher

2014-07-19 Thread Sergey Kirpichev

On Monday, July 14, 2014 11:36:19 AM UTC+4, F. B. wrote:

 x = Wild('x')
 x.is_integer().OddQ()


Hmm.  What's wrong with x = Wild('x', integer=True, odd=True), like Symbol?

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/52b83187-763d-44e7-92da-a424e2922560%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Implementation of equation solvers.

2014-07-02 Thread Sergey Kirpichev

On Wednesday, July 2, 2014 5:32:56 PM UTC+4, Harsh Gupta wrote:

 The ideas is to see equation solving techniques as transformation on 
 equations
 which takes an equation as input and produces a *simpler* equation which 
 has
 the same solutions as the input.


Sounds good so far.

But please keep in mind that some of your examples do not satisfy this 
requirement.  E.g. IF.

- Search method that will that decide the order of transformations and 
 produce
   the trivial equation.


Most tricky part.  But we can adopt the current heuristic.
 

 - A Trivial equation solver.


I don't see why we should restrict yourself here.  Why not keep some simple 
pattern-based database, like dsolve?  E.g.:
1) f(x) where f is in a given set of functions (sin, tan...)
2) rational function of x

 

 Here we can emulate our current method of solving.

 ```
 def solve_univariate(f, x):
  Returns a trivally solvable equation for f 
 if f.is_function:
 f = IF(f)
 if is_trigonometric_function:
 f = WT(f)
 .

 if is_trivial_equation(f):
 # some additional conditions have to be entered here
 # to avoid infinite recursion
 return solve_univariate(f)
 ```


I don't see how this will work for simple poly(sin(x))==0 equation.  Unless 
you miss the decomposition step, as mentioned by Aaron.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/932d297c-6b9e-45e8-95ff-0d4656806dae%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Erroneous series expansion of tan(x)

2014-06-11 Thread Sergey Kirpichev

This does work in my order-point branch, see 
https://github.com/sympy/sympy/pull/2427

In [1]: tan(x).series(x,pi/2,10)
Out[1]: 
 35  7
9   
  ⎛π⎞  ⎛π⎞⎛π⎞  ⎛
π⎞
  ⎜x - ─⎟2⋅⎜x - ─⎟⎜x - ─⎟2⋅⎜x - ─⎟ ⎛   
10 
1 π   ⎝2⎠  ⎝2⎠⎝2⎠  ⎝2⎠x⎜⎛
π⎞   
- ─ - ─ +  + ── +  + ── + ─ + O⎜⎜x - 
─⎟  ;
  π   6  45 945 4725   93555  3⎝⎝
2⎠   
  x - 
─   
  
2   

   
   
  ⎞
 π⎟
 x → ─⎟
 2⎠


On Wednesday, June 4, 2014 8:01:06 PM UTC+4, su225 wrote:

 Hello. I'm new to sympy and took the tutorial today. But I noticed the 
 following error. I'm glad to contribute to this project.
 Where can I get resources to learn more about the sympy?

 The following line produces erroneous output.
 In [129]: tan(x).series(x,pi/2,10)

 I have attached the screenshot of the output
  
 tan(x) cannot be expanded about pi/2.
 The idea is determining the interval of convergence. If the point 
 specified for the expansion is out of this interval then throw an exception 
 telling that the power series expansion doesn't exist about that point.

 Thanks.




-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/7cb3e2cf-c7a1-41cd-b684-2e0605398302%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Behavior of Float objects with .format() method

2014-06-11 Thread Sergey Kirpichev



On Monday, June 9, 2014 5:35:34 AM UTC+4, Amit Saha wrote:

 I came across this: 

  from sympy import Float 
  Float(1.25343) 
 1.253430 

  '{0:.2f}'.format(Float(1.25343)) 
 Traceback (most recent call last): 
   File stdin, line 1, in module 
 ValueError: Unknown format code 'f' for object of type 'str' 

 Is this known/expected? Shouldn't it work? 


Why?  f stands for builtin float, not Float.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/9e8b5557-0576-4f15-857d-5ff2b0ac00ee%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Lazy and relaxed series

2014-06-03 Thread Sergey Kirpichev

On Tuesday, June 3, 2014 10:22:14 PM UTC+4, Camille Gillot wrote:

 I wanted to know if there is any interest in lazy power series computation 
 in Sympy.


Yes, there is.
 

 I would like to contribute by coding a kind of series field operations 
 module, implementing series with on-demand coefficient computation.


In some sense, we have lseries method for on-demand computation of series 
terms.  But the output is not well structured (it's usual sympy's 
expressions), not a some fancy object.
 

 Do some stream-like data structure already exist ?


No.

btw, you should take look on the recent series-related discussions, e.g. 
this:
https://groups.google.com/forum/?fromgroups=#!searchin/sympy/series$20stream/sympy/o214NoJ7tJY/iXT_VSbEOocJ

And we have one accepted GSoC project for series improvement.  You can help 
student by submitting patches or doing reviews if you know the subject 
well...

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/a11c3853-df1a-4e02-b5c6-64060722bbe5%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Unbundle mpmath

2014-04-26 Thread Sergey Kirpichev

Hello,

Perhaps, it's time to revive this old discussion, as sympy devs seems
to be more positive for allowing external dependencies:
https://github.com/sympy/sympy/issues/7339

Thus, it's a ping for mailing list about
https://github.com/sympy/sympy/pull/2192
(as suggested by Ondřej)

On Friday, June 28, 2013 5:42:07 PM UTC+4, Sergey Kirpichev wrote:

 There is a long term issue [1] to make mpmath an external dependency.  It 
 was suggested by
 Aaron (see discussion in PR [2]), that we should discuss this on the 
 mailing list.

 Perhaps, the main reason to prefer bundled mpmath copy is a
 mystical complex installation.  Not clear enough.  For example, I don't
 think that additional steps for package installation from mpmath manual 
 does matter.  For non-root users
 there shouldn't be any problems too, at least for Python 2.6+ (per-user 
 site-package directory).

 [1] http://code.google.com/p/sympy/issues/detail?id=2482
 [2] https://github.com/sympy/sympy/pull/2192



-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/6b54d310-2754-4a4c-acb9-15d77447b6e3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Are all sympy Floats rational?

2014-04-04 Thread Sergey Kirpichev

On Mon, Mar 31, 2014 at 02:17:56PM -0700, Richard Fateman wrote:
I think the answer from a practical perspective is that you can't exclude
floats.

But why we need them as a class, derived from Number and Basic?

An example of real issue:
In [1]: n1=Float(0.001)
In [2]: n2=Float(1.1)
In [3]: n3=Float(1.0)
In [4]: x1=n1*x
In [5]: x2=n2*x
In [6]: x3=n3*x
In [10]: f1=Add(Add(x1, x2, evaluate=False), -x3, evaluate=False)
In [11]: f1
Out[11]: -1.0*x + 0.001*x + 1.1*x
In [13]: f2=Add(x1, Add(x2, -x3, evaluate=False), evaluate=False)
In [14]: f1
Out[14]: -1.0*x + 0.001*x + 1.1*x
In [20]: f1-f2 # Hey, sympy thinks associativity holds
Out[20]: 0

But:
In [21]: f1e=Add(Add(x1, x2), -x3); f1e
Out[21]: 0.101*x
In [22]: f2e=Add(x1, Add(x2, -x3)); f2e
Out[22]: 0.101*x
In [23]: f1e-f2e
Out[23]: -1.11022302462516e-16*x

So are you taking the view, now, that 0.5 is rational?

Yes, for the provided example (x**0.5)

  End of story.  It seems, people want to introduce not floats but a fancy
  notation for rationals.

You can't predict the future.

I'm not attempting to do so.

   If this is only the reason why we keep Float
  object - we should drop Float's and fix sympify to map python's floats
  to Rational's instead (for example: sympify(0.2) - Rational(str(0.2))).

Sometimes floats can be very handy.  Finding approximations to roots
of a polynomial.  Compared to rational arithmetic, where different,
usually much slower methods, might be used.

But I don't suggest to forbid using float at all (python has builtin
float type, we also have mpf from mpmath; we can convert SymPy's numbers to 
these
types in some use cases).

  Is there a real need for floats in symbolic mathematics package?

Just because you might not need them now, doesn't really answer the
question.

Good point.  But if we don't need them now - it's a good idea to
introduce such an object only on demand...

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20140404144411.GA23457%40darkstar.order.hcn-strela.ru.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Are all sympy Floats rational?

2014-04-04 Thread Sergey Kirpichev

On Friday, April 4, 2014 6:44:11 PM UTC+4, Sergey Kirpichev wrote:

 In [13]: f2=Add(x1, Add(x2, -x3, evaluate=False), evaluate=False) 
 In [14]: f1 
 Out[14]: -1.0*x + 0.001*x + 1.1*x 


A little typo, should be:
In [13]: f2=Add(x1, Add(x2, -x3, evaluate=False), evaluate=False)
In [14]: f2
Out[14]: 0.001*x + -1.0*x + 1.1*x

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/82a37017-c8c5-4f13-9efe-9b7eba51a85c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Issue migration from Google Code to GitHub

2014-04-03 Thread Sergey Kirpichev

Hello,

An update for this task.  I will run script 6th April to add
references for github issues on google code.  See this:
https://github.com/sympy/sympy/issues/7235#issuecomment-38809765
and discussion below.

Unfortunately, a lot of email notifications will appear (for issue owner or 
for people in cc list).

If you disagree or have better ideas how to do this - please reply.

On Friday, March 28, 2014 7:10:10 PM UTC+4, Sergey Kirpichev wrote:

 On Fri, Mar 28, 2014 at 11:02:53AM -0400, Jason Moore wrote: 
 Yes, applause! Thanks for doing this. 
 Jason 

 This is not done yet.  See this: 
 https://github.com/sympy/sympy/issues/7235 


-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/f66d5abd-2984-42f0-a4ad-d04609d34b75%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Python 3.2 support

2014-03-31 Thread Sergey Kirpichev



On Monday, March 31, 2014 6:30:31 PM UTC+4, Aaron Meurer wrote:

 Ok, that's fine. I didn't realize some Linux distros were still on 3.2. 


Debian!

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/89cb7cb4-a832-4817-a9eb-ad6e36b9aea1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Are all sympy Floats rational?

2014-03-31 Thread Sergey Kirpichev

On Monday, March 31, 2014 7:38:48 AM UTC+4, Richard Fateman wrote:

 Does python allow you to just write very long integers, e.g.
 x: := 213456789123456789123456789 ?


Yes.  And almost every contemporary language allow you to do so.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/a03ffa02-c9a0-413c-9ccd-9f369b6446b3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Are all sympy Floats rational?

2014-03-31 Thread Sergey Kirpichev

On Sat, Mar 29, 2014 at 01:28:54PM -0700, Richard Fateman wrote:
Common Lisp uses floating-point arithmetic for floats, exact arithmetic
for the data type used for rationals.

Good news.  Python does same.

If you stick with integers and ratios of integers, Common
Lisp does mathematically valid  arithmetic.

Excelent!  Python too.

The problem is: float arithmetic is not a valid arithmetic by
design.  And the question is: should we use float numbers in SymPy
at all.

  But I admit, we shouldn't allow
  using of Float's (or builtin float type) in symbolic mathematics. 

I disagree, from experience.  People want to introduce floats.  For
example, some people
write squareroot(x)   as  x^0.5
Saves typing, and they believe it is the same thing.

But that's not necessary floats!  It's just a different
notation for rational 1/2 (as they believe).

In [1]: Rational('0.123')
Out[1]: 
123 

1000
In [2]: Float('0.123')
Out[2]: 0.123
In [3]: type(_)
Out[3]: sympy.core.numbers.Float
In [4]: sympify('0.123')
Out[4]: 0.123
In [5]: type(_)
Out[5]: sympy.core.numbers.Float

The little problem here is that float(0.3) != 3/10 internally, for
example (despite people think so).  See this:
https://docs.python.org/2/tutorial/floatingpoint.html

In [16]: Decimal(0.3)
Out[16]:
Decimal('0.299988897769753748434595763683319091796875')
In [17]: Rational(0.3)
Out[17]: 
 5404319552844595
 ─
 18014398509481984
In [18]: Rational(str(0.3))
Out[18]: 3/10

End of story.  It seems, people want to introduce not floats but a fancy
notation for rationals.  If this is only the reason why we keep Float
object - we should drop Float's and fix sympify to map python's floats
to Rational's instead (for example: sympify(0.2) - Rational(str(0.2))).

Is there a real need for floats in symbolic mathematics package?

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20140331162823.GB30706%40darkstar.order.hcn-strela.ru.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Are all sympy Floats rational?

2014-03-29 Thread Sergey Kirpichev

On Saturday, March 29, 2014 8:34:36 AM UTC+4, Richard Fateman wrote:

 Your problem is that the notions of +  and - in your computer programming 
 language
 are apparently inadequate.


In which one?  Python, CLisp?
 

 It is certainly possible to do this correctly by converting a,b,c
 into ratios of integers


Sure.  It's possible to convert a,b,c to rationals, but then it will be 
rational
arithmetic, not arithmetic for floats (IEEE 754).
 

 The fact that  +  gets different answers for adding numbers a,band for 
 adding a', b'   where
 a-a' and b=b'can't be a good thing in a system that is supposed to do 
 mathematics.

 
There are still reasons for floats (Christophe point you to major one).  
But I admit, we shouldn't allow
using of Float's (or builtin float type) in symbolic mathematics.  An 
example: https://github.com/sympy/sympy/pull/2801

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/2a4b3706-875b-428a-b6b6-dc0b42f59617%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: ProductSet implementation

2014-03-26 Thread Sergey Kirpichev


On Wednesday, March 26, 2014 5:18:41 PM UTC+4, Amit Saha wrote:

 IMO, a product set is a set (at least Wikipedia tells me so). So, 
 shouldn't it rather be the following? 

 {(1,2), (1,3), (2,2), (2,3)} 


1) It *is* a set:
In [8]: isinstance(p, Set)
Out[8]: True

2) Yes, maybe for FiniteSet - we should evaluate ProductSet.
 

 Again going by the same idea, I also feel, the len() function should 
 return the cardinatlity and not raise a TypeError: 


Maybe.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/781a6ecb-ebcf-4f2b-9221-4ef4e9b96a3e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] subset() : why not is_subset() ?

2014-03-25 Thread Sergey Kirpichev



On Tuesday, March 25, 2014 2:14:10 PM UTC+4, Amit Saha wrote:

 Before I work on the patch, what is the strategy to make backward 
 incompatible changes? Since, existing users may be surprised if we 
 suddenly remove the method and replace it by another one and even 
 break their existing code. 


I suggest you to take look on the code around.  We have
a lot of bugs with  DeprecationRemoval label.

In this particular example, you should rename method and then
add @deprecated decorator.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/d446c0eb-284c-4ac9-b570-caf7aee42b86%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Are all sympy Floats rational?

2014-03-24 Thread Sergey Kirpichev


On Monday, March 24, 2014 4:58:02 AM UTC+4, Richard Fateman wrote:

 So is every double-precision IEEE float  (except inf and NaN) convertible 
 to the
 exact rational number which it represents in the form  integer/integer.


It was noted in this thread several times: we are not interested
in this truism.  The problem is not with the data structure, but with 
operations.
Field properties doesn't hold for floats, it was shown for you several 
times in this
thread (CLisp example included).
 

 2. we  can deduce from the fact that a,b are rational numbers that a+b is 
 a rational number.


But now, suppose a, b and c - rational numbers.  Then we can deduce:
((a + b) + c) - (a + (b + c)) is zero.   But this conclusion will be wrong 
if
we count floats as rationals. 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/7eca18e2-831e-4e5d-bdec-06276ff8926c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Are all sympy Floats rational?

2014-03-20 Thread Sergey Kirpichev


On Thursday, March 20, 2014 4:19:24 PM UTC+4, Kalevi Suominen wrote:

 This leads to problems if used as a basis for decisions.


Sure.   Julien, can you comment this?  This was introduced by commit 
9c359bc (and this is not tested!).

In my view, it's wrong.  Floats are inexact numbers, we shouldn't mix them 
with rational arithmetic.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/4710c28b-4aea-42d3-bc4a-33d6366b37fd%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: GSOC 2014: Solvers

2014-03-18 Thread Sergey Kirpichev



On Tuesday, March 18, 2014 8:20:31 PM UTC+4, Harsh Gupta wrote:

 We will have to be very careful while rewriting the solvers. I just came 
 across another of Professor Fateman's paper which describes why the 
 solution of z**w == y for z isn't simply y**(1/w).

 http://www.cs.berkeley.edu/~fateman/papers/y=z2w.pdf

 Sadly we doing it wrong.

 We are also wrongly simplifying y - (y**(1/(1 + I)))**(1 + I) as 0.

 In [58]: f

 Out[58]: y - (y**(1/(1 + I)))**(1 + I)


 In [59]: simplify(f)

 Out[59]: 0


Indeed (e.g. y=-1 + 4000*I).  Looks a bug somewhere in powsimp().  It 
doesn't look as a new for me, so maybe it was already reported.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/0be1c8c7-7847-406c-85ef-bfc3900322da%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Discussion regarding the series expansion project

2014-03-14 Thread Sergey Kirpichev

Hello,

On Fri, Mar 7, 2014 at 3:29 AM, Avichal Dayal avichal.da...@gmail.comwrote:
 It would be great if you could look at my proposal (just an initial draft) 
 and give suggestions:-
 https://github.com/sympy/sympy/wiki/GSoC-2014-Application-Avichal-Dayal-Series-Expansion

 Formal Power Series:

What about Lambert series?

 Currently series() returns something like 1 + x + x**2/2 + O(x**3) while
 Mathematica and others are also capable of returning something like
 summation(x**n/factorial(n), (n, 0, oo)).

I'm little surpised.  Can you provide an example of this from
the Mathematica?

 Basic class structure: --FormalSeries--

Very unclear.  That's all to say.

 Representation of infinite series: ... I'm planning to represent it
 using ordered term of (coeff, exponent) knows as “lazy evaluation” in
 the literature.

Probably, you should think about multivariable series too.

 We will need a sequence object to represent sequences. This is already
 implemented here:
 https://github.com/goodok/sympy/tree/sequences/sympy/sequences. However
 the code is old and does not play nice with current SymPy structure. I
 plan to re-implement it based on a similar model.

I don't think it's a good idea.  We should have instead a basic
container type (cf. tuple or list) for infinite sequence (take look on
Stream class in the fn python package).  Then, we can
wrap it like Tuple or Dict.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/20140314102414.GA10361%40darkstar.order.hcn-strela.ru.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Discussion regarding the series expansion project

2014-03-14 Thread Sergey Kirpichev



On Friday, March 14, 2014 6:24:15 PM UTC+4, pr...@goodok.ru wrote:

 Do you meant Laurent series? 


Yep, sorry

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/25817a3e-8098-4187-9b7b-f63b3ffc3f59%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Can sympy do tensor based Euler-Lagrange?

2014-03-14 Thread Sergey Kirpichev

On Thursday, March 13, 2014 7:29:40 PM UTC+4, brombo wrote:

   If you are talking about Lagrangian field theory the question is how do 
 you take a derivative with respect to the field and the gradient of the 
 field and not with respect to the position vector the field is a function 
 of.


The docstring of euler_equation function has a trivial example of scalar 
field.  There should not be any problems for more complex field (e.g. for 
electrodynamics).

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/ffb3b43b-fd2c-4e7b-9c47-372d43e441c6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Simplify expression

2014-03-14 Thread Sergey Kirpichev



On Friday, March 14, 2014 9:30:24 PM UTC+4, ecir...@gmail.com wrote:

 3. If I knew that one term is in the form of (a - b), would the help?


Yes, for example:
In [2]: e = a**4 - 4*a**3*b + 6*a**2*b**2 - 4*a*b**3 - a + b**4

In [3]: e
Out[3]: 
 4  32  234
a  - 4⋅a ⋅b + 6⋅a ⋅b  - 4⋅a⋅b  - a + b 

In [4]: e2 = e.subs(a, x + b)

In [5]: e2.expand()
Out[5]: 
  4
-b + x  - x

In [6]: e2.expand().subs(x, a - b)
Out[6]: 
4
-a + (a - b) 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/a50281ef-f3f4-4e09-9103-bb308f5ef481%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: The place where the func and args attribute is set for sympy objects

2014-03-13 Thread Sergey Kirpichev


On Thursday, March 13, 2014 9:32:47 AM UTC+4, Tschijnmo Tschau wrote:

 It looks like AssocOp only have got one level of super class in Sympy, the 
 Basic class, whose constructor sets the _args attribute but not for the 
 func and args attribute.


The Basic class has *properties* func and args (@property decorator).

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/5c979e13-9899-4008-a7d2-f8ca79d64270%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sympy] Re: Can sympy do tensor based Euler-Lagrange?

2014-03-13 Thread Sergey Kirpichev

On Monday, March 10, 2014 9:24:27 PM UTC+4, Joseph Smidt wrote:

Does anyone know if sympy is equipped to handle tensor based 
 Eular-Lagrange equations equations such as these?


Yes, with euler_equations from calculus module.  But it doesn't support 
tensor notation directly.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/ac94ef0a-2527-41be-91dd-341d8bc7e21e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Re: Issue migration from Google Code to GitHub

2014-03-07 Thread Sergey Kirpichev

Everything is on the github.

https://github.com/sympy/sympy/issues/7235

On Thursday, March 6, 2014 12:32:21 AM UTC+4, Aaron Meurer wrote:

 This is done. It took some time because I was only able to update 200 
 issues at a time. I accidentally sent out an email on one of them, so 
 sorry about that. 

 Now only the owners of the Google Code repo can edit issues, which is 
 just me and Ondrej. If you want to take a shot at forward links on all 
 the issues, let me know and I'll give you access. 

 Aaron Meurer 

 On Wed, Mar 5, 2014 at 1:29 PM, Sergey B Kirpichev 
 skirp...@gmail.comjavascript: 
 wrote: 
  On Wed, Mar 05, 2014 at 12:43:17PM -0600, Aaron S. Meurer wrote: 
  Yes, let's get it done. Should I freeze the google code issues? 
  
  Yes.  Then I'll run export script and then 
  1) upload zip file with exported data and provide the link for you 
  2) ask support to import it to the test repo 
 https://github.com/skirpichev-test/test3/issues 
 (this should be as fast as the export stage) 
  3) ask for final import (or you can do this instead) 
  
  -- 
  You received this message because you are subscribed to the Google 
 Groups sympy group. 
  To unsubscribe from this group and stop receiving emails from it, send 
 an email to sympy+un...@googlegroups.com javascript:. 
  To post to this group, send email to sy...@googlegroups.comjavascript:. 

  Visit this group at http://groups.google.com/group/sympy. 
  To view this discussion on the web visit 
 https://groups.google.com/d/msgid/sympy/20140305192931.GA26996%40darkstar.order.hcn-strela.ru.
  

  For more options, visit https://groups.google.com/groups/opt_out. 


-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/a49dc856-6eaf-4776-b758-e09c415864e0%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sympy] Addition of 'BigInteger' to Numbers

2014-03-05 Thread Sergey Kirpichev


On Wednesday, March 5, 2014 5:08:47 AM UTC+4, Ondřej Čertík wrote:

 Can you post your code where you need large integers in Python that 
 runs out of memory?


A good example is this one:
http://code.google.com/p/sympy/issues/detail?id=3736
:)

But I think it's a very different (and very difficult!) problem (mostly
related with rules of evaluation for sympy's objects. 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/26bc3365-0553-4653-a432-6dc3a7c6f887%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sympy] Re: Issue migration from Google Code to GitHub

2014-03-04 Thread Sergey Kirpichev

Any feedback?

I don't have a good idea what to do next. Every mentioned
problem in the test repo was solved one way or another (see
labels Migration-Bugs and Migration-Bugs-Done).

We can do a new test (final?) import as before, in one repo,
or we can also split issues, labeled Live (~80) to a separate repo.

BTW, there is no references in the imported issues to
these people from the team:
https://github.com/tjl
https://github.com/moorepants
https://github.com/lidavidm
https://github.com/raoulb
https://github.com/srjoglekar246
Perhaps, their contacts differs for googlecode and github.

On Monday, February 17, 2014 10:32:35 PM UTC+4, Sergey Kirpichev wrote:

Hello,

On Sun, Feb 16, 2014 at 1:05 PM, Sergey Kirpichev
skirpic...@gmail.comwrote:

[...]
Other tasks for volunteers (in random order):
1) Find problems in the test import (to be in:
https://github.com/skirpichev-test/issues-test2
Please report issues in the same repo and add label:
Migration-Bugs.
[...]

Everything is in this repo and ready to test.

Issue numbers are shifted with delta = 4.

Some problems:
1) assignee option doesn't work sometimes (Only if
I add mentioned collaborators first. Currently, this was
done for certik and asmeurer after importing ~200 issues)
2) formatting errors for old issues with textile syntax:
https://github.com/skirpichev-test/issues-test2/issues/26
Fixed and should be ok for issues with number = 300.
3) #123 for autolinking issues does work only for markdown.
Old issues are formatted with textile syntax.
4) Empty title:
https://github.com/skirpichev-test/issues-test2/issues/182
5) Bad unicode:
https://github.com/skirpichev-test/issues-test2/issues/4099
6) Too long issue/comment body:
https://github.com/skirpichev-test/issues-test2/issues/293

--
You received this message because you are subscribed to the Google Groups
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/b134c541-2f92-40cd-9f38-f43f740fbc75%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[sympy] Re: complex numbers

2014-03-04 Thread Sergey Kirpichev

On Tuesday, March 4, 2014 9:38:22 AM UTC+4, Chris Smith wrote:

 Something that seems a little counter-intuitive to me (having now worked 
 with the assumptions a bit) is the decision to let 2, I and 2 + I all be 
 considered complex


I'm little surprised, what could be counter-intuitive here from the 
mathematical point of view?
 

 It seems like it would be more useful to consider them be onlny real, 
 imaginary and complex, resepectively. 


Some shortcuts for internal use may be convenient, however.  E.g. 
Q.nonzero_imaginary or something.

The way it is right now, the only way to say that something has a non-zero 
 real and imaginary part is to write it as a + I*b, e.g. ask(Q.real(I**(a + 
 I*b)), Q.real(a)  Q.real(b))


But this does not mean a+I*b has a non-zero re/im parts!

Also, consider some arithmetic, e.g. complex + complex != complex with your 
definition.  By killing field properties - you complicate other things, I 
guess...

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/98a6227e-cbd9-44e7-899f-3449e7f1b042%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[sympy] Re: asymptotic expansion of sin and cos at infinity

2014-03-01 Thread Sergey Kirpichev

On Friday, February 28, 2014 9:02:25 PM UTC+4, Avichal Dayal wrote:

 Currently the series method does not allow to return only the Order term.


Why do you think so?

In [2]: O(1).series(x,n=100)
Out[2]: O(1)

 Since sin and cos do not converge at infinity and are bounded functions,
 in my opinion asymptotic expansion of sin and cos at [oo] should be simply
 Order(1), right?

Typo?  Perhaps, O(1, (x, oo))

Why you think, that that's
1) an expansion
2) the right one.

O(1) - just a set of function with some specific behaviour around x=0.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/bf58345d-3c0a-4198-936f-0965408a767c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[sympy] Re: asymptotic expansion of sin and cos at infinity

2014-03-01 Thread Sergey Kirpichev

On Saturday, March 1, 2014 5:35:01 PM UTC+4, Avichal Dayal wrote:

 Perhaps, something to indicate an error. 

 But there are instances where series(sin(x), x, oo) is used by other 
 methods
 For e.g.:- gruntz((sin(x) + cos(x)/x**2, x, oo) tries to find that series
 If we raise an error, then those limits won't work (which should) 


Limit may work, if we implement algorithms to solve this kind
of problems.   gruntz() wont work, but it's expected.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/f3470041-9480-48a3-aa74-dd289b6c3434%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[sympy] Re: Discussion regarding the series expansion project

2014-02-28 Thread Sergey Kirpichev

On Friday, February 28, 2014 12:01:10 AM UTC+4, Avichal Dayal wrote:

 Yes, that's very *big* amount of work.

 You mean enough for the summer
 or too much for the summer?


It depends on what is exactly you are planning to do.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/7fcf5052-2110-44a9-a4e8-76c16d29ad4c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sympy] Re: New PEP for infix operator

2014-02-27 Thread Sergey Kirpichev


On Wednesday, February 26, 2014 11:49:47 PM UTC+4, Aaron Meurer wrote:

  Because it turn python to be a DSL for numeric computations. 

 How is that a bad thing? Python is already used heavily for numeric 
 computations.


It's good for this specific domain, but bad for the Python in general,
as it is not a DSL.

btw, here is a discussion of the possibility to have generic infix ops (the 
whole thread):
https://mail.python.org/pipermail/python-dev/2008-July/081509.html
and some arguments against matrix mul as well, e.g.:
https://mail.python.org/pipermail/python-dev/2008-July/081537.html

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/8e7613c2-2d43-4f5c-b2df-77e5b8de7cbd%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[sympy] Re: Discussion regarding the series expansion project

2014-02-27 Thread Sergey Kirpichev


On Thursday, February 27, 2014 9:14:30 AM UTC+4, Avichal Dayal wrote:

 I am planning to work on formal power series, asymptotic expansion and 
 order
 term arithmetic. Will that be the right amount of work for the summer?


Yes, that's very *big* amount of work.  btw, I think now (see pr #2630) we 
have
some functions in polys module for first task.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/1ef45d6c-e230-4719-a490-3d8b8b83bbaf%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sympy] Discussion regarding the series expansion project

2014-02-26 Thread Sergey Kirpichev



On Wednesday, February 26, 2014 6:03:22 AM UTC+4, Aaron Meurer wrote:

 There was some paper about computing limits with oscillating 
 functions. Maybe you can find the reference if you search the mailing 
 list archives (or probably Raoul will remember it). 


Probably, it's a good idea to search first in google code issues, tagged 
with Series.

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/db1e82c9-2288-4ce0-9a2b-d5052da9493b%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[sympy] Re: New PEP for infix operator

2014-02-25 Thread Sergey Kirpichev

On Tuesday, February 25, 2014 8:31:02 PM UTC+4, Aaron Meurer wrote:

 Some people on this list might be interested in this. Some folks are 
 giving another shot at writing a PEP to add a new operator to Python 
 for matrix multiplication. The discussion is at 
 https://github.com/numpy/numpy/pull/4351. 


Really bad idea.  It's better to have an ability to define arbitrary
infix operator (like Haskell, for example).

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/21d80555-342e-4ae1-b7f0-9db077bc37f6%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sympy] Re: GSOC 2014 idea: Adding parsing functionality to Sympy Live

2014-02-24 Thread Sergey Kirpichev



On Monday, February 24, 2014 7:38:22 AM UTC+4, Aditya Shah wrote:

 Then I have another idea in mind. A language specifier configuration file 
 can be given as input by the developer and sympy takes care of the rest. 
 It's quite like yacc. This would essentially reduce the workload of the 
 developer substantially regarding the development of a new parser.


What if there is no language specification, like for Mathematica? 

-- 
You received this message because you are subscribed to the Google Groups 
sympy group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sympy+unsubscr...@googlegroups.com.
To post to this group, send email to sympy@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/73fbc371-03a7-4631-b3e0-d64502fc6eed%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Re: [sympy] Re: Limit - Bug

2014-02-18 Thread Sergey Kirpichev

On Tuesday, February 18, 2014 12:51:50 PM UTC+4, Christophe Bal wrote:

 Hello,
 what is the origin of this name gruntz ?


From the name of algorithm author.


 2014-02-18 1:57 GMT+01:00 Aaron Meurer asme...@gmail.com javascript::

 Ah, so it seems that these incorrect limit heuristics have still not
 been completely abolished from the code base. I would recommend just
 using gruntz() for now. limit() is not that much smarter, but it does
 have a bad tendency to give wrong answers when gruntz() works just
 fine.

 Aaron Meurer

 On Mon, Feb 17, 2014 at 1:29 PM, Christophe Bal 
 proj...@gmail.comjavascript: 
 wrote:
  Thanks for this.
 
 
  2014-02-17 19:21 GMT+01:00 Avichal Dayal avicha...@gmail.comjavascript:
 :
 
  However gruntz gives the correct result:-
  gruntz((x*exp(x)) / (exp(x)-1), x, -oo) gives 0
 
  The part of code that is going wrong is already labeled as XXX: todo
  More specifically the following:-
  if abs(z0) is S.Infinity:
  # XXX todo: this should probably be stated in the
  # negative -- i.e. to exclude expressions that should
  # not be handled this way but I'm not sure what that
  # condition is; when ok is True it means that the leading
  # term approach is going to succeed (hopefully)
 
  It should exclude some expressions and I guess this is one of them.
  I'll try to understand and fix it.
 
 
  --
  You received this message because you are subscribed to the Google 
 Groups
  sympy group.
  To unsubscribe from this group and stop receiving emails from it, send 
 an
  email to sympy+un...@googlegroups.com javascript:.
  To post to this group, send email to sy...@googlegroups.comjavascript:
 .
  Visit this group at http://groups.google.com/group/sympy.
  For more options, visit https://groups.google.com/groups/opt_out.
 
 
  --
  You received this message because you are subscribed to the Google 
 Groups
  sympy group.
  To unsubscribe from this group and stop receiving emails from it, send 
 an
  email to sympy+un...@googlegroups.com javascript:.
  To post to this group, send email to sy...@googlegroups.comjavascript:
 .
  Visit this group at http://groups.google.com/group/sympy.
  For more options, visit https://groups.google.com/groups/opt_out.

 --
 You received this message because you are subscribed to the Google Groups 
 sympy group.
 To unsubscribe from this group and stop receiving emails from it, send an 
 email to sympy+un...@googlegroups.com javascript:.
 To post to this group, send email to sy...@googlegroups.com javascript:
 .
 Visit this group at http://groups.google.com/group/sympy.
 For more options, visit https://groups.google.com/groups/opt_out.




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

[sympy] Re: Feedback on step-by-step thought

2014-02-17 Thread Sergey Kirpichev

It doesn't seems to be a good idea to modify almost every function in the 
SymPy.

It's Python, it has a great introspection and debugging capabilities!  
Please take look to something like trace module from stdlib.

btw, if you are going to modify the SymPy sources - please fork and do your 
modification on the branch.

https://github.com/sympy/sympy/wiki/Development-workflow

On Sunday, February 16, 2014 8:41:54 PM UTC+4, RAJAT AGGARWAL wrote:

 Hi Matheww
 You can look at the following link. I had just posted some methods of the 
 classes so that you can have the idea of some modifications.
 https://github.com/rajat974/StepDerivation

 Rajat Aggarwal

 On Sunday, February 16, 2014 3:04:19 AM UTC+5:30, RAJAT AGGARWAL wrote:

 Based on suggestions given by Matheww for step-by-step implementation, I 
 tried implementing the step derivation first. Attached file is the output 
 of some basic derivations which show steps. It needs to be arranged yet. 
 Also, the printing may depend upon the certain requirements. Waiting for 
 the feedback.

 Thanks

 Rajat



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

Re: [sympy] Re: Issue migration from Google Code to GitHub

2014-02-17 Thread Sergey Kirpichev

Hello,

On Sun, Feb 16, 2014 at 1:05 PM, Sergey Kirpichev skirpic...@gmail.comwrote:
 [...]
 Other tasks for volunteers (in random order):
 1) Find problems in the test import (to be in:
https://github.com/skirpichev-test/issues-test2
Please report issues in the same repo and add label:
Migration-Bugs.
 [...]

Everything is in this repo and ready to test.

Issue numbers are shifted with delta = 4.

Some problems:
1) assignee option doesn't work sometimes (Only if
   I add mentioned collaborators first.  Currently, this was
   done for certik and asmeurer after importing ~200 issues)
2) formatting errors for old issues with textile syntax:
   https://github.com/skirpichev-test/issues-test2/issues/26
   Fixed and should be ok for issues with number = 300.
3) #123 for autolinking issues does work only for markdown.
   Old issues are formatted with textile syntax.
4) Empty title:
   https://github.com/skirpichev-test/issues-test2/issues/182
5) Bad unicode:
   https://github.com/skirpichev-test/issues-test2/issues/4099
6) Too long issue/comment body:
   https://github.com/skirpichev-test/issues-test2/issues/293

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

1 2 >

1 - 100 of 138 matches

Mail list logo