Do you like sets? I like sets! Full sets Module! by latot · Pull Request #585 · gnu-octave/symbolic

latot · 2016-09-30T01:19:02Z

Hi @cbm755 this time i bring the full sets module (i skip some redundant functions), because we add intervals function with some things but is very incomplete, when i need works with it, it feels... without tools to handle it.

So i think with this we will can works properly with sets.

Obvs, the function names don't are absolutes, if you like we can discuss names for it, in the most of functions i use the sympy name, in others i change it to conserve the octave/matlab prefix.

And this time i take time to check every format!

This pr is required for improve solveset correctly.

Here are two special functions it not belongs directly from sympy sets:

domain: return a domain from sympy 'Complexes', 'Reals', etc (the only function added to user env)
point: express a point or a set of it (don't is from sets but its a feature of sympy).

Obvs you can like split this heavy pr, but i think before it select ready things and that type of things.
I think here no are unnecessary functions, things with which we can work.

The original issue: #575
have the original names of sympy functions and some data of this work.

i don't know if all this functions are available in the travis sympy, i'll check it.

As a note, sadly you can't call sets from sym, why?, in a example, to theoretically you need call like this:

sym('UniversalSet()')

but sadly for the S function don't like "" in the expression with sets:

>>> S("UniversalSet()")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib64/python2.7/site-packages/sympy/core/sympify.py", line 322, in sympify
    expr = parse_expr(a, local_dict=locals, transformations=transformations, evaluate=evaluate)
  File "/usr/lib64/python2.7/site-packages/sympy/parsing/sympy_parser.py", line 894, in parse_expr
    return eval_expr(code, local_dict, global_dict)
  File "/usr/lib64/python2.7/site-packages/sympy/parsing/sympy_parser.py", line 807, in eval_expr
    code, global_dict, local_dict)  # take local objects in preference
  File "<string>", line 1, in <module>
  File "/usr/lib64/python2.7/site-packages/sympy/core/symbol.py", line 157, in __call__
    return Function(self.name)(*args)
  File "/usr/lib64/python2.7/site-packages/sympy/core/function.py", line 762, in __new__
    obj = super(AppliedUndef, cls).__new__(cls, *args, **options)
  File "/usr/lib64/python2.7/site-packages/sympy/core/cache.py", line 93, in wrapper
    retval = cfunc(*args, **kwargs)
  File "/usr/lib64/python2.7/site-packages/sympy/core/compatibility.py", line 807, in wrapper
    result = user_function(*args, **kwds)
  File "/usr/lib64/python2.7/site-packages/sympy/core/function.py", line 431, in __new__
    pr = max(cls._should_evalf(a) for a in result.args)
ValueError: max() arg is an empty sequence
>>> S(UniversalSet())
UniversalSet()

but if you remove the quotes you will can't make symbols. other point its it will be to confuse to the user, call a lot of possibles things from sym, seems a bad idea, i think its more clear a new function to handle it.

I decide use the S.tuple instead of Point as point for two reasons, first the point function have this issue: sympy/sympy#11683, and in the sets doc the tuple is used to represent point in sets, so it should be safe for now (i write in the doc what happen when you plus two points).

Thx. Cya.

cbm755 · 2016-09-30T16:47:47Z

Wow! This is pretty cool.

Pytave might make at least some of this possible without us writing code (by exposing things like .openleft automatically).

latot · 2017-01-26T03:46:30Z

Hi, all updated, but i was trying to check this in travis but isn't working D:

latot · 2017-01-26T03:49:11Z

(travis don't is checking the pr)

cbm755 · 2017-01-26T06:25:30Z

looks like it has run now (?)

latot · 2017-01-26T18:00:15Z

Yay!, all updated and working!

latot · 2017-04-06T22:13:58Z

Hi, is possible start merging this?

Thx.

cbm755

Ok, I started reading through this. A detailed review for correctness will take quite some time!

cbm755 · 2017-04-07T12:10:53Z

inst/@sym/ainterval.m

+%% @documentencoding UTF-8
+%% @defmethod @@sym ainterval (@var{x})
+%% Return the union of intervals of x when, @var{x} is in rectangular form
+%% or the union of intervals of r when self is in polar form.


First line need to fit on one line (its used for lookfor).

cbm755 · 2017-04-07T12:12:26Z

inst/@sym/boundary.m

+%!test
+%! a = interval (sym (0), 1);
+%! b = interval (sym (0), 1, true, false);
+%! assert( isequal( boundary (a), boundary (b)))


I know I've been very sloppy in the past, but for new code, let's try to use core Octave style (except for sym()).

cbm755 · 2017-04-07T12:14:31Z

inst/@sym/complexregion.m

+
+
+%% This function is tested in @sym/ainterval, @sym/binterval
+%% @sym/ispolar, @sym/psets, @sym/sets and @sym/normalizethetaset


add an %! assert true or something here too so automated checks don't flag this as untested.

cbm755 · 2017-04-07T12:15:35Z

inst/@sym/complexregion.m

+%% @documentencoding UTF-8
+%% @defmethod @@sym complexregion (@var{x}, @var{y})
+%% Represents the Set of all Complex Numbers.
+%% It can represent a region of Complex Plane in 


blank line between the "lookfor" and the rest of the body. But this looks wrong? Should it represent all complex numbers if no arguments are passed or what does the lookfor summary line mean?

cbm755 · 2017-04-07T12:16:36Z

inst/@sym/complexregion.m

+
+%% -*- texinfo -*-
+%% @documentencoding UTF-8
+%% @defmethod @@sym complexregion (@var{x}, @var{y})


y should be called something else like @var{polar}. And please list both ways of calling it:

@defmethod @@sym complexregion (@var{x}) @defmethodx @@sym complexregion (@var{x}, @var{polar})

cbm755 · 2017-04-07T12:21:07Z

inst/@sym/infimum.m

+%% -*- texinfo -*-
+%% @documentencoding UTF-8
+%% @defmethod @@sym infimum (@var{x})
+%% The infimum of @var{x}.


The infimum of the set @var{x}.

cbm755 · 2017-04-07T12:22:06Z

inst/@sym/iscomplement.m

+%% -*- texinfo -*-
+%% @documentencoding UTF-8
+%% @defmethod @@sym iscomplement (@var{x})
+%% Retrurn True if @var{x} is complement.


spelling mistake and this lookfor line doesn't tell me any more than the name iscomplement already does. At least put the word "set" in there somewhere.

cbm755 · 2017-04-07T12:22:33Z

inst/@sym/iscomplement.m

+%%   @result{} ans = (sym) True
+%% @end group
+%% @end example
+%%


Some/most of these methods need @seealso

cbm755 · 2017-04-07T12:23:05Z

inst/@sym/isdisjoint.m

+%% -*- texinfo -*-
+%% @documentencoding UTF-8
+%% @defmethod @@sym isdisjoint (@var{x}, @var{y})
+%% Returns True if @var{x} and @var{y} are disjoint.


make sure the word "set" appears in most of these "lookfor" summary lines,

cbm755 · 2017-04-07T12:24:18Z

inst/@sym/isleftunbounded.m

+%% -*- texinfo -*-
+%% @documentencoding UTF-8
+%% @defmethod @@sym isleftunbounded (@var{x})
+%% Return True if the left endpoint is negative infinity.


"return True" is probably not needed (I forgot exactly, you could check the core Octave guidelines on this sort of thing). Use the word "Interval" somewhere.

cbm755 · 2017-04-07T12:31:22Z

My biggest concern about all this is whether we should just wait until (and if?) sym objects support x.is_leftopen calling directly without us writing any code.

That would keeps things as close to sympy as possible, keeping things slim.

OTOH, this is here now and looks like it will be good. I am undecided what to do. Which is probably why I've let this sit so long...

latot force-pushed the more_sets branch 3 times, most recently from ad926a8 to 7e4f112 Compare September 30, 2016 02:01

latot changed the title ~~[WIP] You like sets? I like sets! Full sets Module!~~ You like sets? I like sets! Full sets Module! Sep 30, 2016

latot changed the title ~~You like sets? I like sets! Full sets Module!~~ Do you like sets? I like sets! Full sets Module! Sep 30, 2016

latot force-pushed the more_sets branch from 7e4f112 to db18726 Compare September 30, 2016 20:28

latot force-pushed the more_sets branch from 36849e2 to 50ba50a Compare January 26, 2017 00:03

latot added 22 commits January 26, 2017 00:41

improve inst/@sym/ainterval.m

bc8cb1a

improve inst/@sym/binterval.m

8adba2e

improve inst/@sym/boundary.m

1ca789d

improve inst/@sym/complement.m

0432c94

improve inst/@sym/complexregion.m

f7aaa18

improve inst/@sym/contains.m

ea687f7

improve inst/@sym/imageset.m

48f1392

improve inst/@sym/infimum.m

c4f5d14

improve inst/@sym/iscomplement.m

bd7aa6f

improve inst/@sym/isdisjoint.m

21fba6e

improve inst/@sym/isemptyset.m

6afe6ec

improve inst/@sym/isintersection.m

f4ea2ba

improve inst/@sym/isleftopen.m

f9275ef

improve inst/@sym/isleftunbounded.m

0bde17b

improve inst/@sym/ispolar.m

8211f68

improve inst/@sym/ispropersubset.m

8d2e67a

improve inst/@sym/ispropersuperset.m

504e490

improve inst/@sym/isrightopen.m

e88914c

improve inst/@sym/isrightunbounded.m

4b79f56

improve inst/@sym/issubset.m

0b33991

improve inst/@sym/issuperset.m

b3dc9ca

improve inst/@sym/isuniversalset.m

7c8e22e

latot added 10 commits January 26, 2017 00:41

improve inst/@sym/measure.m

b628130

improve inst/@sym/normalizethetaset.m

a5678f5

improve inst/@sym/point.m

187a14c

improve inst/@sym/powerset.m

31d984c

improve inst/@sym/productset.m

4b4f6d1

improve inst/@sym/psets.m

02e356a

improve inst/@sym/rangeset.m

a40c5cd

improve inst/@sym/reversed.m

3a0e1ef

improve inst/@sym/sets.m

4a7a4c4

improve inst/@sym/supremum.m

ea86f00

latot force-pushed the more_sets branch from 50ba50a to 118a24b Compare January 26, 2017 03:42

improve inst/domain.m

38a19ad

latot force-pushed the more_sets branch from 118a24b to 38a19ad Compare January 26, 2017 03:45

latot force-pushed the more_sets branch 3 times, most recently from 071448d to 9817c11 Compare January 26, 2017 17:20

fix tests

14d4008

latot force-pushed the more_sets branch from 9817c11 to 14d4008 Compare January 26, 2017 17:23

cbm755 requested changes Apr 7, 2017

View reviewed changes



		%% This function is tested in @sym/ainterval, @sym/binterval
		%% @sym/ispolar, @sym/psets, @sym/sets and @sym/normalizethetaset

Conversation

latot commented Sep 30, 2016 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cbm755 commented Sep 30, 2016

Uh oh!

latot commented Jan 26, 2017

Uh oh!

latot commented Jan 26, 2017

Uh oh!

cbm755 commented Jan 26, 2017

Uh oh!

latot commented Jan 26, 2017

Uh oh!

latot commented Apr 6, 2017

Uh oh!

cbm755 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cbm755 commented Apr 7, 2017

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

latot commented Sep 30, 2016 •

edited

Loading