make symbolic series subclass of Expression

[rwst]

Making Expression.series create SymbolicSeries (a subclass of Expression) and moving the series code into a separate file series.pyx

• reflects the fact that expressions and series don't commute
• allows upcoming fixes that don't clutter Expression methods with if ex.is_a_series():
• makes for better documentation via having a Symbolic Series ref man page

In refactoring language this is "replace conditional with polymorphism".

16203, #17400, and #17402 depend on this.

Depends on #19948

Component: symbolics

Author: Ralf Stephan

Branch/Commit: 1832f4a

Reviewer: Vincent Delecroix

Issue created by migration from https://trac.sagemath.org/ticket/17659

[rwst]

comment:38

That happens in 6 out of 10 attempts here. Bummer.

[rwst]

comment:39

For the record, only when doctesting and in different processes does it fail often enough to be noticed (regardless of if testing in parallel or not). When not leaving the Sage process the results seem always the same. One possbility would be therefore an uninitialized variable in Pynac affecting the result depending on what's in the memory at Sage start.

[vbraun]

comment:40

Might also depend on Python memory locations (are the underlying ginac objects cached or not?). If its an uninitialized pynac variable then valgrind should find it.

[rwst]

comment:41

I would have found it if it were in the upcoming pynac-0.5.2 because I used valgrind extensively there. So it might just disappear with #19312.

[rwst]

comment:42

Okay, for some reason printing order here is according to hash value, and hash values of symbols are not preserved over different processes. Putting sage: hash(x) in an arbitrary doctest and testing multiple times shows this clearly, also in Sage-6.6. It's just that such series manipulations produced an error until recently, so the differences over time couldn't be observed.

[rwst]

comment:43

See https://github.com/pynac/pynac/issues/22

This is now fixed in pynac and depends on the upgrade to pynac-0.6.1.

[rwst]

Dependencies: #19448

[rwst]

Changed dependencies from #19448 to #19948

[vbraun]

comment:46

sage -t --long src/sage/symbolic/series.pyx
**********************************************************************
File "src/sage/symbolic/series.pyx", line 71, in sage.symbolic.series
Failed example:
    x*ex1
Expected:
    (1*x + (-1/6)*x^3 + Order(x^4))*x
Got:
    x*(1*x + (-1/6)*x^3 + Order(x^4))
**********************************************************************
1 item had failures:
   1 of  23 in sage.symbolic.series
    [42 tests, 1 failure, 0.04 s]

[rwst]

comment:47

Beware the dependency!

[vbraun]

comment:49

Patchbot also shows failure

[rwst]

Changed branch from u/rws/17659-1 to u/rws/17659-2

[rwst]

comment:51

Yes, a remainder from the time before #19948.

---

New commits:

c3be6a1	Merge branch 'u/rws/17659-1' of trac.sagemath.org:sage into tmp04
1832f4a	17659: fix doctest

[rwst]

Changed commit from d5cc424 to 1832f4a

[tscrim]

comment:53

Ralf, did you intend for this to wait to 7.2 or did you mean 7.1?

[vbraun]

Changed branch from u/rws/17659-2 to 1832f4a