delta/python-packages/sqlalchemy.git, branch rel_2_0_old

- black -l79 update

2019-07-06T20:15:39+00:00

Change-Id: Ie6ebff96f3b21fd35c9ceb22575b9b172373da7b

negate True/False separately from other elements

2019-07-06T20:15:39+00:00

Fixed issue where double negation of a boolean column wouldn't reset
the "NOT" operator.

Fixes: #4618
Change-Id: Ica280a0d6b5b0870aa2d05c4d059a1e559e6b12a

Remove threadlocal engine strategy, engine strategies pool threadlocal

2019-07-06T20:15:35+00:00

Change-Id: I8a351f9fa1f7dfa2a56eec1cd2d1a4b9d65765a2

Black updates

2019-07-06T20:13:36+00:00

Updating for a few black adjustments

Change-Id: I9c8abadd1ccd6173e6d68dd62f14cec208cc304a
(cherry picked from commit 17413bbf10c3b993d343c28035713a23c745712c)

Remove lib/ source prefix

2019-07-06T20:13:36+00:00

This prefix was a holdover from early days when Python
standard practices were not as clear and we had come from
a Perl background as far as scripting languages were concerned.

Specifically, this document:

https://docs.python.org/2/distutils/setupscript.html#listing-whole-packages

suggested using lib/ was perhaps still a common thing, and maybe it was.
But soon after, it was definitely not.  git seems to travel well as far
as things like cherry-picks and such (also not a thing we had back then).

Change-Id: Id528da1c8ec902b5d50fb360af62db366b03b419

- version number is also hardcoded in conf.py

2019-07-06T20:13:36+00:00

Change-Id: Ia8d0c0c47e2229ac404319328c8b865cde29e12e

- initial 2.0 setup

2019-07-06T20:13:36+00:00

Change-Id: Ib5af75df94b23104eebe0e918adcf979d798ea3b

class level docstring for Subquery

2019-07-06T19:52:23+00:00

Remove inherited member doc from most selectables, it's too verbose

Change-Id: I1b6635fe73342705846c0fab635b46557e560734

Merge "SelectBase no longer a FromClause"

2019-07-06T19:24:49+00:00

SelectBase no longer a FromClause

2019-07-06T17:02:22+00:00

As part of the SQLAlchemy 2.0 migration project, a conceptual change has
been made to the role of the :class:`.SelectBase` class hierarchy,
which is the root of all "SELECT" statement constructs, in that they no
longer serve directly as FROM clauses, that is, they no longer subclass
:class:`.FromClause`.  For end users, the change mostly means that any
placement of a :func:`.select` construct in the FROM clause of another
:func:`.select` requires first that it be wrapped in a subquery first,
which historically is through the use of the :meth:`.SelectBase.alias`
method, and is now also available through the use of
:meth:`.SelectBase.subquery`.    This was usually a requirement in any
case since several databases don't accept unnamed SELECT subqueries
in their FROM clause in any case.

See the documentation in this change for lots more detail.

Fixes: #4617
Change-Id: I0f6174ee24b9a1a4529168e52e855e12abd60667