Closes #10802.

pg, fix Infinity and NaN values conversion.

Closes #13334.

* rebased
* removed guard
* inlined guard into case statement

Before this patch `Infinity`, `-Infinity` and `Nan` were read as `0`.

Handle other pk types in PostgreSQL gracefully.

Closes #10505.

- rebased
- test slightly modified

In #10410 it was noted that you can no longer create PK's with the
type of bigserial in PostgreSQL in 4.0.0.rc1. This is mostly
because the newer adapter is checking for column type with the
id column instead of just letting it pass through like it did
before.

Side effects:
You may just create a PK column of a type that you really don't
want to be your PK. As far as I can tell this was allowed in 3.2.X
and perhaps an exception should be raised if you try and do
something extremely dumb.

Modify assert conditions not to be affected
by each database bind variable naming conventions

Closes #14999.

by each database bind variable naming conventions

Remove redundant tests method.

Improve documentation

correcting method documentation [ci skip]

Refactor AR's validations_test.rb

Replace :base with :in [ci skip]

Namely, if the mday is omitted but any other upper components are, then instead
of supplying the mday from the current time, it defaults to 1.

This is a follow up to #15058.

This exception is regularly raised during development. This means it will enter
the user realm. We should provide an API page to show that this exception is public API.

/cc @schneems

adding complete message in documentation [ci skip]

There may be situations where you need to tunnel SSL connections over
port 80 so we shouldn't remove it if it has been explicitly provided.

Change Gemfile’s double quotation marks in plugin generator

Since the generated application’s Gemfile is using single quotation marks, it
seems fair to use it on plugin’s Gemfile as well.

Keep method's description/documentation consistent [ci skip]

Closes #15024.

These operations happen before instances are created.
The current behavior is misleading and can result in broken behavior.

Fix broken proc syntax for 1.9.3

This lets us avoid

1. A slow call to Hash#slice
2. An is_a? test
3. Extra hash allocations (from slice)
4. String allocations

etc.

avoids extra hash allocations on each call

this is due to:

  https://bugs.ruby-lang.org/issues/5994