GitLab

The unit tests now create the schema before each test and destroy
it afterwards.  This will be more flexible than needing to update
the schema on test machines each time.  The drawback is the need
to keep the database creation script in step with the hardcoded
set of SQL statements in the test program.

It is easier to add a new statement to the head of the file rather
than searching through the file for the method in which the prepared
statements are created.

... where the elase is selected by address only.

Clear up after every test and check that commit works.

... traced to not initializing a "struct" properly.

... after moving creation of MYSQL_BIND structures to a separate class.

This change adds the code for prepared statement creation and the
template for retrieving information from the database.

Also includes first unit test, which fails.

with fixing Conflicts:

with fixing conflicts in
	ChangeLog
	examples/README

this works around a bit awkward installation setup where there's a link
from "python3" to "python3.x" but not from "python3-config" to the
corresponding "python3.x-config".  That happens for recent versions of
Homebrew.

Sunstudio is unhappy with plain NULL inside EXPECT_EQ and thinks it is
int. This makes it think otherwise.

This should fix compilation issue on sunstudio.

so it sounds more natural as an English term.

this branch introduced is_failed() (and failed() was already defined for
a different a purpose), so for consistency it would be better to name
the running version is_xxx too.