Given two connections, it outputs a file that represent the diference
between schemas. It means that if you run the output file into the
"first connection", it will have the same schema as the "second
connection". The main use case is to deploy changes into testing,
staging, and production environments.

pgquarrel doesn't rely on any tools (like pg_dump) instead it connects
directly to PostgreSQL server, obtain metadata from catalog, compare
objects and output SQL commands necessary to turn "first database" into
"second database".

It could work on different PostgreSQL versions. However, if the "second
database" version is greater than "first database" version, some SQL
commands could not work on a prior version when you try to run the
output file.

It was tested on Linux and FreeBSD. I didn't try Windows yet.

README file has a list of supported features. Not all of the objects
are covered but I will try to do my best to supported the most common
SQL commands.