Fix missing CSV delimiter

Pass in CSV reader in order to set delimiter

Signed-off-by: Byron Ruth <b@devel.io>

Ignore drop table or view errors

Ignore errors during drop table or view

Other than a connection issue, the only error would be
the view or table does not exist.

Signed-off-by: Byron Ruth <b@devel.io>

1.2.3: Add method to initialize fields in profiler

This is only required if there are no data records in the CSV file.

Signed-off-by: Byron Ruth <b@devel.io>

Increase partition size, skip view creation if too wide

Increase max column partition size

Signed-off-by: Byron Ruth <b@devel.io>

Skip unique indexes on text columns

Skip unique indexes for text column types

Signed-off-by: Byron Ruth <b@devel.io>

Add auto-split support

Implement auto-split for very wide tables

Postgres has an upper limit of 1600 columns per table, but a CSV file
can theoretically have any number. This feature implicitly partitions
the columns into separate tables and creates a view on top of it to
join the data back together.

Signed-off-by: Byron Ruth <b@devel.io>

Support really long lines in CSV

Increase max scanner buffer size

Signed-off-by: Byron Ruth <b@devel.io>

Bulk loading

Change repo to library, move command to cmd/sql-importer

In addition, support for a bulk load has been added. The sql-importer
program can take a directory name and recursively imports CSV files
using the parent directory as the schema name.

Signed-off-by: Byron Ruth <b@devel.io>

1.0.0 Final

Add files for creating releases

Signed-off-by: Byron Ruth <b@devel.io>