A Julia REPL mode for PostgreSQL powered by ReplMaker.jl, LibPQ.jl and SQLStrings.jl.
Install the package with Julia's Pkg mode:
pkg> add SQLREPL
To connect the REPL mode, you'll need a connection string for your Postgres database. You can then use:
julia> using SQLREPL
julia> SQLREPL.connect("your_connection_string")
REPL mode SQL initialized. Press ) to enter and backspace to exit.
"Prompt(\"SQL> \",...)"Now press ) to enter the REPL mode. You can create tables and do some simple
data insertion and extraction with standard SQL syntax:
SQL> create table foo (x text, y int);
SQL> insert into foo values ('hi', 1);
SQL> insert into foo values ('ho ho', 2);
SQL> select * from foo
2×2 DataFrame
Row │ x y
│ String? Int32?
─────┼─────────────────
1 │ hi 1
2 │ ho ho 2Thanks to SQLStrings.jl, you can also interpolate local Julia values into your
expression. Let's set min_y in the Julia Main module:
julia> min_y = 2And we can now use $min_y within our queries:
SQL> select * from foo where y >= $min_y
1×2 DataFrame
Row │ x y
│ String? Int32?
─────┼─────────────────
1 │ ho ho 2For more complex data manipulation, the REPL mode can be combined with programmatic access via the normal Julia REPL:
julia> using LibPQ, SQLStrings
julia> conn = LibPQ.Connection("");
julia> for y=1:10
msg = "Hi $y"
LibPQ.execute(conn, sql`insert into foo values ($msg, $y)`)
endthence
SQL> select * from foo
12×2 DataFrame
Row │ x y
│ String? Int32?
─────┼─────────────────
1 │ hi 1
2 │ ho ho 2
3 │ Hi 1 1
4 │ Hi 2 2
5 │ Hi 3 3
6 │ Hi 4 4
7 │ Hi 5 5
8 │ Hi 6 6
9 │ Hi 7 7
10 │ Hi 8 8
11 │ Hi 9 9
12 │ Hi 10 10To edit multi-line SQL statements easily, surround your statement with brackets:
SQL> (select * from foo
where y > 5
and y <= 7)
2×2 DataFrame
Row │ x y
│ String? Int32?
─────┼─────────────────
1 │ Hi 6 6
2 │ Hi 7 7Alternatively, to insert a line, the usual key binding ALT+Enter can always be used.
The resulting DataFrame is available in the ans variable back in the Julia
REPL. Starting with
SQL> (select * from foo
where y > 5
and y <= 7);we then have
julia> ans
2×2 DataFrame
Row │ x y
│ String? Int32?
─────┼─────────────────
1 │ Hi 6 6
2 │ Hi 7 7To inspect table schema you can use the psql-like meta-command \d:
SQL> \d foo
2×5 DataFrame
Row │ column_name data_type character_maximum_length column_default is_nullable
│ String? String? Union{Missing, Int32} String? String?
─────┼───────────────────────────────────────────────────────────────────────────────
1 │ x text missing missing YES
2 │ y integer missing missing YESIn the future we might implement more of the
psql meta-commands.
This package arose from a discussion on Julia discourse.