Import

The IMPORT statements imports tables and functions into an SQRL script. Tables can either be imported from an external data source or another SQRL script.

Tables and functions are imported from packages. The import path is the name of the package followed by the name of a specific table or function to import or the asterisk character * to import all tables and functions from the package. Import paths use the dot character . to separate path components.

IMPORT datasqrl.seedshop.Orders;
IMPORT time.*;

The first statement imports a table named Orders from the package datasqrl.seedshop. The second imports all functions from the time package.

Package Resolution

Packages are either part of the standard SQRL library, declared as a dependency, or stored in a local directory.

Locally

Packages are looked up locally first. The package name path is converted to a directory path relative to the directory in which the DataSQRL command is invoked.

A package with the name my.data is translated to the relative directory path my/data (on Unix based systems) or my\data (on Windows).

If this directory exists, imported tables and functions are located inside that directory.

Dependency

Dependencies are declared in the package configuration and downloaded from a repository at compile time.

By default, DataSQRL looks up any missing packages in the repository. A package is missing if it is not declared as dependencies and cannot be resolved locally. If the missing package can be located in the repository, a dependency on the most recent version of that package is added to the package configuration.

Take a look at the DataSQRL repository to find public packages you can import into your script. (coming soon)

You can publish your own packages to the repository, so they can be used as dependencies in your SQRL scripts.

Standard Library

The SQRL standard library contains function packages that can be imported without any configuration. For example, time is a standard library function package that contains functions for manipulating timestamps. See the list of function packages for more information.

Changing Names

To import a table or function under a different name, use AS directly after the import path:

IMPORT datasqrl.seedshop.Orders AS MyOrders;
IMPORT time.epochMilliToTimestamp AS epoch2TS; 

This statement imports the same Orders table from the datasqrl.seedshop but makes it available under the name MyOrders in the script. Likewise, the epochMilliToTimestamp is imported under the name epoch2TS.

You can change the name of imported tables and functions to resolve naming conflicts or choose names that are a better fit for your script.

Setting Timestamps

Timestamps are important for stream tables. In many cases, the DataSQRL compiler can determine the timestamp of a stream table. In other cases, you can declare the timestamp of a data source table explicitly:

IMPORT datasqrl.seedshop.Orders TIMESTAMP timestamp;

This statement declares that the timestamp column on the Orders table must be used as the timestamp.

The timestamp declaration can also be an expression to process the input data to produce a valid timestamp.

IMPORT time.epochMilliToTimestamp;
IMPORT datasqrl.seedshop.Customers TIMESTAMP epochMilliToTimestamp(changed_on) AS timestamp;

In this import statement we convert the integer changed_on column to a timestamp via the epochMilliToTimestamp function. When an expression is used to declare a timestamp it is added as a column to the table and must be given a name after the AS keyword.

info

Declaring a timestamp is only possible for tables that are imported directly from a data source.

Creating A Package

To create a new package to import your own tables or functions into a SQRL script, read the following documentation:

• Data Source Package: to create a package that contains tables from an external data source
• Function Package: to create a package that contains custom function implementations
• SQRL Script (coming soon): to import tables defined in another SQRL script