Working with Sources (Edge)

A Source connects Activate to the data system that holds the data you want to send to your business applications. It defines the connection to your data system, rather than the specific data you want to send.

You do not need a separate Source for each table or query. You can create multiple datasets on the same Source, and each dataset determines which rows and columns Activate reads through that connection. However, a dataset can read from only one Source.

Read Familiarizing with the Sources UI for the pages you use to view and manage your Sources. For the steps to create or edit a Source, refer to the documentation for your respective Source type.

Note: You cannot delete a Source while it is being used by a dataset or an Activation.

---

Source Database and Schema

Your Source specifies the database, and in some cases the schema, in which your dataset queries run. This context lets you reference a table in your dataset query without naming the database or the schema. The fields that provide it depend on your Source type:

• Snowflake: The Default Database Name and Default Schema Name fields. Both are optional, so you can leave either one blank.

• Amazon Redshift: The Database Name field, which is required. Every dataset on the Source reads from that database. An Amazon Redshift Source does not specify a schema.

The way you write a table name in your dataset query depends on what your Source specifies:

• Neither a database nor a schema: Use the fully qualified <DATABASE_NAME>.<SCHEMA_NAME>.<TABLE_NAME> format. This applies to a Snowflake Source that leaves both fields blank.

• A database but no schema: Use the <SCHEMA_NAME>.<TABLE_NAME> format. This applies to every Amazon Redshift Source, and to a Snowflake Source that specifies only the default database.

• Both a database and a schema: You can reference a table using <TABLE_NAME>. This applies to a Snowflake Source that specifies both fields.

Note:

• A default schema applies only when the default database is also specified.

• The database and schema names are case-sensitive.

---

Bookkeeping

Activate uses bookkeeping to identify the rows that changed since an Activation’s last successful run.

For each Activation, Activate keeps a copy of the data returned by the previous successful run and compares it with the current result. The comparison allows Activate to identify which rows have changed and send only those rows during subsequent runs.

Activate stores bookkeeping data in a dedicated schema in your data system. The schema also stores the records that Activate could not send to the Target. Activate reports these records as failed. Where this schema lives depends on your Source type:

• Snowflake: A database named HEVO, and a schema inside it that is also named HEVO.

• Amazon Redshift: A schema named hevo, in the database that your datasets read from.

You must consider the following about bookkeeping:

• Activate does not modify the tables your datasets read from. The bookkeeping schema is the only location in your data system where Activate writes data.

• You must create the bookkeeping schema yourself. Activate does not create it when you configure a Source, so create it as part of setting up your Source. The Source connection test fails until the schema exists.

• Do not store your own data in the bookkeeping schema. Activate reserves it for bookkeeping and manages the tables within it, so Activate may modify or remove any data you store there.

Refer to the documentation for your respective Source type for the steps to create the bookkeeping schema.

---

See Also

• Familiarizing with the Sources UI

• Amazon Redshift

• Snowflake

• Deleting Sources