Bigeye lineage exports come in two files that work together: a nodes file, which contains every asset in your lineage graph, and an edges file, which contains the connections between the assets in your lineage graph. Use the nodes export to export a list of "what exists" and the edges export for a list of "how it flows."

Why Export Nodes, Edges, or Both

Export nodes for an asset inventory, to audit reports, to catalog what exists across sources, and perform coverage checks. It tells you what exists, not how assets are connected.

Export edges when performing impact and root-cause analysis (what feeds a dashboard, what breaks if a table changes), tracing end-to-end flow across warehouses, BI, and ETL tools, and loading lineage into an external graph or BI tool since each row already has both endpoints.

Nodes Export Details

In the nodes export, there will be one row per asset. Below is the data model for each row.

Column Name	Description
ID	The asset's unique ID in Bigeye.
type	What kind of asset it is - a schema, table, or column in a warehouse, or a BI/ETL object like a Looker dashboard or a PowerCenter load target.
name	The asset's display name.
sourceType	The kind of system it comes from (e.g. Snowflake, Oracle, Looker, PowerCenter).
sourceName	The name of the specific connected source. Warehouse assets show a source name; BI/ETL objects often leave this blank.
schemaName	The schema and table the asset lives in. Populated for warehouse tables/columns; blank for BI/ETL objects that aren't table- or column-level.
path	The full location, with levels joined by " > ". This is your cleanest field for uniquely identifying an asset.
originatingSrc	The originating source of the data, when captured via the “Originating Source” attribute.
transformExpr	Any transformation expression applied, when captured via any of the following keys: "Transformation Expression", "Transform Expression", "Sql Transform Expression", or "SQL Transform Expression"

Custom attributes: If your workspace has configured extra catalog attributes, they appear as additional columns after the transformExpr column. Custom attributes can only be added when enabled by Bigeye.

Edges Export Details

In the nodes export, there will be one row per connection, carrying the full details of both ends side by side: an "upstream" set of columns followed by the identical "downstream" set. Any configured extra attributes appear as paired "upstream " and "downstream " columns.

Read a row as "upstream feeds downstream," so this file is what you use for impact analysis, root-cause tracing, and mapping end-to-end data flow across your warehouses, BI, and ETL tools.

Below is the data model for each row.

Column Name	Description
upstreamId	The source asset's unique ID in Bigeye.
upstreamType	The kind of asset it is: a schema, table, or column in a warehouse, or a BI/ETL object like a Looker dashboard or a PowerCenter load target.
upstreamName	The source asset's display name.
upstreamSourceType	The system associated with the asset (e.g. Snowflake, Oracle, Looker, PowerCenter).
upstreamSourceName	The name of the specific connected source name that shows in the Bigeye catalog.
upstreamSchemaName	The schema the asset lives in. Populated for warehouse tables/columns; blank for BI/ETL objects that aren't table- or column-level.
upstreamTableName	The table the asset lives in. Populated for warehouse columns; blank for BI/ETL objects that aren't table- or column-level.
upstreamPath	The full location, with levels joined by " > ". This is your cleanest field for uniquely identifying an asset.
upstreamOriginatingSrc	The originating source of the data, when captured via the “Originating Source” attribute
upstreamTransformExpr	Any transformation expression applied, when captured via any of the following keys: "Transformation Expression", "Transform Expression", "Sql Transform Expression", or "SQL Transform Expression".
downstreamId	The source asset's unique ID in Bigeye.
downstreamType	The kind of asset it is: a schema, table, or column in a warehouse, or a BI/ETL object like a Looker dashboard or a PowerCenter load target.
downstreamName	The source asset's display name.
downstreamSourceType	The system associated with the asset (e.g. Snowflake, Oracle, Looker, PowerCenter).
downstreamSourceName	The name of the specific connected source. Warehouse assets show a source name; BI/ETL objects often leave this blank.
downstreamSchemaName	The schema the asset lives in. Populated for warehouse tables/columns; blank for BI/ETL objects that aren't table- or column-level.
downstreamTableName	The table the asset lives in. Populated for warehouse columns; blank for BI/ETL objects that aren't table- or column-level.
downstreamPath	The full location, with levels joined by " > ". This is your cleanest field for uniquely identifying an asset.
downstreamOriginatingSrc	The originating source of the data, when captured via the “Originating Source” attribute
downstreamTransformExpr	Any transformation expression applied, when captured via any of the following keys: "Transformation Expression", "Transform Expression", "Sql Transform Expression", or "SQL Transform Expression".

Custom attributes: If your workspace has configured extra catalog attributes, each one adds two columns - an "upstream " and a "downstream " - appended after the built-in columns (upstream ones first, then downstream). Custom attributes can only be added when enabled by Bigeye.