Is there an existing issue for this?
Category of feature request
Profiler
Problem statement
We need a programmatic way to identify the source system that was used as input to generate a profiler output (e.g., Redshift, Snowflake, etc.).
Currently, this information has to be inferred either from the file name or by inspecting the DuckDB file for source-specific tables. Both approaches are brittle and error-prone, making them unsuitable for reliable automation.
The profiler output should include explicit source metadata so that consumers can determine the originating source system programmatically without relying on heuristics.
Proposed Solution
Add a source-independent metadata table to the DuckDB file generated by the profiler.
At a minimum, this table should include the source system used as input (e.g., Redshift, Snowflake, etc.), allowing consumers to determine the originating source programmatically instead of relying on file names or source-specific tables.
Since this table is intended for generic metadata, it can also be extended in the future to capture additional useful information, such as the profiler version, generation timestamp, or other run-level metadata.
Additional Context
No response
Is there an existing issue for this?
Category of feature request
Profiler
Problem statement
We need a programmatic way to identify the source system that was used as input to generate a profiler output (e.g., Redshift, Snowflake, etc.).
Currently, this information has to be inferred either from the file name or by inspecting the DuckDB file for source-specific tables. Both approaches are brittle and error-prone, making them unsuitable for reliable automation.
The profiler output should include explicit source metadata so that consumers can determine the originating source system programmatically without relying on heuristics.
Proposed Solution
Add a source-independent metadata table to the DuckDB file generated by the profiler.
At a minimum, this table should include the source system used as input (e.g., Redshift, Snowflake, etc.), allowing consumers to determine the originating source programmatically instead of relying on file names or source-specific tables.
Since this table is intended for generic metadata, it can also be extended in the future to capture additional useful information, such as the profiler version, generation timestamp, or other run-level metadata.
Additional Context
No response