Metadata Analyzer
Analyze, document, and publish metadata. The action walks a project folder for
pipelines (*.hpl), runs Hop’s impact analysis on each one, and writes a
documentation tree describing which tables and fields every pipeline reads and
writes — with field-to-source tracing per output table. Output is HTML or Markdown.
Options
Section titled “Options”| Option | Description |
|---|---|
| Source folder (pipelines) | Root folder scanned for *.hpl files. |
| Output folder | Where the documentation is written. |
| Output format | HTML or Markdown. |
| Include sub-folders | Scan sub-folders recursively. |
Output
Section titled “Output”For each pipeline the action writes one document at the output root, named after the
pipeline, listing every output table. Each output table also gets its own detail
document under the table/ sub-folder. Recognized output transforms are Bulk
Loader, Staging Upsert, and Table Output.
Every written column is documented with four columns:
| Column | Meaning |
|---|---|
| Field | The column written to the output table. |
| Expression | The source field or expression it comes from. |
| Source | Where the value originates. When the column comes from a database lookup, this is a link to the looked-up (for example dimension) table’s document; otherwise it is the source table or the transform name as plain text. |
| Type | The column data type. |
Main source table
Section titled “Main source table”When a pipeline has a description, the action shows it as the pipeline’s main source table at the top of the document. Set the pipeline description to the main source table name to make it appear.