Skip to main content

Configure Action reference

Reference

The Configure Action modal is the dense, single point of configuration for every source, target, script, and notification node. The reference below covers each family in turn (Generic API, Email/SMTP, File Transfer, Database, Salesforce) with field-by-field guidance.

Many nodes use "Configure Action" to define what the node does (query/proc/script/write).

Common Modal Behavior​

Across systems:

  • Configure Action modals open maximized by default.
  • A footer contains key actions (Save, Preview when supported).
  • Some modals support a maximize/minimize toggle near the close button.

Generic API Configure Action (Generic API / REST)​

Generic API actions let you configure HTTP requests as reusable actions tied to a Generic API connection.

Supported methods:

  • Source actions: GET, POST (emit rows from the response)
  • Target actions: POST, PUT, PATCH, DELETE (send incoming rows to the API)

Request configuration:

  • Method
  • Path (relative to the connection Base URL)
  • Headers (key/value; enable/disable per row)
  • Query Params (key/value; enable/disable per row)
  • Templates:
    • Path/body fields support templates like {field:SomeColumn}.
    • Preview Variables (JSON) can be used during Preview to render templates without needing real incoming rows.

Preview + schema options:

  • Preview button:
    • Executes the request via the backend (avoids browser CORS issues).
    • Shows URL, status, rows, and raw JSON.
    • Saves the preview response JSON into the action draft as a response example.

Email (SMTP) Configure Action (Target)​

If a Target node selects an Email (SMTP) connection, Configure Action defines the email to send and (optionally) how to attach the incoming rows.

Fields:

  • To, Cc, Bcc: comma/semicolon-separated email list.
  • Subject
  • Message (plain text)

Attachments (optional):

  • Attach rows as file: when enabled, the target converts incoming rows into file(s) and attaches them to the email.
  • Formats: CSV, JSON, XLSX
  • Compress attachments: gzip (creates .gz attachments)
  • Split modes:
    • Single file: one attachment for the entire incoming stream/chunk.
    • Per row: one attachment per input row (small datasets only).
    • Group by: split into one attachment per group key (comma/semicolon/newline list of fields).
  • File naming:
    • File name template: supports tokens like {runId}, {nodeId}, {chunkIndex}, {groupKey}, {YYYYMMDD}, {HHmmss}, {SSS}, {epochMs}.
    • Append timestamp: appends a millisecond-level timestamp to reduce overwrites.

Output rows:

  • Email targets emit status rows so downstream nodes (or run inspection) can show what was sent:
    • dl_success, dl_error, dl_subject, dl_attachment_name, dl_bytes, and related fields.
    • Infers an actionSchemaInfo so downstream Transform nodes can map fields immediately.
  • Schema Strategy:
    • Preview from API (recommended when the API is reachable)
    • Import response example (paste JSON, infer schema)
    • Import schema (paste columns JSON)

Target mapping schema (for Transform output-side):

  • Provide Request Fields (for mapping) as a comma-separated list to control which fields appear for mapping into the target API action.
  • You can also paste a request example JSON object and click Use Example to populate request fields.

File Transfer Configure Action (SFTP/FTP/FTPS)​

File Transfer actions are configured on Source and Target nodes.

Source node operations:

  • List Directory
    • Reads a directory and outputs file rows downstream.
  • Download (Get File)
    • Downloads a remote file into runtime staging and outputs a contentRef column downstream.
    • Note: staging under stage/<runId>/... is temporary and is purged after the run completes (success/failure). If you need to persist the file, upload it to a target system or explicitly write it to a durable store.

Target node operations:

  • Upload File
    • Converts incoming rows into 1 or more files and uploads them to the remote server.
  • Create Directory
  • Delete Directory
  • Delete File
  • Rename

Remote directory validation and creation:

  • In Upload/Create/Delete directory operations, the UI includes a refresh/validate button beside "Remote Directory".
  • At runtime, uploads attempt to mkdir the remote directory first (best-effort) to reduce failures on fresh servers/volumes.

Upload File (Target) options:

  • Output format: CSV, JSON, XLSX
    • XLSX is supported for Excel output; .xls is not used.
  • Compress: checkbox (gzip)
    • When enabled, the uploaded file name includes a .gz suffix.
  • Dynamic file names:
    • Use a file name template with tokens.
    • Example: export_{YYYYMMDD}_{HHmmss}_{SSS}_{runId}.csv
    • Common tokens:
      • {runId}, {nodeId}, {nodeName}
      • {chunkIndex}, {partIndex}
      • {YYYYMMDD}, {HHmmss}, {SSS}, {epochMs}
      • {field:MyField} to inject a value from the current row (useful for per-row or group naming)
  • File splitting:
    • Single file (all rows)
    • Per row (one file per row)
    • Group by field(s) (one file per group value; composite keys supported with comma-separated fields)
      • If one of the group-by fields is blank for a row, the runtime will group it under a __blank__ group key (which typically produces an extra file). If you want a single output file, use Split = Single file.

Append timestamp:

  • When enabled, the runtime ensures a millisecond-safe suffix (adds {YYYYMMDD}_{HHmmss}_{SSS} when missing).
  • This is recommended when uploading multiple files quickly to prevent accidental overwrites.

List/Download filters:

  • Include (glob), Exclude (glob), Regex (advanced)
  • Example include: *.csv
  • Example exclude: *_tmp.csv

Parse File Content (Source):

  • For Download/Get File actions, you can enable parsing so the Source outputs rows (not just file metadata).
  • CSV is supported today:
    • Configure delimiter.
    • Choose whether the CSV has a header row.
    • If "CSV has header row" is enabled, use Use Header to infer column names from the remote file header.
      • If the remote file path does not exist or is a directory, the UI shows an explicit error.
  • Parsed rows include __fileName and __filePath for traceability.

Database Source Configure Action (SQL Server, MySQL/MariaDB, PostgreSQL, Oracle)​

Operation: Query (source nodes read data).

Modes:

  • Standard Query (No Code)
    • Choose table/view.
    • Choose fields:
      • select specific fields
      • or Select All (recommended for first setup)
    • Optional filters:
      • Where condition text area
      • Limit (Preview may cap at a fixed max)
  • Custom Query
    • Write a custom SELECT statement.
    • Preview executes with a safety limit (for example, Preview 20).
  • Proc Mode
    • Choose a stored procedure (and functions where supported).
    • Parameters are auto-populated and can be set to static values or mapped later.
    • For Source nodes:
      • Output columns become the action output schema available for downstream mapping.
  • Script Mode
    • Write a SQL script to execute (query a view/function/procedure).
    • In Script Mode:
      • Table/view/proc browsing is not required (script text is the source of truth).
      • Preview should execute the script directly.

Preview:

  • Source-node preview runs a limited number of rows for safety and speed.

Database Target Configure Action (SQL Server and other DB targets)​

Operation: Insert, Update, Delete, Merge (Merge is used instead of Upsert for SQL Server targets).

Modes:

  • No Code
    • Choose table.
    • For Update/Delete:
      • you must specify an Id/Key field for matching.
    • For Merge:
      • configure match behavior (for example, update target when matched; insert when not matched).
  • Proc Mode
    • Choose a stored procedure.
    • For Target nodes:
      • input parameters should be exposed as action input schema so they can be mapped in the Transform node.

Database Script Configure Action (SQL Server, PostgreSQL, MySQL/MariaDB, Oracle, Snowflake)​

Database Script actions run database-side scripts or procedures without requiring a table-based source or target action.

Modes:

  • Script Mode
    • Write executable SQL or PL/SQL text directly.
    • The Database selector is available in script mode and now loads metadata the same way as other DB action flows.
  • Procedure Mode
    • Choose a stored procedure to execute.
    • Procedure output can be returned as rows, first row, or no output depending on the action configuration.

Database selection behavior:

  • Override connection default database
    • Off: the action uses the linked connection's default database/service.
    • On: the action stores its own database override.
  • You can also pick a database directly from the dropdown; choosing one switches the action to use that override.
  • Default (from connection) keeps the action aligned with the linked connection settings.

Notes:

  • DB Script actions do not require table metadata to save.
  • In library projects, DB Script setup steps are commonly used to build and populate isolated lab assets before source and target workflows run.

Salesforce Source Configure Action​

Operation: Query.

Modes:

  • Standard Query
    • Choose object.
    • Choose fields (by describing the object).
    • Optional: WHERE clause text box to filter rows.
  • Custom Query
    • Write SOQL.
    • If a user enters SELECT * ..., the system should expand * to a full field list using describe (because SOQL does not support *).

Preview:

  • Preview shows the first 10 records (REST/SOQL).

API Type:

  • Depending on the action, you can choose REST or Bulk where supported.

Salesforce Target Configure Action​

Operations:

  • Insert
  • Update
  • Upsert (requires External Id field selection)
  • Delete

API Type:

  • REST or Bulk (Bulk API 2.0 Ingest for write operations, where enabled).

Output schema:

  • For DML actions, output includes status fields (success/id/errors) so you can join results downstream.