Selectors define which data to extract from a source and how to label it. The extracted data populates the connector's preview table, which is then available for transforms and field mapping.
Each selector has two parts:
- Header: The column name as it appears in the preview table. Used to reference the column in transforms and mapping. Must be unique across all selectors.
- Path: The location of the data to extract in the source. The supported path format depends on the source data type (e.g., JMESPath for JSON, CSS or XPath for HTML, column index or header value for spreadsheets). The path does not need to be unique — the same data can be extracted into multiple columns with different headers.
Note: Paths are not validated against the source data. This means that a selector may be added even if it does not represent a valid path containing data, which will result in an empty column. This is to allow for cases where the same connector configuration should be used across different source data sets where the format varies.
Default Selectors
Clicking + Add Default Selectors automatically adds selectors based on the format of the source data currently populating the preview table.
JSON
All non-null leaf nodes detected in the API response are added as selectors. The header and path are set to the same value.
HTML (including Crawler source)
Default selectors include Page Title, Page ID, and Page URL. CSS or XPath notation can be used for additional selectors.
.XLS/CSV
| Condition | Behavior |
|---|---|
HasHeaderRow = true |
All detected columns are extracted, using the first-row value as the header. If duplicate header names exist, the selector falls back to Alphabetic Index (CSV) or Numeric Index (Excel), and a number is appended to ensure uniqueness (e.g., two "Name" columns become Name[0] and Name[1]). |
HasHeaderRow = false |
All detected columns are extracted using Alphabetic Index (Excel) or Numeric Index (CSV). |
| Merging sheets vertically | Default headers are extracted from the first sheet only. |
| Merging sheets horizontally | Headers are extracted across all sheets. Duplicate headers across sheets are not allowed. |