Business Data Reading Rules
A form definition tells Soln what fields exist. A data reading rule tells it where each business record and field value is stored. Configure reading rules before using a scanned form in a dataset.
Configure The Record Source
- Open the form model's Data reading rules.
- Select the business-record source first.
- Configure the column that identifies the form when several forms share one table. The selector can use the main source or a merged source.
- Choose how field values are read and configure any additional sources.
- Review the field matching results and adjust individual fields.
- Select what each result row represents, preview the data, and save.
A shared table's form identifier must match the scanned form ID or code. For example, form_records.form_id matches a form ID; it does not identify an individual business record.
Field Value Locations
| Reading method | Example |
|---|---|
| Database column | Customer name comes from the name column |
| JSON path | Customer name comes from the customer_name property in a data column |
| XML path | Customer name comes from a node in an XML document |
| EAV attribute | Separate attribute rows contain the record identifier, field identifier, and field value |
Use actual source samples to select value locations. System fields may be stored outside the JSON/XML document even when ordinary fields are stored inside it.
Merge Business Data Sources
Use Form data merge when one form's business data is spread across multiple tables.
- Select the merge table and the existing source it matches.
- Add corresponding column pairs. Every condition must match.
- Add fixed-value conditions when required, such as an entity type or binding type.
- Choose whether unmatched records are retained with empty fields or omitted.
- Configure field matching so corresponding field codes or external field IDs can generate mappings from the merged table.
At least one condition must match fields between sources. Fixed values restrict the join and do not replace its record linkage.
For a profile-binding design, a typical chain is:
entities.id → bindings.entity_id
bindings.form_record_id → form_records.id
Include tenant or form matching where the source requires it, and limit the binding type as appropriate. A field may read its value from any configured source. Existing mappings take priority; if several sources could provide one field, choose it explicitly.
Choose The Row Meaning
This setting applies to the complete reading configuration in the editor.
| Row meaning | Use when | Multiple matches |
|---|---|---|
| One complete record | Each business record has at most one matching row in each merge source | Reading fails if a merge returns multiple matches |
| One related record | One business record can have several bindings or related rows | Returns a row for each match |
| One detail item | The result represents line items or expanded details | Returns a row for each detail |
For example, a member with two profile bindings does not satisfy One complete record unless matching conditions select one binding. Select a relationship/detail row meaning when the multiple rows are intentional.
Parent values repeat in one-to-many results. Summing a parent amount across those rows can count it more than once; choose the dataset grain and aggregations accordingly.
Adjust And Reuse A Field Mapping
In Field matching results, open Adjust for a field. Set its value location and reference value format.
Reference values can be direct IDs/codes, objects containing an identifier, or lists of identifiers/objects. Choose the identifier location inside objects when needed. Reference display names are resolved separately from the original value.
Use the apply-to-other-forms option for a shared field. Only forms with the same field code, type, and reading sources are listed. Saving replaces that field's value location and reference value format in the selected forms while keeping their names, options, and other fields.
The matching result distinguishes fields configured in the current rule from fields already covered by another rule. A field covered elsewhere does not need to be configured again solely because a new rule is being edited.
Expand Details And Combine Equivalent Tables
Expand JSON/XML detail items when a list represents line items. Select the list location, match its fields, and choose whether records without details are retained.
Use the operation for combining records from other tables when equivalent records are stored in separate tables, such as yearly partitions. Confirm the field mapping for every source; this differs from adding columns through a join.
Verify The Preview
Check several records for missing values, reference labels, date/time fields, and repeated rows. If a field is empty, compare its selected source and path with the actual sample before changing its display format.
Save and apply the reading configuration before creating or refreshing a dataset.