a!gridField_25r3 Component

Function

a!gridField_25r3( label, labelPosition, instructions, helpTooltip, emptyGridMessage, data, columns, pageSize, initialSorts, secondarySorts, pagingSaveInto, selectable, selectionStyle, selectionValue, selectionSaveInto, selectionRequired, selectionRequiredMessage, disableRowSelectionWhen, validations, validationGroup, showWhen, spacing, height, borderStyle, shadeAlternateRows, rowHeader, accessibilityText, refreshAlways, refreshAfter, refreshInterval, refreshOnReferencedVarChange, refreshOnVarChange, userFilters, showSearchBox, showRefreshButton, showExportButton, recordActions, openActionsIn, actionsDisplay, actionsStyle, maxSelections, showSelectionCount, loadDataAsync, similarityScoreThreshold )

Displays a grid of data as rich text, links, images, buttons, tags, record actions, or progress bars. This grid component also supports searching, filtering, selecting, sorting, and paging. For example:

Scroll or drag to pan · ⌘+/⌘− to zoom · Esc to close

See also:

• To learn how to work with the read-only grid, see Configuring the Read-Only Grid or the Grid Tutorial.
• To organize editable inputs in a tabular layout, use an editable grid.

Parameters

Usage considerations

Choosing the source of the grid

The fastest way to configure a read-only grid is from design mode. When you drag and drop the component into the live view, you can select one of the following options as the grid's data source:

Using records-powered grids

Consider the following when creating a records-powered grid.

Using records-powered grids in a portal

When you use a records-powered grid in a portal, there are a few things to keep in mind:

• Some record components don't work in portals: Certain record components, such as record links and record actions, are incompatible with portals. When you create a read-only grid using a record type, it may automatically use these components, so be sure to remove the incompatible components before publishing your portal.
• Users can't save their filters in a portal: The record list search, refresh button, export to Excel, and user filters will all function in a portal. However, users can't save their filters in a portal since they aren't signed in. The Manage filters button is replaced with a Clear Filters button in a published portal.
• Asynchronous loading is not compatible with portals.

Scroll or drag to pan · ⌘+/⌘− to zoom · Esc to close

See Working with Data in Portals for more information about using records as the source for read-only grids in portals.

Autopopulate a grid with a record list

When a record type is used as the data source for your read-only grid, Appian automatically generates grid columns based on the columns configured for the record list.

Specify which fields to query and when

By default, a grid will query and display all the data specified in the columns, pagingSaveInto, and selectionSaveInto parameters. This default behavior can potentially impact performance if your grid includes dynamic behavior since it may result in querying more data than necessary.

For example, let's say you configured the showWhen parameter on the columns in your grid so that certain columns only appear to managers. Even when columns are hidden from the grid, those fields will still be queried since they're specified in the columns parameter of the grid.

To alter this default behavior, you can specify when the grid should query the fields in your columns using the fields parameter in a!recordData(). Learn more about using the fields parameter in a!recordData().

Reference related data

If you use a record type with data sync enabled as the source of your grid, you can also reference related record fields in your grid.

You can use related record fields in your columns, or to sort and filter the grid.

Filter, limit, and sort related one-to-many data

If you reference a related record field from a one-to-many relationship as a column in the grid, the maximum number of related records returned for each base record is 100.

To filter, limit, and sort the related record set returned from the one-to-many relationship, you can leverage the relatedRecordData parameter in a!recordData().

Avoid using relationship references to display related one-to-many data. Instead, you should only reference the specific related record field you want to display. This avoids querying more data than necessary, which can impact the grid's performance and export behavior.

For example, to display a list of cases associated with each customer, you should use the related record field reference recordType!Customer.relationships.cases.fields.title in the grid column. You should not use the relationship reference recordType!Customer.relationships.cases.

Use records-specific components

You can display the following components on any records-powered grid:

• User filters
• Search
• Export to Excel button
• Record actions

Export data to Excel

When set to true, the showExportButton parameter allows users to export the grid data to Microsoft Excel. By default, any columns visible in the grid will export to Excel. You can specify which columns are exported by configuring the exportWhen parameter on individual grid columns.

Users should only export grid data from applications and sites. You should not export grid data from an interface in Appian Designer.

Review how much data can be exported from the grid, and how certain display elements are exported to Excel.

Display and style record actions

The recordActions parameter allows you to display record actions, like related actions or record list actions, on the grid. You can display actions from different record types.

You can then use the actionsStyle parameter to determine whether all actions display as with standard styling, or if the first record action displays as a primary button style. Appian recommends using the "TOOLBAR_PRIMARY" option when you want the first action, typically the most common action, to stand out.

When actionsStyles is set to "TOOLBAR", the buttons display like this:

Scroll or drag to pan · ⌘+/⌘− to zoom · Esc to close

When actionsStyles is set to "TOOLBAR_PRIMARY", the buttons display like this:

Scroll or drag to pan · ⌘+/⌘− to zoom · Esc to close

See the record action component for other details to consider when adding record actions to a grid.

Record action security

Each action specified in the recordActions parameter has security that is checked to ensure only the right users can see the right actions. In most cases where you use security rules, Appian evaluates the security for the entire grid before it is shown to the user. Because security rules are verified in bulk, the overall grid loading time will be shorter than if security expressions are used.

Row limit for records-powered components

Interface components that use a record type as the data source are known as records-powered components.

Records-powered components can display a maximum of 5,000 rows. As a best practice, you should not display all 5,000 rows in a component since it may impact performance.

Using the loadDataAsync parameter

When you use a record type as the grid's data source, you can set the loadDataAsync parameter to true to allow the interface to load without waiting for the grid data. A placeholder will display in the grid's place, enabling users to interact with other parts of the interface while the grid's data loads in the background.

If the grid's data is reused elsewhere in the interface, consider using a!asyncVariable() to create an async variable instead. See Asynchronous Loading for more information on choosing the right method.

If you are querying data directly in the data parameter of the read-only grid, you can still use the loadDataAsync parameter. However, it is typically simpler to just use an async local variable for the grid data.

Configuring grid display

• If the value for a column is empty or null, the grid still renders the column with just the label.
• The selectionStyle of ROW_HIGHLIGHT does not have any visual indicator when selection is disabled through the disableRowSelectionWhen parameter.
• For accessibility purposes, every grid should have a row header configured. The first column containing text is usually the correct choice for row header. See the SAIL Design System for more information.

Configuring grid paging

• To reduce clutter on smaller grids, the following happens:
  • No paging controls are displayed when all rows are on one page.
  • Paging controls to jump to the first or last page are not displayed when a grid has less than 3 pages.
• When a read-only grid is used offline, paging and sorting don't function and images won't render.
• If you are managing your own paging, you cannot pass a datasubset which has a paging configuration that uses a batch size of -1.
• When entering a query in the data parameter (either directly, or indirectly from a rule reference), the query must return the total count (fetchTotalCount: true).

Grid height and column width

• Grid heights behave as a fixed height on web but a maximum height on mobile.

• For columns using the "AUTO" width, each column's width is determined by the length of the longest value in that column.

  • See the read-only grid column component for more information on using the "AUTO" width.
  • For complete guidance on when and how to use column widths, see Column Widths.

Design mode guidance

The following guidance applies when configuring a read-only grid in design mode:

• Generated columns:
  • When you populate the data parameter from design mode and let the grid manage the paging, the grid will automatically generate an initial set of columns in the order they appear.
  • Grid columns are not automatically generated when you select VARIABLE as the data source and use a local variable to power the read-only grid.

  • The initial values for automatically-generated columns are:

    Parameter	Value	Description
    label	fv!row.<fieldname>	The field name is converted to title case. If the field name is in camel case, it will be split and converted to title case. If the field name is too long, it will be truncated with ellipses (...).
    value	fv!row.<fieldname>	The field value to display.
    sortField	<fieldname>	The field name of the primary key.
    align	"END" or fixed()	The alignment depends on the field's data type. If the data type for the column value is numerical or date-based, then it will set align : "END". If the data type for the column value is a decimal, then it will be wrapped in the fixed() function to two decimal places. Ex. value: fixed(fv!row.cost, 2).

• Updating columns:
  • After initial columns have been generated, the grid will automatically update with new columns to match any new fields added to the data source.
  • However, automatic updates do not occur when the data source is an RULE or EXPRESSION. To update the columns displayed in the grid, you must update the data source, and add or remove the columns manually.
• Selecting QUERY as the data source:
  • If you select QUERY as the data source, you'll use the query editor to configure the query that powers your grid.
  • Appian will automatically use the following query values in your grid:
    • The query's batchSize will be the value of the grid's pageSize parameter.
    • The query's sortInfo will be used as the grid's initialSort parameter.
• Batch size limitation:
  • The grid cannot use a query that has a paging info with a batch size of -1 in design mode.

Using smart search

You can use smart search across multiple text fields or a document field in a grid.

To use smart search in a read-only grid:

1. Enable smart search for Text or Document fields on a synced record type.
2. In an interface, add a Read-Only Grid component and set the source to your record type.
3. In the read-only grid, ensure the showSearchBox parameter is set to true.
4. For the similarityScoreThreshold parameter, change the value to less than 1. For example: 0.5. This will allow the grid to use both keyword search and semantic search. If you leave this parameter at 1, then only keyword search is applied.
5. (Optional) Add an initial sort on recordType!<Record Type Name>.searchResults.allSearchFields.similarityScore in descending order so the best matches appear first. This sort is ignored when no search term is entered and applies only after a user starts a search.

Search behavior in a!gridField

By default, the search box in a read-only grid uses a keyword search to find exact matches to a user's search terms. You can implement semantic search in this default search box, or by building a custom search box for the grid.

Depending on which search box you’re using, the way you implement semantic search and how it behaves will differ:

• In the default search box:
  • When implemented, both smart search and keyword search are applied to the grid. Smart search will apply to all columns that reference fields with smart search enabled, and keyword search will apply to all columns with text and number (integer) fields.
  • Learn how to implement smart search in the default search box.
• In a custom search box:
  • When implemented, smart search is only applied to fields that are filtered by a "search" operator.
  • Learn how to implement smart search in a custom search box.

Primary similarity score

When you search across multiple fields, a row may have several search matches. The highest similarity score found in that row is called the primary similarity score. Use the primary similarity score to filter and sort data by the strongest search matches, regardless of which field contained the match.

For example, if smart search is enabled on the title, description, and name fields, a semantic search for "electrical outage" in a read-only grid evaluates all three fields. To ensure users see the most relevant results first, sort the grid by the primary similarity score.

Sorting by similarity score

You can sort a read-only grid by similarity score to show the most relevant search matches first.

When you sort by similarity score, you must also configure a secondary sort. The secondary sort determines the grid's order when a user has not entered a search term, which means the similarity score is null.

To sort the grid by the highest similarity score:

1. In the initialSorts parameter of the read-only grid, define an array {} containing two a!sortInfo() functions.
2. In the first a!sortInfo() function, configure the following:
   • For field, enter recordType!<Record Type Name>.searchResults.allSearchFields.similarityScore. This will sort by the primary similarity score.
   • For ascending, enter false. This will sort from highest to lowest similarity score, placing more relevant results first in the list.

3. In the second a!sortInfo() function, choose a field to sort by when the grid first loads or when the search box is empty.

   For example:

1
2
3
4
5
6
7
8
9
10
11
12
  initialSorts: {
    /* Sorts by similarity score in descending order if there is a search */
    a!sortInfo(
      field: recordType!SSTA Case.searchResults.allSearchFields.similarityScore,
      ascending: false()
    ),
    /* Defaults to sorting by id field when there is no search */
    a!sortInfo(
     field: recordType!SSTA Case.fields.id,
     ascending: true
    )
  },

Tips for smart search

• Verify the fields you want to search are enabled for smart search.
• Make sure the smart search indexing status is Completed and not Failed or Running.
• Search time increases as the number of records and searchable fields grows. To improve performance, limit the number of fields and records being searched when possible. For example, searching 10,000 records is faster than searching 1 million records.
• Use allSearchFields.similarityScore whenever possible to sort and filter. Only use <<fieldName>>.similarityScore if your use case requires it. This field behaves like related records and may return unexpected results without a search. Add logic to check for a search term first.
• Don't filter or sort by sourceContent. sourceContent is designed to be a result of the search, not a criteria for filtering or sorting.

Examples

You can find example configurations of the grid at Configuring the Read-Only Grid and Grid Tutorial.

New version details

The read-only grid component was evolved to support a simplified similarity score range for smart search. This evolution ensures that the simplified score range is applied when smart search is implemented on the default search box or in a custom search box built for the grid.

The similarity score's decimal distribution was rescaled for greater clarity, moving from granular values (like 0.0026) to a more intuitive and readable range (like 0.4).

When updating to the latest function version, use our general guidance for interpreting similarity scores as a starting point. You'll need to test smart search with your data to determine what makes a score meaningful for your use case.

For example, if you previously used a similarity score of 0.0026 for the similarityScoreThreshold, you will need to experiment with the updated scale. You may try a score of 0.6 but it may not return enough results, so you might try a value of 0.4 to return more results instead. The significance of a similarity score will vary across data and use case.

Feature compatibility

Feedback