Grids AI Column Demos: Add Descriptions #31830
Conversation
There was a problem hiding this comment.
Pull request overview
This PR adds description files for the AI Columns demo in both DataGrid and TreeList components. These descriptions document the new AI column feature, which allows components to generate additional data fields using AI/LLM integration.
Key Changes
- New demo documentation explaining AI column integration and configuration
- Instructions for setting up AI columns with proper API references
- Explanation of how to optimize AI resource usage via the
onAIColumnRequestCreatinghandler
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| apps/demos/Demos/DataGrid/AIColumns/description.md | Adds comprehensive documentation for AI columns feature in DataGrid, including setup instructions and configuration details |
| apps/demos/Demos/TreeList/AIColumns/description.md | Adds comprehensive documentation for AI columns feature in TreeList, mirroring the DataGrid documentation with TreeList-specific API references |
You can also share your feedback on Copilot code review for a chance to win a $100 gift card. Take the survey.
| @@ -0,0 +1,12 @@ | |||
| DevExtreme DataGrid supports AI columns. These columns augment your data and transform our DataGrid into an AI-powered data exploration tool. This demo implements an AI column (fixed on the right side of the component) to display one additional AI-generated data field. You can also integrate multiple AI columns to expand the scope of generated insights. | |||
There was a problem hiding this comment.
The phrase "transform our DataGrid" should use "the" instead of "our" for consistency with other demo descriptions. Throughout the codebase, component references use "the DataGrid", "the TreeList", etc. The possessive "our" is typically reserved for marketing contexts (e.g., "our .NET-based DevExpress Reports") rather than component references.
| DevExtreme DataGrid supports AI columns. These columns augment your data and transform our DataGrid into an AI-powered data exploration tool. This demo implements an AI column (fixed on the right side of the component) to display one additional AI-generated data field. You can also integrate multiple AI columns to expand the scope of generated insights. | |
| DevExtreme DataGrid supports AI columns. These columns augment your data and transform the DataGrid into an AI-powered data exploration tool. This demo implements an AI column (fixed on the right side of the component) to display one additional AI-generated data field. You can also integrate multiple AI columns to expand the scope of generated insights. |
| @@ -0,0 +1,12 @@ | |||
| DevExtreme TreeList supports AI columns. These columns augment your data and transform our TreeList into an AI-powered data exploration tool. This demo implements an AI column (fixed on the right side of the component) to display one additional AI-generated data field. You can also integrate multiple AI columns to expand the scope of generated insights. | |||
There was a problem hiding this comment.
The phrase "transform our TreeList" should use "the" instead of "our" for consistency with other demo descriptions. Throughout the codebase, component references use "the DataGrid", "the TreeList", etc. The possessive "our" is typically reserved for marketing contexts (e.g., "our .NET-based DevExpress Reports") rather than component references.
| DevExtreme TreeList supports AI columns. These columns augment your data and transform our TreeList into an AI-powered data exploration tool. This demo implements an AI column (fixed on the right side of the component) to display one additional AI-generated data field. You can also integrate multiple AI columns to expand the scope of generated insights. | |
| DevExtreme TreeList supports AI columns. These columns augment your data and transform the TreeList into an AI-powered data exploration tool. This demo implements an AI column (fixed on the right side of the component) to display one additional AI-generated data field. You can also integrate multiple AI columns to expand the scope of generated insights. |
| @@ -0,0 +1,12 @@ | |||
| DevExtreme DataGrid supports AI columns. These columns augment your data and transform our DataGrid into an AI-powered data exploration tool. This demo implements an AI column (fixed on the right side of the component) to display one additional AI-generated data field. You can also integrate multiple AI columns to expand the scope of generated insights. | |||
There was a problem hiding this comment.
I like Sveta's remark about specifying the demo test case
After the first paragraph, add a note explaining that the demo includes a default prompt for the AI column. This prompt is used to determine the origin country of a car item. Users can modify this prompt or enter a completely custom one directly through the AI column UI.
| - Specify the column's [name](/Documentation/ApiReference/UI_Components/dxDataGrid/Configuration/columns/#name). | ||
| - Configure **columns[]**.**ai** options, including the [prompt](/Documentation/ApiReference/UI_Components/dxDataGrid/Configuration/columns/ai/#prompt) and generation [mode](/Documentation/ApiReference/UI_Components/dxDataGrid/Configuration/columns/ai/#mode). | ||
|
|
||
| This demo also configures the [onAIColumnRequestCreating](/Documentation/ApiReference/UI_Components/dxDataGrid/Configuration/#onAIColumnRequestCreating) handler to limit data included in AI requests. The default DataGrid behavior is to include all data from visible rows in AI requests, including data not bound to a column and data of hidden columns. This gives LLMs broader context, but increases the component's use of AI resources. To limit data included in AI requests, modify the **AIColumnRequestCreatingEvent**.[data](/Documentation/ApiReference/UI_Components/dxDataGrid/Types/AIColumnRequestCreatingEvent/#data) parameter. No newline at end of file |
There was a problem hiding this comment.
AFAIK we have some limitations for AI service that we use in this demo.
So, it's better to include some Note about it to warning users that the demo is for testing purposes and we've limited tokens...
Please check similar notes in other DX /not-DX products and consult with developer to finalize the text
3 participants
No description provided.