Superset: Dataset Creation Without Immediate Charting
Hey data enthusiasts! Have you ever felt a little boxed in when creating datasets in Superset? Currently, the platform forces you into chart creation immediately after you set up a new dataset. This can be a bit of a drag, especially when you're just organizing things, preparing for future visualizations, or handling batch operations. Well, there's a cool feature request brewing, and it's all about giving you more control over the dataset creation process. Let's dive into the details, shall we?
The Problem: Friction in Dataset Creation
The current workflow in Superset can be a tad clunky. When you create a new dataset, you're automatically whisked away to the chart creation page. This assumes that every dataset immediately needs a visualization. While that's often the case, it's not always true. Sometimes, you just need to set up the dataset, save it, and come back to the charting later. This current approach adds unnecessary steps and makes the process feel less flexible.
Imagine you're setting up a bunch of datasets for a new project. You want to get them all prepped and organized before you start building your charts. With the current system, you'd have to either create a dummy chart just to get back to the dataset list, or you'd have to cancel out of the chart creation every time. It's not the end of the world, but it definitely adds some friction to your workflow. This new feature aims to smooth things out and let you work your way. This enhancement will significantly improve the user experience by providing flexibility in the dataset creation workflow, allowing users to choose whether they want to proceed to chart creation or return to the dataset list after creating a dataset.
Current Behavior Breakdown
Let's break down exactly what happens when you create a dataset right now:
- You navigate to the "Add Dataset" page.
- You choose your database and table.
- You click the "Create" button.
- Boom! You're instantly redirected to the chart creation page (
/chart/add/?dataset={dataset_name}). - There's no way to just create the dataset and stay put on the dataset management page without first canceling out of chart creation.
This is where the proposed change comes in. It's all about giving you choices.
The Solution: More Control Over Your Workflow
The goal of this enhancement is to give users more control over their workflow. The proposal suggests changing the "Create" button to a dropdown, offering two distinct options:
- Create and go to chart creation: This maintains the current default behavior, keeping things familiar for existing users.
- Create and return to the dataset list: This is the new option, allowing you to create the dataset and go back to managing your datasets.
This will make it a lot easier to manage your datasets, especially when working on larger projects where you might not be ready to create visualizations right away.
Expected Behavior Explained
Here's what the team is aiming for with this update:
- The "Create" button will transform into a dropdown.
- The default action (clicking the main part of the button) will still take you to chart creation.
- The dropdown menu will include a "Create dataset only" option.
- Selecting "Create dataset only" will redirect you back to the dataset list page (
/tablemodelview/list/). - The button will remain disabled if no table is selected or if the dataset already exists.
This setup keeps the current default behavior intact while adding a valuable new option. This ensures backward compatibility, and the new feature enhances the user experience. All existing tests will pass and new tests cover both navigation paths.
Testing and Verification
To ensure everything works as expected, here's the testing plan:
Manual Testing Steps:
- Go to the "Add Dataset" page.
- Choose your database and table.
- Check that the create button appears as a dropdown with an arrow.
- Click the main part of the button and verify you're taken to the chart creation page.
- Repeat steps 2-3, and then click the dropdown arrow.
- Select "Create dataset only" from the dropdown.
- Confirm you're redirected to the dataset list page (
/tablemodelview/list/). - Double-check that the dataset was created successfully.
This testing process ensures that the new feature integrates seamlessly into the existing workflow.
Benefits of the New Feature
So, what's the big deal? Why is this change so important?
- Enhanced Flexibility: Users have more control over the dataset creation process, allowing them to tailor their workflow to their specific needs. It helps users who want to organize data without immediately creating charts.
- Improved User Experience: Reduces unnecessary steps and navigation, leading to a smoother and more efficient workflow. Makes Superset more intuitive and user-friendly.
- Better Organization: Enables users to create and manage datasets more effectively, especially when preparing for future visualizations or handling batch operations. Users can build their datasets and later create charts.
- Increased Productivity: By streamlining the dataset creation process, users can save time and focus on more important tasks, boosting overall productivity. Reduces friction when setting up multiple datasets.
Conclusion: A Step Towards a Better Superset
This enhancement is a thoughtful addition to Superset that addresses a common pain point. By giving users the option to create datasets without immediately jumping into chart creation, it improves flexibility, streamlines workflows, and enhances the overall user experience. The changes are designed to be intuitive and non-disruptive, preserving the current functionality while adding valuable new options. It's a win-win for anyone who uses Superset to manage and visualize their data.
By following the proposed changes, Superset will become a more powerful and flexible tool for data exploration and visualization. It's about giving users the power to work the way they want, making data analysis a more enjoyable experience for everyone. So, let's look forward to this update that simplifies the dataset creation process and makes data workflows smoother and more effective!