Unsaved changes warning or auto save option on tab change

Theoretically tabs inside drawers would be a very convenient way to structure large input forms (much better than using those group dividers one can use to give a bit of orientation in a large form). But practically this is no option since switching to another tab leads to a loss of the data you just entered into the form inside of the tab you intend to leave without any warning. So either there should be a warning dialogue popping up when one clicks on a tab link or an autosave option should be configurable for this purpose.

Please use https://docs.nocobase.com/handbook/block-multi-step-from instead of tabs.

Thank you for the hint. But it seems that a multi step form could only be the solution for a linear data collection approach. If you need to jump between different aspects of your entity frequently a multi step form might not be the way to go.

You’re right. However, tabs are not meant to be used for splitting forms. Could you provide a specific example of a form?

I’m thinking of an applicant tracking system that could be used to collect important information about the candidate during an interview. This information can be divided into the following areas, for example, which could be mapped in tabs:

  • Education and professional experience (career steps, CV, references, …)
  • Current conditions (salary, bonus model, vacation, mobile working, company car, …)
  • Desired conditions (salary, vacation, mobile working, company car, …)
  • Change ambition and general conditions (reason for a change, notice period, remaining contract term, earliest start date, …)
  • Fit of the candidate to the individual requirements of the position profile

You can try to conduct such interviews in the structure and order of a long form. However, experience shows that you always have to skip back and forth because the interview develops differently and the candidate provides information that has not yet been asked for, for example. To enable a natural flow of conversation, data collection should adapt to the course of the interview and not the other way around.