Shopify Excel-Like json_string metafield type

We have a namespace that contains several metafields of type "json_string". When we create new instances of these metafields using Excel-Like Bulk Product Manager for Shopify, they are created as type "string".

With the exception of this behaviour, Excel-Like Bulk Product Manager works very well for our needs. Is there a method to change the type of a metafield? Alternatively, can the behaviour of the application be changed to respect the type of those metafields which already exist?

Good day,

Thank you for your attention and responsiveness in adding the feature to edit metafields of type JSON_string.

Upon setting the metafield type to "JSON string" and the editor type to either "simple" or "JSON editor" we notice two different behaviours:

First, the simple editor appears to still create type "string" (as revealed by exporting using Excelify; see attachment).



Second, when using the tabular JSON editor the type appears to be correctly set to "JSON string" (as seen in Excelify export). However, the tabular JSON editor converts existing arrays [...] to objects {...} which is not expected, and not handled properly by our theme, or other apps that edit metafields (e.g. Accentuate; see attachments).






Can you advise as to whether this behaviour is in error, or can be corrected?

Thank you again for your care and attention.

I am pleased to report that the new JSON Array string type functions as expected, and that the tabular editor works well.


The results render properly in our template, and other metafield apps. Well done!




Choosing the simple editor also allows for whole arrays to be copied and pasted from one field to another.

The only unexpected behaviour occurs in split view, where the HTML/code editor appears, but changes don't appear to be saved.



(As a feature suggestion, it would be useful to provide access to both the simple editor in the main grid view, and the tabular editor in the split view, for flexibility. Accentuate also helpfully un-escapes the content of each element and renders the resulting HTML, which is extremely user-friendly.)

For the most part, the main issue is resolved and we can now edit our data without error. Thank you for your responsive support.