Month: August 2018

Using the built-in JSON schema generator

Do you want to quickly generate a basic JSON schema in your editor from any JSON input to avoid some typing that needs to be done every time you start writing a schema? A quick skeleton schema as a starting point with all current properties defined and their integral types already resolved? If yes, you can do this in JSONBuddy with ease.

This is how you do it in the editor:

  1. Open your JSON data as document. It doesn’t matter if you use the text or grid view.
  2. Use the following command to open a new document with the generated JSON schema: “JSON | Generate JSON schema from document”.
Generate a JSON schema from the current document in the editor

The generated schema is opened as a new document in the editor. Integral types are resolved and all objects which are not part of the top level are added as definitions to the schema. Equal object definitions will only appear once.

Automatic schema generation in the background

But even if you don’t use the command to generate a schema, a JSON schema is automatically generated in the background for any JSON document with currently no schema assigned. This schema is used to fill the Elements pane and to provide the entry-helper windows.

JSON editor and validator for large schemas

Sometimes it can happen that a JSON schema is generated from existing data and the resulting schema is quite huge. And maybe this schema is just taken as a starting point and you need to extend or modify it to get the final specifications for your JSON input.

One example of a really huge JSON schema is the Ansible schema available from schemastore.org (Please take a look at this blog entry to see how easy it is to open any schema from schemastore.org in JSONBuddy). The Ansible schema has currently more than 100.000 lines and 5 MB of JSON text. For this task, you need a good JSON editor which can also handle quite large files.

JSONBuddy: The editor for large JSON schemas

JSONBuddy is capable of loading big schema documents. The editor will apply syntax-coloring, entry-helpers and code folding as it does for a typical schema. You also don’t have to work on your schema without the context-sensitive help of getting the specification text for the keyword you are currently editing. All of this is also functional for big schemas.

But there is more. For example, besides being able to edit a JSON schema as you are used to, you get an instant evaluation from the built-in JSON schema analyzer. The schema analyzer runs in the background and is also active for large schema documents. For example, if a certain definition is not used in the local schema, there is no need to run an explicit find operation:

Inserting a new definition in a JSON schema with more than 100k lines

And this also works the other way around, whenever you add a $ref keyword to your JSON schema in the editor, the analyzer will tell you if the referenced definition can be successfully resolved.

“And do I get those nice validation error indicators which are saving so much time on editing my JSON content in the editor?” Yes, you do:

Active background validation on using a huge JSON schema

JSON editor in a nutshell: Using JSON pretty-print

Maybe, you often have to deal with JSON data which is not formatted as you like or has no white space at all. As an example, it is hard to edit this JSON text in the editor:

JSON data with no white-space is hard to read
JSON data with no white-space

What you need is an editor that a) understands the JSON syntax and b) can quickly format your JSON data hassle-free.

Format JSON data with no white space in the editor

In JSONBuddy, just use the pretty-print feature to format the current text content at any time. The keyboard shortcut for the pretty-print command is Ctrl+Shift+p (you can change the shortcut in the “Options…” dialog). There is also a menu item at “JSON | Pretty-print JSON”. After the formatting the JSON data is much more readable and can be easily modified in the editor:

JSON data after running the pretty-print command.

Note that short array content is aligned at a single line (available with JSONBuddy 5). This makes the JSON more compact and is saving space in the editor window. This is also true for arrays with only primitive types and having no long string values. As a consequence, arrays with many items of only primitive types, often numbers, are still displayed in a compact way with no new-lines separating the single items. In contrast, arrays with long string values are displayed with each item on a new line to increase readability. The editor applies an adaptive JSON formatting based on the actual content.

Use pretty-print in combination with the navigation history

The pretty-print functionality is getting even more powerful in combination with the navigation history of the JSON editor. If you format your JSON data while you are adding new properties, objects or arrays, the navigation history sets the text selection to the current property, after you applied the formatting operation.

You can also use the “Navigate Backward Ctrl+Shift+,” and “Navigate Forward Ctrl+Shift+.” commands to navigate through the selection history as you need.