Google changes how it parses double-escaped JSON-LD entities

Google has modified the way it extracts JSON-LD information, bringing it extra in keeping with JSON and different internet requirements. As Google Search Central introduced on LinkedIn, Google will solely do a single go of HTML unescaping.

What builders ought to know

If builders are utilizing JSON-LD for structured information, it’s time to replace your code. As a substitute of counting on double-escaped entities, swap to straightforward JSON escapes or Unicode hexadecimal escapes.

Since Google’s parser now solely unescapes as soon as, any code that relied on double-unescaping gained’t work the way in which it used to.

Google Search Central posted about this on LinkedIn:

“To convey our parser as much as JSON and different requirements, we modified our JSON-LD extraction and at the moment are solely making use of a single go of HTML unescaping. Virtually talking, because of this double-escaped entities (like & or ✔) will not be unrolled. If you happen to’re utilizing JSON-LD for structured information, make sure you replace your code to straightforward JSON escapes or Unicode hexadecimal escapes (like u0026).”