Linking image comments to elements using kix anchors in Google Docs API

[sarahj]

I’m trying to connect comments with images in Google Docs using the API but running into issues.

When I get document elements using DocumentApp.getActiveDocument() and fetch comments through Drive API’s comment retrieval, I notice something weird.

Text comments include a context field like this:

{
  "message": "feedback on paragraph",
  "resolved": false,
  "state": "active",
  "context": {
    "format": "text/html", 
    "content": "highlighted text content"
  },
  "reference": "kix.abc123def456"
}

But image comments are missing the context:

{
  "message": "note about picture",
  "resolved": false, 
  "state": "active",
  "reference": "kix.xyz789ghi012"
}

Since image comments don’t have context data, how can I match them to specific images in the document? I’m building a converter tool and need to pair comments with their target elements. Should I assume all comments without context are on images? Is there a better approach to handle this anchor matching problem?

[Liam23]

yeah, this one’s tricky but i’ve handled it before. don’t assume missing context means it’s an image comment - tables and other elements also skip context fields. i loop through the document elements and match their anchor ids with the kix references. it’s messy but reliable for linking comments to actual images.

[HappyDancer99]

I hit the same issues building a document processing system last year. Google Docs API handles image comments weirdly in their internal structure. Here’s what worked for me: use a two-pass approach. First, iterate through all document elements and build a map of anchor IDs with their element types. Then cross-reference that with your comment data. The tricky bit is some images won’t have direct anchor associations depending on how they were inserted. When direct matching fails, check the parent element’s anchors - that often reveals the connection. Also, comments on grouped objects or drawings behave inconsistently, so add fallback logic for those edge cases.

[JessicaDream12]

The absence of the context field for image comments is expected since the Google Docs API does not provide contextual data for images as it does for text. Instead of relying on that field, you should utilize the kix anchor IDs for matching. When you access the document elements with DocumentApp, each element linked to comments will have anchor information correlating with the reference field from your Drive API results. To identify image elements, employ getType() to detect either DocumentApp.ElementType.INLINE_IMAGE or DocumentApp.ElementType.POSITIONED_OBJECT, ensuring you extract the related anchor data. However, be cautious not to presume that every comment without context pertains to an image, as drawings and tables also lack such fields. It’s essential to verify the element type to streamline your matching process.