Linking inline image to attachment using API

[adam_carter]

Within some fields in JIRA, images can be added inline such as in the description or in comments. Although they appear inline, they are seemingly being added as attachments and then JIRA is maintaining a link somewhere. Unfortunately, I'm struggling to figure out what that link is when using the API.

Calling /rest/api/3/issue/<issue-key> against an issue where an image has been added inline returns the following:

{

                    "type": "mediaSingle",

                    "attrs": {

                        "layout": "align-start"

                    },

                    "content": [

                        {

                            "type": "media",

                            "attrs": {

                                "id": "056f8363-192c-4f50-85a6-9ce2f4dca583",

                                "type": "file",

                                "collection": "",

                                "width": 522,

                                "height": 176

                            }

                        }

                    ]

                },

Unfortunately, the ID shown here is not the id that is needed to access the attachment content. That is - rest/api/3/attachment/content/38477.

The Attachment property of the issue response does show the 38477 ID but there is no reference to the 056f ID shown in the inline comment that would enable a link to be established:

"attachment": [

            {

                "self": "/rest/api/3/attachment/38477",

                "id": "38477",

                "filename": "image-20220322-152329.png",

                "author": {<removed for brevity},

                    "displayName": "<user name>",

                    "active": true,

                    "timeZone": "Europe/London",

                    "accountType": "atlassian"

                },

                "created": "2022-03-22T15:24:03.145+0000",

                "size": 25485,

                "mimeType": "image/png",

                "content": "/rest/api/3/attachment/content/38477",

                "thumbnail": "/rest/api/3/attachment/thumbnail/38477"

            }

        ],

Obviously this is fine all the time there is only a single attachment, but as soon as more than one is added then I need to see the link somehow.

At present, I'm just trying to figure out how to make that link in order to get data out. Once I'm done, then I'll switch my attention to how to add an inline image via the API which I had previously assumed was a case of adding the image as an attachment and then referencing the returned ID when attempting to create the inline image. That now seems like that is not going to work either.

Any help anybody can give would be much appreciated.

[Sam Watkins]

After uploading an image attachment, you can do an HTTP request to fetch the attachment content, which will redirect to a media URL. The media ID that you need can be extracted from this media URL. You can get the media URL by doing a HTTP GET without automatically following redirects. There's no need to actually fetch the image, just look at the redirect URL (in the HTTP "Location" header).

For example, when adding an inline image to the description, you can get the media ID in this way, then update the description or comment to use that image.  I've tested this just now, and it works.

Here's an example.

If this is the main attachment URL:

https://example.atlassian.net/rest/api/3/attachment/12345

When we fetch it, the JSON contains a "content" property, which would be:

https://example.atlassian.net/rest/api/3/attachment/content/12345

When we do an HTTP GET on this content URL, it redirects to another URL using the Location header:

Location: https://api.media.atlassian.com/file/12345678-1234-1234-1234-123456789abc/binary?token=WHATEVER&client=WHATEVER&dl=true&name=something.jpg

We need that UUID "12345678-1234-1234-1234-123456789abc" that comes after /file/ in the redirected URL, this is the media ID.

You can then use that media id in the ADF, something like this:

{
  "version": 1,
  "type": "doc",
  "content": [
    {
      "type": "mediaSingle",
      "attrs": {
        "layout": "align-start",
        "width": 100.0,
      },
      "content": [
        {
          "type": "media",
          "attrs": {
            "id": "12345678-1234-1234-1234-123456789abc",
            "type": "file",
            "collection": ""
          }
        }
      ]
    }
  ]
}

If you need to go the other way, and find the attachment ID (12345) given the media ID (12345678-1234-1234-1234-123456789abc), you could scan all the issue's attachments one by one in this way.  In our case, we are tracking these attachment IDs and media IDs along with some other information in a database, so we don't have to query every attachment each time.

[Board Genius]

This is crazy. Thank you for figuring this out and sharing.

[Jehan Bhathena]

Three cheers to @Sam Watkins .

:-)

This is a great workaround.

[Lam Le]

Hey @Jehan Bhathena , I just tested this and wasn't able to find the location header. Are you? 

[Jehan Bhathena]

Hi @Lam Le ,

yes, I checked it via my browser console.

Unfortunately not sure how we can check this on a API tool like postman.

[Lam Le]

Ah, thanks for the guidance. Found it using my browser console, but still stuck on trying to find this location header using http get through 'Automation For Jira' as a header response. 

[Jenni]

Is there any better way to do this..? very painful.....T^T

[David Bakkers]

If you need a quick and dirty way to get the attachment ID when getting the contents of a rich text field such as Comments or Description, get the HTML rendered output.

You do this by adding ?expand=renderedBody to your get request for that field, which will produce some extra JSON content, like this:

"renderedBody": "<p>This comment refers to the </p><p><span class=\"image-wrap\" style=\"\"><img src=\"/rest/api/3/attachment/content/10017\" height=\"492\" width=\"708\" style=\"border: 0px solid black\" /></span></p><p> attachment with the name test.txt</p>"

See, there is the ID of the attachment (in this case '10017') at the end of the path to it.

[Phil Maddaloni]

@David Bakkers this is not the same id that is being discussed. The ID that you refer to here is the attachment id. There is another ID that is the media services ID (a longer UUID-like ID that is used for embedding images directly within content).

[Roy Davis]

Hi adam.carter

Did you ever figure this out?

thanks

- Roy