Use the REST Api to pull COMMENT attachments from an issue

[Fabian Dittrich]

Pulling regular attachments attached directly to an issue is no problem. 

When parsing comments though, this is how comment attachments are returned by the API:

____

"body"=>

  {"type"=>"doc",

   "version"=>1,

   "content"=>

    [{"type"=>"paragraph", "content"=>[{"type"=>"text", "text"=>"Hi ABC,"}]},

     {"type"=>"paragraph",

      "content"=>

       [{"type"=>"text",

         "text"=>

          "TEXT ABC?"}]},

     {"type"=>"mediaSingle",

      "content"=>[{"type"=>"media", "attrs"=>{"type"=>"file", "id"=>"5ac80cf3-82b7-4f59-a015-09dcbc3f18de", "collection"=>"", "width"=>2534, "height"=>266}}],

      "attrs"=>{"layout"=>"align-start"}},

     {"type"=>"paragraph", "content"=>[{"type"=>"text", "text"=>"Many thanks"}]}]},

___

Only the bold id 5ac80cf3-82b7-4f59-a015-09dcbc3f18de is returned, no filename, no URL and there seems to be no way to access an attachment with the above ID.

How can I get more info from the API about comment attachments?

[Hariharan Iyer]

Hi @Fabian Dittrich ,

There doesn't seem to be a way to do this using the v3 API. However with the v2 API it seems possible, albeit in a convoluted way. e.g. with this URL

'https://xxxx.yyyyy.net/rest/api/2/issue/zzz-9?fields=comment,attachment'

I get the following (output snipped for brevity)

"fields": {
"attachment": [
{
"self": "https://xxxx.yyyy.net/rest/api/2/attachment/10000",
"id": "10000",
"filename": "Screenshot 2024-02-07 at 9.02.56 PM.png",
...
"comment": {
"comments": [
{
"self": "https://xxxx.yyyy.net/rest/api/2/issue/10030/comment/10017",
...
"body": "!Screenshot 2024-02-07 at 9.02.56 PM.png|width=1248,height=727!\n\ntesting attachment",

 Now you can correlate the content in body with the filename in attachment  and use the url in the self field therein to download the actual attachment.

I know it's not the cleanest approach, but seems like the best we can do with what's available.

Hope that helps!

[Hariharan Iyer]

One other caveat - this would only work if the filenames are unique.