Has anyone used Microsoft Flow HTTP connector to call Confluence REST APIs?

Byron Stoker

I have a MS Flow setup to detect changes in Azure Dev Ops.  When triggered I want to copy the content from the description of the work item and create a new page in Confluence containing the same content.  

I am using the HTTP connector and passing the most simple JSON to demonstrate the idea.

Headers

{ "Content-Type": "application/json", "Accept": "application/json"}

Body

{ "type": "page", "title": "New Page Test6", "space": { "key": "TS" }, "body": { "storage": { "value": "<p>This is <br/> a new page</p>", "representation": "storage" } }}

 

When the flow is triggered I get a success code (i.e. Status = 200) even though Confluence did not actually create the page.

When I run the exact same JSON structure as CURL or through our Confluence API plug (../restbrowser#/resource/api-content/POST) it works just fine.

 

Looking for some guidance on what could possibly be the difference between the two approaches.

Thanks

Chii

Can you find and put here the full request being sent by the Microsoft Flow Http Connector (e.g., include all the headers, and the method and body etc)? There must be something different between what is being sent by CURL and the Flow Http Connector.

Also, how is the authentication done? The REST api browser in the administration uses the cookies in the current user's browser. I expect you will need to use BASIC auth instead when making the REST api call from Microsoft Flow Http Connector.