Pulling a link from the body of a Yammer post

[Tono_Analyst2]

Hi,

There is a Yammer post created each week with a web link in the body. What would be the best way to pull this link to parse it in a following step.

Thanks

[Expiscornovus]

Hi @Tono_Analyst2,

When it is an attachment to the yammer post you could a HTTP request to retrieve the file and recreate it as a file in SharePoint, so you easily reuse it in a News Post.

Below is an example of that approach.

1. The HTTP action uses the web_url of the first attachment and tries to retrieve the content.

triggerOutputs()?['body/attachments'][0]['web_url']

2. The Create File uses the body of the response of the HTTP action. I also use the expression below to reuse the same file name.

Interestingly enough I didn't see a property for that in the json so I had to use an expression to refactor it 😅

substring(triggerOutputs()?['body/attachments'][0]['web_url'], add(lastIndexOf(triggerOutputs()?['body/attachments'][0]['web_url'], '/'),1), sub(length(triggerOutputs()?['body/attachments'][0]['web_url']),add(lastIndexOf(triggerOutputs()?['body/attachments'][0]['web_url'], '/'),1)))

[Tono_Analyst2]

Hi again @Expiscornovus 🙂 

Unfortunately, I cannot use premium connectors with my licensing (sorry - I should have mentioned this).

Is there another solution that you know of by any chance? If it helps the structure of the post would look something like.

____________________________________________________________________________________________________

Yammmer Post Title abc

Link https://.....

Maybe more words at the end.... 

____________________________________________________________________________________________________

[Tono_Analyst2]

I have managed to publish a random image using the image link as follows. 

However, I am still stuck on how to extract the image url from the yammer post.

[Expiscornovus]

Hi @Tono_Analyst2,

In your setup you could probably use the parsed body and use a couple of split functions in an expression to retrieve the hyperlink.

Below is an example expression:

if(not(equals(indexOf(triggerOutputs()?['body/body/parsed'], 'https://'), -1)), concat('https://', split(split(triggerOutputs()?['body/body/parsed'], 'https://')[1], decodeUriComponent('%0A'))[0]), null)

Btw, as far as I am aware the Reposts endpoint is to create News Links and not News Posts. That is a different approach, and a News Link is a bit more limited with what you can set. However, if you are happy with that we can obviously stick with News Links instead 😁

The reason why I had to recreate the file for a News Post was I couldn't use an external file link in the Banner of a News Post. That was why I was downloading and creating it 😁