Closed eurubkov closed 4 years ago
I would say collect all repos from all of the projects for now. We can add support for the extra layer later.
Ok I implemented getting all readmes across the whole organization for azure devops. One thing I don't necessarily like is HttpClient. Different resources give different opinions, but initializing a new HttpClient for each request will exhaust all sockets on the machine. So I moved it to be static. I also had to have two of them because I couldn't figure out how to dynamically change the media type.
In AzureDevOps, the hierarchy is as follows: Organization -> Project -> Repository -> Items. Each organization can have multiple projects, so we need to either let the user pass in the project name/id or get readmes across all projects. That's up for the discussion.
Still need to figure out what to do when a request fails for any reason (log somewhere, notify someone, retry?).
I also changed the GetReadmes method to have the return type of Task.