Ok, I will create these JIRA tickets. Thank you all.

Just one thing about the versioning question raised by Lucas: if we go for the first option (and it seems that's where we're going), there will be no breaking change. Of course that doesn't mean the subject shouldn't be discussed :)

On Wed, Sep 14, 2016 at 5:42 PM, Stefan Negrea <snegrea@redhat.com> wrote:

On Wed, Sep 14, 2016 at 3:42 AM, Joel Takvorian <jtakvori@redhat.com> wrote:
1st Option: keep the API in the current form and bring all endpoints up to the same level of functionality, that is:
  * Create "@get */raw" that can take ids or tags
  * Make "@post */raw/query" understand tags
  * Create "@post */stats/query" that can take ids or tags

Thanks for the detailed explanation.

Option 1 might be the only way to go. Adding end-points for tags would just increase the API without any benefits. Since the output is identical, tags or ids is just a query parameter. Also, we need to think about backwards compatibility and expectations that we already set. We already have APIs that support both, so we need to continue that path.

For making these changes, I would break your proposal into several PRs and Jira tickets. A single massive PR will make the review process more difficult.

Thank you,
Stefan


_______________________________________________
hawkular-dev mailing list
hawkular-dev@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/hawkular-dev