Closed jpthiery closed 6 years ago
Hi, thanks for the PR. Sorry for the delay, I wasn't available recently. This seems like a great addition, I'll review it this week-end.
Don't worry, it's fine ;) I may want to improve test coverage since existing mock not tests headers which i change. I'm not sure to not breack existing behavior.
I will produce documentation update to explain news entry.
For information, i have successfully produce a a report from Docker image stored in an private ECR
repository this morning using change wrote in this PR.
I may want to improve test coverage since existing mock not tests headers which i change. I'm not sure to not breack existing behavior.
Maybe the default value of token_type should be 'Bearer' instead of 'Basic'.
+1 can't wait to see ECR supported. Is there any type of workaround to make ECR work with current version while I'm waiting on this PR to work through the process?
@yebinama Are you going to cut a new revision for release 3.0 for pip installation? Or are you waiting for documentation before doing that?
@dustinvanbuskirk I'll do a new release probably ths week-end.
Yes, i may proposed a documentation PR to allow you to adopt this feature. I try to done it this week.
I've added some documentation to release the 3.1.0 version saturday.
Been working on a custom Docker image to support registry scans as part of CI pipeline.
Was waiting on the release to put in my own PR.
Thank you both for your contributions.
Since some Docker registry like
AWS ECR
not provide a Token URL, we can push toclair
some layer to analyse from the private registry.In order to allow
paclair
to retrive private docker image layer fromAWS ECR
, this Pull Request propose to allow to define the ̀token` using conf.In a our case, we may use the
AWS CLI
to fetch this token, then inject it topaclair
.