@jeff_reynolds It's not just you. You should post this over in the ideas section. I completely agree with your sentiment. I see a couple of key cons to the new cache input option: 1) if you were to add a tool after the input tool and cache it, it would not use the cached input; it would instead re-run the query. I have posted an idea here that would address this. 2) as you pointed out as well, I have some workflows with 10+ inputs where I've checked the cache option. Previously, my first run would take a while but then every run after that would use cached inputs and allow for easier development. Now, it appears that I would have to go through and click cache and run my workflow for every tool every time. I have posted an idea here that may address this.
I feel the same. I thought the new version would be much easier to cache and run from any point. However, removing the cache function on input node is a big mistake. I have to right click cache and run then have to stop the work flow and start over on the next node. Not fun.
Thank you for sharing. I did vote on @patrick_digan's idea submission. Thank you very much.
I agree here. Why would the new Cache and Run necessitate removing the Cache Data option on the Input Tool. I like the new Cache and Run concept, but it is a bit limited. Often times, my workflow has a dozen or more data sources. They also regularly have multiple branches and the Cache and Run only supports a single cache point. I get that caching too much data can cause performance issues, but wouldn't that decision be better left to the analyst?