Closed anngvu closed 1 year ago
I made some comments and suggested changes, but this is looking great!
I have a broader question: if a user has a very nested file structure, e.g. for a sequencing experiment, two files per sample, but one sample for folder, will they still get a large list? Is there any way to "crawl" up to the highest common folder for which there are no annotations in on any of the files (recursively) within that folder, to ensure we are providing them with the most concise summary possible?
So I get what you mean about the nested folders, but I think it would be time-consuming enough that it's probably not worth it. There would be lots of extra queries going up the tree and trying to get the best summary common ancestor folder, and we're still trying to limit this to under 10 minutes.
So I get what you mean about the nested folders, but I think it would be time-consuming enough that it's probably not worth it. There would be lots of extra queries going up the tree and trying to get the best summary common ancestor folder, and we're still trying to limit this to under 10 minutes.
Yeah, that's kind of suspected would be the issue. Let's give this a shot and see how it goes!
Yeah, that's kind of suspected would be the issue. Let's give this a shot and see how it goes!
@allaway I think this is low-priority development feature because of the likely poor tradeoff, so I won't spend time testing it in this update, but you can put in an official enhancement issue.
Close issue #51. Did quite a bit of refactoring to make things cleaner and faster.
@cconrad8 You're added for more optional ui-type review (sent an example email), so you can validate how the new message template has been heavily influenced by your suggestions.