+ [2020-07-17T16:34:15Z] cos I guess 'repo' will allow all of the above, and nothing less than 'repo' will suffice for any of the above, but it's not clear.
+ [2020-07-17T16:32:48Z] cos Also, what is the minimum scope needed to be able to read pull requests and their comments?
+ [2020-07-17T16:32:20Z] cos Anyone know, for example, what is the minimal scope to allow a release to be created? To allow assets to be uploaded to a release?
+ [2020-07-17T16:31:57Z] cos Looking at https://developer.github.com/apps/building-oauth-apps/understanding-scopes-for-oauth-apps/ but it isn't specific enough about some things, or doesn't have all the right terms to be searchable.
+ [2020-07-17T16:31:32Z] cos Trying to figure out the minimal set of scopes required for personal access tokens to do certain things.
+ [2020-06-11T01:40:07Z] cos the way I ended up solving it was to copy the files I had edited to a temporary directory, using -D to get rid of the brnach, recreating the branch, and copying my changed files back in and committing.
+ [2020-06-11T01:38:56Z] cos git restore, git reset, and git checkout all failed me in different ways - or I didn't use them right.
+ [2020-06-11T00:57:41Z] cos In case branch-B ends up getting merged to master before branch-A does, I don't want the branch-A merge to delete those files. I want it to not have any effect on those files.
+ [2020-06-11T00:57:07Z] cos I was working in two different branches today. At some point, I did a commit in branch-A that accidentally included a couple of new files meant for branch-B. Later, I commited those same files in branch-B as well. I've done at least one additional commit in both branch-A and branch-B since then. How can I get rid of the extra files from branch-A, not as a file deletion commit, but as if they were never there?
+ [2020-06-03T02:01:37Z] cos Well, I learned a lot from you, thank you!
+ [2020-06-03T01:42:57Z] cos so going by date+timestamp seems best
+ [2020-06-03T01:42:44Z] cos ahh so that commit that was the parent of the commit with the tag. but that might not work if the tag isn't the first commit on its branch.
+ [2020-06-03T01:37:42Z] cos huh, TAGNAME^..HEAD actually goes back to one commit *before* the tag. weird.
+ [2020-06-03T01:36:39Z] cos (well, obviously since I dont want it to be brittle the script will do some error handling on the output of git tags, but that's not something I have any trouble doing)
+ [2020-06-03T01:35:57Z] cos "the" tag is always there, since I can select it from "git tags".
+ [2020-06-03T01:33:55Z] cos I probably do, but just as importantly I want a method of listing commits that is not vulnerable to failing any time the tag isn't there
+ [2020-06-03T01:30:01Z] cos I want something that works regardless of whether the tag is or not, especially since I can't predict, and since we have no clue how this tag ended up not on master.
+ [2020-06-03T00:43:57Z] cos oh :( well there goes that idea. maybe I can look up the datestamp of the tag and maybe there's a way to show all commits newer than that datestamp...
+ [2020-06-03T00:31:29Z] cos But will that not give me all commits to master that are newer than that tag?
+ [2020-06-03T00:31:07Z] cos and I think that is what I want - though git is convoluted enought that if it's not what I want it may take a long time to figure it out :/