Expand link to handle aliases. DRY tests mechanics. - #1237
Conversation
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
|
@MilesCranmer would you mind reviewing this. Thanks |
link to handle aliaseslink to handle aliases. DRY tests mechanics.
|
Resolved my comments as I put them into #1242. By the way, will this mess up any of the existing serialized files at all? I'm assuming no but I really don't have a clear enough understanding for the internals to say. |
|
From a design perspective I'm a little worried by how LinkedChannel and AliasChannel seem to be so different. e.g., LinkedChannel looks to be creating shell scripts, whereas AliasChannel is using runtime resolution. I don't see a strong reason for this split. I think they should use a similar mechanism. |
|
To be honest I think the alias solution is better than the existing LinkedChannel solution. The LinkedChannel solution seems to rely on generating shell scripts (?) which seems kinda fragile. |
Maybe if people downgrade after using this new feature? I don't think we need to worry about that, we can try it out in the prerelease. I think your other comment could be a follow on consideration? Are you happy to approve this so we can move ahead? |
|
I think it's probably good to go. I haven't messed around with it locally though which is something I'd like to try. Have you given it some testing in your terminal at all? I think we should try to break things a little |
|
We had a bug and a test gap for a general |
|
@IanButterworth FYI I think both of my PRs #1241 and #1242 got squashed into a single commit without me included as coauthor |
|
Odd. Squash merges usually retain authorship. |
|
I think if you do a |
|
I didn't edit it. Maybe the GitHub default changed it. |
|
Nope the website. I have an enhanced GitHub chrome extension, it could be that. I'll look when at my computer. |

Alternative to #1115
See discussion around #1115 (comment)
Closes #838
Closes #830
Closes #1115
Developed with Claude