james hadfield

It's possible for root nodes to belong to a clade (see #1199 for some more context). If the root node doesn't have mutations explicitly set then you can supply the...

I guess we could also implement some syntax (in the clades TSV) instructing it to simply label the root with this clade. This somewhat breaks the approach of `augur clades`...

Wow, I have almost zero memory of writing these tests. It does cover some general entry points to the pipeline which are good to include in CI, so I'd vote...

> Since 2018, I've wanted to integrate env var management for nextstrain build via first-class env dir support (so we don't have to do this hacky thing) Beyond the immediate...

Bumping this as it's happened twice now with LIMC collaborators. On [a related slack thread](https://bedfordlab.slack.com/archives/C01LCTT7JNN/p1719286802460679) there's a few votes to stop our CLI rejecting iat values from the future. @tsibley...

> I think real users will only ever have access to one user pool, so this seems developer-facing. In this case it'd be helpful to not only provide instructions to...

(Three years later) this would still be incredibly useful. I'm currently uploading data for a job and have been looking at the same line for an hour now... ``` zipping:...

> Require --cpus for the AWS Batch runner. Downside is this is a breaking change and erodes the runtime portability properties of nextstrain build's interface. This is my preference

I imagine we could reuse a lot of the code behind the `--download` functionality here ```py if supplied_exclude_patterns: force_exclude = glob_matcher(patterns) else: force_exclude = lambda path: False hardcoded_excluded = path_matcher([...

> but results are often useful for debugging or quickly rerunning a rule locally without having to rerun time-intensive steps Yup, I agree, but I think we're thinking of slightly...