Pete Gonzalez

> The run is triggered by the SASS task because Heft's file watcher support doesn't actually monitor the output of globs, it monitors all the file system calls made to...

@dmichon-msft when we discussed this, we thought that upgrading the decoupled local dependencies in the Rush Stack repo would eliminate this error. Did that work? (We can look at ways...

We originally standardized on CRLF because it was handled correctly by all editors on all OS's, whereas LF was mishandled by some editors (e.g. Notepad.exe and cmd.exe on Windows). The...

@iclanton @patmill @lahuey If we make this change, it would impact SPFx as well.

Cool, let's do it.

> I'd think the gold standard would be to detect the existing line endings and then use those when we update the file This would produce nondeterministic results, and the...

@iclanton and I discussed this today, and came to these conclusions: - Which behavior is ideal depends on Git's [core.autocrlf](https://git-scm.com/book/en/v2/Customizing-Git-Git-Configuration) setting, but it's impractical for our tools to query that...

Footnote: A possible fourth option would be to set `core.safecrlf=false` which suppresses Git's `CRLF will be replaced by LF` warning. That would technically resolve this GitHub issue, but it seems...

@kevin-y-ang FYI

In [this Zulip thread](https://rushstack.zulipchat.com/#narrow/stream/262521-api-extractor/topic/zulip-requires-a-topic-so-heres-one/near/310657683), @davidmurdoch said that this problem repros for his project when using the latest API Extractor but NOT with v7.20.1. @zelliott do you think it could be...