Jon Parise

> There are domains with underscores in the wild. Based on my understanding of the relevant RFCs, that isn't standards-compliant and would challenge our definition of "valid" without loosening our...

I think if we can define and document the rules for allowing `_` characters, that seems reasonable to me. For a function like this that returns a "valid" boolean response...

I re-implemented this as `HostName.validate` now that `std.Io` has landed. The code and tests are the same as the original version. A few notes: 1. I'm now using `HostName.max_len`, which...

With the latest revision, the pattern now yields correct results for a complete set of possible `history` values: ```bash #!/usr/bin/env bash declare -a lines=( "10000 history 1" " 1000 history...

We could get more precise (and more closely reproduce the `sed` pattern) if we (temporarily) used `extglob`, but that might be overkill for this set of constrained inputs.

> Your script produces a wrong result for the last input: > > ``` > $ bash bash-completion-167.sh > history 1 > history 1 > history 1 > history 1...

> @jparise do we know if this reduces the overhead time of running bash-preexec? Would be curious how much this speeds us up. In this artificial test, the new (non-`sed`)...

As a proof point, [Ghostty 1.1.0](https://ghostty.org/docs/install/release-notes/1-1-0) shipped with this change to its bundled bash-preexec, and we haven't heard of any issues through the pre- or post-release period yet.

> As a proof point, [Ghostty 1.1.0](https://ghostty.org/docs/install/release-notes/1-1-0) shipped with this change to its bundled bash-preexec, and we haven't heard of any issues through the pre- or post-release period yet. @rcaloras...

It looks like some users of this function were expecting the octal behavior.