On Tuesday August 27 2024 05:38:33 Ryan Carsten Schmidt wrote: >No there is no such thing and there should be no need to implement anything >like that since ports should not fail
There are lots of things that shouldn't happen but still do. >and if they do the log can be consulted to discover what happened. The log will only show the immediate reason for a failed build, without any form of explanation. Sometimes the underlying cause is easy to address in a way that relaunching the failed command will result in a successful build/install. (I had an example but already forgot what it was.) Rather than letting the user figure it out or having to file a ticket, why not provide the necessary feedback as a courtesy if you already have such information. Anyway, I'm happy to take suggestions how to implement the feature myself without hacking "base". I already tried renaming the `return` procedure, and `trace add execution return enter <proc>`, both of which cause things to break when I print something in the callback. R
