Created attachment 223453 [details] 2.3.0 Attached patch updates caddy to 2.3.0. Also included: - fix PLIST_FILES (should be relative, not absolute) - remove pkg-message with incorrect instructions - remove CGO_ENABLED=0 which appears to no longer be necessary
Hi, I don't use this software anymore, I'd be happy to pass on maintainership. Adam, would you be willing to pick it up? Thanks!
I'd be very happy to! Thank you for all the time you spent on this port.
A commit references this bug: Author: adamw Date: Sun Mar 21 17:27:55 UTC 2021 New revision: 568926 URL: https://svnweb.freebsd.org/changeset/ports/568926 Log: www/caddy: Update to 2.3.0 and take maintainership Thanks to Daniel Tihanyi for all his work maintaining this port! While here, a number of cleanups: - Simplify Makefile - Remove pkg-message whose instructions were superseded by @sample - Re-enable cgo. It seems to work fine now. - Rename the rc(8) 'validate' subcommand to 'configtest' for clarity and consistency - Add a little more info to the sample Caddyfile - Point to instructions for getting access logs working Changes: https://github.com/caddyserver/caddy/releases/tag/v2.3.0 PR: 254436 Approved by: Daniel Tihanyi (outgoing maintainer) Changes: head/www/caddy/Makefile head/www/caddy/distinfo head/www/caddy/files/Caddyfile.sample.in head/www/caddy/files/caddy.in head/www/caddy/files/pkg-message.in head/www/caddy/pkg-descr
Hi Adam, Not sure of your reasons for changing the non-standard validate command to configtest? I feel configtest has less clarity and consistency with Caddy's validate command.
(In reply to Basil Hendroff from comment #4) It does present an inconsistency with caddy's command, but it brings consistency with the majority of FreeBSD rc(8) scripts, which use `configtest' for that purpose. The intent of rc(8) scripts is to present a common, abstracted interface, not to mirror the end binary's command structure. If people want to test the config file, `configtest' is the first command to try. I don't feel super strongly about it, though, and if enough other people indicate a strong preference I'm certainly willing to reconsider.