I don't know, if this is an BUG of the ebuild or something else. I know from my SuSE days, that there's a daemon called 'idled'. net-mail/cyrus-imapd-2.2.8 installs /usr/share/man/man8/idled.8.gz but no /usr/lib/cyrus/idled. strange... what happend? Reproducible: Always Steps to Reproduce:
missing configure option --with-idled=idled
you will fix it? ;-)
Created attachment 39063 [details] Updated ebuild I noticed the same thing today. idled really is a useful feature, and I don't think people should have to resort to editing the ebuild to enable it. In any case, installing the manual page for a feature not installed seems confusing. I made a new version of cyrus-imapd-2.1.15.ebuild, which is what I have installed, just to test out a couple of ideas. If someone could let me know if I'm completely barking up the wrong tree I'd be grateful. With the attached ebuild, if you have "cyrus-imapd" in your USE flags, --with-idle=idled will be passed as a configure option, and if not, --with-idle=poll (the current default) will be given. If the USE flag is not set, the manual page is not installed. If the new USE flag is set, idled is enabled in cyrus.conf, otherwise it is left disabled - I'm not sure the approach I've taken to do this is the right one though.
Created attachment 39064 [details, diff] Patch that enables idled
Comment on attachment 39064 [details, diff] Patch that enables idled cyrus_2-conf-idled.patch
*** Bug 55377 has been marked as a duplicate of this bug. ***
and now? nothing happend? remind, remind, remind. ;-)
there are 3 options for with-idle: idled, poll, and no. The attached ebuild doesn't cover all of them.
I think a USE flag to choose between --with-idle=idled and --with-idle=poll is enough since it is possible to disable IDLE poll support with "imapidlepoll: 0" in imapd.conf. From the manpage: This option is used when idled can not be contacted or when polling is used exclusively. The minimum value is 1. A value of 0 will disable polling (and disable IDLE if polling is the only method available).
swegener, thanks! I'll include this in the next bump.
fixed in 2.2.8-r1