Bug 192607 - [patch] sysutils/roottail error during install with NO_INSTALL_MANPAGES
Summary: [patch] sysutils/roottail error during install with NO_INSTALL_MANPAGES
Status: Closed Not Accepted
Alias: None
Product: Ports & Packages
Classification: Unclassified
Component: Individual Port(s) (show other bugs)
Version: Latest
Hardware: Any Any
: --- Affects Some People
Assignee: freebsd-ports-bugs (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-08-12 08:30 UTC by dmitry2004
Modified: 2014-08-12 10:00 UTC (History)
1 user (show)

See Also:


Attachments
Patch (387 bytes, patch)
2014-08-12 08:30 UTC, dmitry2004
no flags Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description dmitry2004 2014-08-12 08:30:02 UTC
Created attachment 145700 [details]
Patch

After migrate to stagedir error during install with NO_INSTALL_MANPAGES
Comment 1 John Marino freebsd_committer freebsd_triage 2014-08-12 08:37:37 UTC
NO_INSTALL_MANPAGES is not a thing.
Man pages, by policy, are always installed.  There is a distinction between DOCS and man pages.

This solution is not valid.

Now, what is the actual problem?
Comment 2 John Marino freebsd_committer freebsd_triage 2014-08-12 08:56:08 UTC
Okay, I talked with antoine and he tells me NO_INSTALL_MANPAGES is valid for imake-based ports.  Let me look at this again.
Comment 3 dmitry2004 2014-08-12 08:59:26 UTC
Many years (~14) i have NO_INSTALL_MANPAGES=yes in make.conf, all works fine. 

Mk/Uses/imake.mk also support NO_INSTALL_MANPAGES, many ports support this option too.

But migrate to stagedir broke separation man pages and other files.
Comment 4 dmitry2004 2014-08-12 09:06:26 UTC
Need an opportunity not install man pages when required (example jailed environment).
Comment 5 John Marino freebsd_committer freebsd_triage 2014-08-12 09:12:32 UTC
Antoine says NO_INSTALL_MANPAGES is only valid for imake when there is no install.man target in the imake file.  This port's imake file has that target, so NO_INSTALL_MANPAGES should not be used.


Are you trying to set NO_INSTALL_MANPAGES in make.conf to prevent man pages from installing?  What's the actual problem this PR is trying to solve?
Comment 6 John Marino freebsd_committer freebsd_triage 2014-08-12 09:13:02 UTC
oops, I mean to post that before your responses.
Comment 7 John Marino freebsd_committer freebsd_triage 2014-08-12 09:15:58 UTC
(In reply to dmitry2004 from comment #3)
> Many years (~14) i have NO_INSTALL_MANPAGES=yes in make.conf, all works
> fine. 


No, this is not fine.  NO_INSTALL_MANPAGES is not a make.conf variable.  You are misusing it.


> 
> Mk/Uses/imake.mk also support NO_INSTALL_MANPAGES, many ports support this
> option too.

Any non-imake port that supports this is incorrect and must be fixed.


> But migrate to stagedir broke separation man pages and other files.


No, it didn't.  Man pages are not separate.


> Need an opportunity not install man pages when required (example jailed environment).

Well, jailed environment isn't a strong example.  Most people use "embedded environment".  There is a request for no-manpage installs.  It's not implemented yet.  This isn't the implementation.
Comment 8 dmitry2004 2014-08-12 09:27:15 UTC
(In reply to John Marino from comment #7)

> > Need an opportunity not install man pages when required (example jailed environment).
> 
> Well, jailed environment isn't a strong example.  Most people use "embedded
> environment".  There is a request for no-manpage installs.  It's not
> implemented yet.

You know anything about the implementation plans?
Comment 9 John Marino freebsd_committer freebsd_triage 2014-08-12 09:38:52 UTC
If I were to speculate, code would be added to bsd.port.mk that would skip installing any file located at MANPREFIX for every port.  It would be controlled by a make.conf-supported variable.
Comment 10 dmitry2004 2014-08-12 09:43:07 UTC
Thanks for your help.
I will wait for a desired function or imlement it and send patch ;-)
Comment 11 John Marino freebsd_committer freebsd_triage 2014-08-12 10:00:59 UTC
(In reply to dmitry2004 from comment #10)
> Thanks for your help.
> I will wait for a desired function or imlement it and send patch ;-)

I think such a patch will be complex as it has to interact with staging functions.  You might want to search bugzilla (ports infrastructure) to see if a PR already exists.  If not, open a PR with a feature request and describe something that that..  At least it will be documented.