octave-bug-tracker
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-bug-tracker] [bug #60695] Presence of qtchooser in host system b


From: anonymous
Subject: [Octave-bug-tracker] [bug #60695] Presence of qtchooser in host system breaks build with Qt in another prefix (pkgsrc here)
Date: Sat, 29 May 2021 14:29:57 -0400 (EDT)
User-agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:88.0) Gecko/20100101 Firefox/88.0

URL:
  <https://savannah.gnu.org/bugs/?60695>

                 Summary: Presence of qtchooser in host system breaks build
with Qt in another prefix (pkgsrc here)
                 Project: GNU Octave
            Submitted by: None
            Submitted on: Sat 29 May 2021 06:29:55 PM UTC
                Category: Configuration and Build System
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Build Failure
                  Status: None
             Assigned to: None
         Originator Name: Thomas Orgis
        Originator Email: thomas.orgis@uni-hamburg.de
             Open/Closed: Open
                 Release: 6.2.0
         Discussion Lock: Any
        Operating System: GNU/Linux

    _______________________________________________________

Details:

The configure script assumes that presence of a qtchooser (e.g.
/usr/bin/qtchoooser) means that moc and friends are actually handled by it, by
being symlinks to qtchooser.

This is a wrong assumption if the host is some GNU/Linux distro and you build
Octave with another install of Qt. In my case it is installing it via the
pkgsrc collection. This means that in my case moc is /data/pkg/qt5/bin/moc,
totally unrelated to /usr/bin/moc, which is a symlink to qtchooser.

As pkgsrc does not user the qtchooser machinery, this confused setup leads to
the correct Qt tools being selected, but the builds breaks on the first call
to `moc -qt=5` as this is _not_ as symlink to qtchooser and hence that
argument triggers an error.

The m4 macro should either check if the tools really accept the assumed
MOCFLAGS etc., or test if the located commands are symlinks to the qtchooser
and _then_ assume that -qt=$version is desired.

I hope that the scenario of using a custom prefix for Qt to override the
system on is not too outlandish when we come from the perspective of a user
doing a custom build of Octave already.

My use case of pkgsrc in HPC especially is about installing lots of software
from sources concurrently, to be independent of the distro in such matters
like Qt version.




    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?60695>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]