Index | Thread | Search

From:
izzy Meyer <izder456@disroot.org>
Subject:
Re: stumpwm: avoid remove .fasl after udpate
To:
OpenBSD ports <ports@openbsd.org>
Cc:
"Kirill A. Korinsky" <kirill@korins.ky>, Timo Myyra <timo.myyra@bittivirhe.fi>, Klemens Nanni <kn@openbsd.org>
Date:
Wed, 9 Oct 2024 13:20:44 -0500

Download raw body.

Thread
  • izzy Meyer:

    stumpwm: avoid remove .fasl after udpate

  • On Wed, 09 Oct 2024 01:29:41 +0200
    Kirill A. Korinsky <kirill@korins.ky> wrote:
    
    > ports@, izzy, Timo,
    > 
    > I was hunting this issue for quite some time and the last stop was
    > here: https://marc.info/?l=openbsd-ports&m=171368973700549&w=2
    > 
    > The cause of issue is a bug inside stumpwm which I've fixed:
    > https://github.com/stumpwm/stumpwm/pull/1214
    > 
    > So, here the diff which backports the fix to us.
    > 
    > After this fix stumpwm keeps all .fasl inside
    > ~/.cache/common-lisp/sbcl-2.4.9.openbsd.sbcl-2.4.9-bsd-x64
    > where it should be.
    > 
    > The diff:
    > 
    > Index: Makefile
    > ===================================================================
    > RCS file: /cvs/ports/x11/stumpwm/Makefile,v
    > retrieving revision 1.23
    > diff -u -p -r1.23 Makefile
    > --- Makefile	30 Sep 2024 15:12:58 -0000	1.23
    > +++ Makefile	8 Oct 2024 23:23:15 -0000
    > @@ -2,7 +2,7 @@ COMMENT=		easily hackable Common LISP wi
    >  
    >  VERSION=		23.11
    >  DISTNAME=		stumpwm-${VERSION}
    > -REVISION=		7
    > +REVISION=		8
    >  
    >  CATEGORIES=		x11
    >  
    > Index: patches/patch-make-image_lisp_in
    > ===================================================================
    > RCS file: /cvs/ports/x11/stumpwm/patches/patch-make-image_lisp_in,v
    > retrieving revision 1.1
    > diff -u -p -r1.1 patch-make-image_lisp_in
    > --- patches/patch-make-image_lisp_in	18 Feb 2024 07:46:17
    > -0000	1.1 +++ patches/patch-make-image_lisp_in	8 Oct
    > 2024 23:23:15 -0000 @@ -7,3 +7,20 @@ Index: make-image.lisp.in
    >   (in-package #:cl-user)
    >   
    >   (let* ((expected-warnings
    > +@@ -48,6 +50,16 @@
    > +     (uiop:symbol-call '#:asdf '#:register-immutable-system
    > system-name)))
    > + 
    > + (sb-ext:save-lisp-and-die "stumpwm" :toplevel (lambda ()
    > ++                                                ;; stumpwm migth be
    > build in fake enviroment, so use uiop:restore-image ++
    >                                 ;; to compute real uiop:*user-cache*
    > on that user ++
    > (uiop:restore-image) ++
    >  ;; and clean asdf configuration to avoid some cached paths as well
    > ++
    > (asdf:clear-configuration) ++
    >        (asdf:clear-output-translations) ++
    >                     (asdf:initialize-output-translations ++
    >                                       '(:output-translations ++
    >                                             :enable-user-cache ++
    >                                               :ignore-inherited-configuration))
    > +                                                 ;; asdf requires
    > sbcl_home to be set, so set it to the value when the image was built
    > +
    > (alexandria:when-let ((home #.(sb-ext:posix-getenv "SBCL_HOME")))
    > +                                                   (sb-posix:putenv
    > (format nil "SBCL_HOME=~A" home)))
    > 
    > 
    
    Did some testing of this diff on my dev laptop, and things seem to work
    as expected. I agree with Timo here. Lets wait to see what upstream
    says before merging this into the port.
    
    Good work Kirill!
    
    -- 
    iz (they/them)
    
    > i like to say mundane things,
    > there are too many uninteresting things
    > that go unnoticed.
    
    izder456 (dot) neocities (dot) org
    
    
    
  • izzy Meyer:

    stumpwm: avoid remove .fasl after udpate