Re: PR/51485 CVS commit: pkgsrc/www/seamonkey

[Thomas Klausner <wiz%NetBSD.org@localhost>]

The following reply was made to PR pkg/51485; it has been noted by GNATS.

From: Thomas Klausner <wiz%NetBSD.org@localhost>
To: Joerg Sonnenberger <joerg%bec.de@localhost>
Cc: NetBSD bugtracking <gnats-bugs%NetBSD.org@localhost>
Subject: Re: PR/51485 CVS commit: pkgsrc/www/seamonkey
Date: Mon, 19 Sep 2016 22:37:01 +0200

 On Mon, Sep 19, 2016 at 10:17:49PM +0200, Joerg Sonnenberger wrote:
 > On Mon, Sep 19, 2016 at 10:05:48PM +0200, Thomas Klausner wrote:
 > > On Mon, Sep 19, 2016 at 06:10:12PM +0200, Joerg Sonnenberger wrote:
 > > > On Mon, Sep 19, 2016 at 05:26:34PM +0200, Thomas Klausner wrote:
 > > > > On Mon, Sep 19, 2016 at 05:00:54PM +0200, Joerg Sonnenberger wrote:
 > > > > > On Sun, Sep 18, 2016 at 11:20:00AM +0000, Thomas Klausner wrote:
 > > > > > > The following reply was made to PR pkg/51485; it has been noted by GNATS.
 > > > > > > 
 > > > > > > From: "Thomas Klausner" <wiz%netbsd.org@localhost>
 > > > > > > To: gnats-bugs%gnats.NetBSD.org@localhost
 > > > > > > Cc: 
 > > > > > > Subject: PR/51485 CVS commit: pkgsrc/www/seamonkey
 > > > > > > Date: Sun, 18 Sep 2016 11:09:42 +0000
 > > > > > > 
 > > > > > >  Module Name:	pkgsrc
 > > > > > >  Committed By:	wiz
 > > > > > >  Date:		Sun Sep 18 11:09:42 UTC 2016
 > > > > > >  
 > > > > > >  Modified Files:
 > > > > > >  	pkgsrc/www/seamonkey: distinfo
 > > > > > >  Removed Files:
 > > > > > >  	pkgsrc/www/seamonkey/patches: patch-mozilla_gfx_ycbcr_yuv__row__arm.S
 > > > > > >  
 > > > > > >  Log Message:
 > > > > > >  Remove patch that adds a .s file as .S file.
 > > > > > >  
 > > > > > >  I can't believe both copies are needed.
 > > > > > >  
 > > > > > >  Addresses "pkgsrc cannot be checked out on a case-insensitive file system"
 > > > > > >  by Jonathan Schleifer, PR 51485.
 > > > > > 
 > > > > > This is wrong. The .s version doesn't work as is because it won't get
 > > > > > preprocessed first.
 > > > > 
 > > > > So we only need the .S patch and can remove the .s one?
 > > > 
 > > > Well, the .S patch is opaque and almost impossible to understand...
 > > 
 > > I think the .S patch is just a copy of the patched .s file.
 > 
 > Exactly. So it is impossible to determine from the copying patch, what
 > actually changed in the file. As such, it can't be maintained properly
 > after updates.
 
 That's true. But even if I add back the patch under a non-conflicting
 name, the package still won't build on a case-insensitive file system.
 It just fixes the checkout.
 
 What's your suggested solution?
  Thomas