screen-devel
[Top][All Lists]
Advanced

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

Re: [screen-devel] GNU Screen v.4.2.0


From: Axel Beckert
Subject: Re: [screen-devel] GNU Screen v.4.2.0
Date: Fri, 25 Apr 2014 20:09:03 +0200
User-agent: Mutt/1.5.20 (2009-06-14)

Hi Amadeusz,

On Fri, Apr 25, 2014 at 03:09:30PM +0200, Amadeusz Sławiński wrote:
> > A less comfortable variant is at
> > http://anonscm.debian.org/gitweb/?p=collab-maint/screen.git;a=tree;f=debian/patches
> > 
> > Most of the patches should be suitable for upstream inclusion.
[...]
> I've went through debian patches and I've applied most of them.

Yay!

With regards to 52fix_screen_utf8_nfd.patch: I forgot to mention that
we disabled that patch(*), as it introduced a regression.

(*) It's commented out in quilt's series file -- which is not obvious,
    I know. The view in the patch tracker would have not shown it
    either...

Changelog entry where I disabled it:

  * Disable patch 52fix_screen_utf8_nfd.patch since it introduced a
    regression (Closes: #677512, Reopens: #600246)

According Debian bug reports:
https://bugs.debian.org/600246 (inclusion request)
https://bugs.debian.org/677512 (regression)

Since I've found no way to fix the initial bug report without the
regression, #600246 is still open.

You may want to have a look at the regression and maybe revert that
commit. Maybe you can also know how to fix it without the according
regression...

Sorry for that noise. I should have taken more time and commenting on
each patch individually... Well, better late than never.

> Here are the ones I've not applied with comments.
> 
> 11replace_doc_paths.patch - seems debian specific

Indeed. Should have mentioned that, too.

> 13split_info_files.patch - also debian specific?

Not sure. While I had to touch it last year (else the build failed
with newer texinfo versions), it was added by my predecessor.

It was introduced in 2007 with the following changelog entry:

  * Fix the commands in doc/Makefile.in to generate splitted info files
    (because that's how they are distributed by upstream).

I think, Jan Christoph Nordholz refers to these files in the Screen
4.0.3 tar ball:

-rw-r--r-- mls/suse      46865 2003-12-05 14:52 screen-4.0.3/doc/screen.info-1
-rw-r--r-- mls/suse      46574 2003-12-05 14:52 screen-4.0.3/doc/screen.info-2
-rw-r--r-- mls/suse      49249 2003-12-05 14:52 screen-4.0.3/doc/screen.info-3
-rw-r--r-- mls/suse      50393 2003-12-05 14:52 screen-4.0.3/doc/screen.info-4
-rw-r--r-- mls/suse      25534 2003-12-05 14:52 screen-4.0.3/doc/screen.info-5

Since the package rebuilds the docs, he wanted to regenerated them
(with the patched paths, see above) in a similar fashion.

But at least with 4.2.0, the pre-generated info files seem to be no
more part of the tar ball anyway.

As the patched caused hassles with newer texinfo versions in the past,
I think yoy can ignore it. I'll also check if I can remove it safely
from the Debian package as the reason for it vanished.

> 26source_encoding.patch - cosmetic change for source code

Yes, cosmetic. Added before my time as maintainer of Debian's screen
package. The according changelog entry from 2007 is:

  * Change the comments in the digraph table to give a description of the
    composed char, which retains its informational value regardless of the
    viewer's encoding. Closes: #437024. Thanks, Ben Finney!

According bug report in Debian with some more discussion:
https://bugs.debian.org/437024

> 45suppress_remap.patch - I've checked the logs and it seems like there
> was already a revert of revert... are you sure it doesn't break
> anything?

Also added before my time as maintainer of Debian's screen package.
According changelog entry from 2008:

  * Unbreak <End> and <^A Bksp> using a patch from Loïc Minier
    (thanks!) - introduced as new 45suppress_remap.dpatch.

According Debian bug report: https://bugs.debian.org/484647

> 47screen-cc.patch - I will first check patches from Fedora, instead of
> monolithic one ;)

Sounds like a good plan. :-)

> 48screen-ipv6.patch - it's a feature -> v5
> 58-show-encoding-hardstatus.patch - feature -> v5

Fine for me. :-)

> 49long-usernames.patch - already fixed
> 50increase-max-TERM-length.patch - already fixed

Yes, thanks!

> 60-644788-screen-4.1.0-4.0.3-interoperability.patch - still not sure
> about this one

Not by me, but I was very happy about someome wrote that for me. I
though don't really understand some parts of it, so not much
explanation from expect that it fixed most of the interoperability
issues. (Resizing the terminal while being connected to a 4.0.3 server
does not work properly with that patch. Detaching and reattaching
helps, though, so it was considered a minor issue as it is only
present during an upgrade.)

I actually thought about dropping that patch since it was primaryly
for the upgrade from Debian 6 to Debian 7 while we are now working
towards Debian 8. But since you explicitly mentioned the issues it
mitigates, I thought it may be still interesting for you. 

So feel free to skip it.

> 80EXP_session_creation_time.patch - not applied -> v5

Great!

And again: Thanks for reviewing the patches and including them were
appropriate!

                Kind regards, Axel
-- 
/~\  Plain Text Ribbon Campaign                   | Axel Beckert
\ /  Say No to HTML in E-Mail and News            | address@hidden  (Mail)
 X   See http://www.nonhtmlmail.org/campaign.html | address@hidden (Mail+Jabber)
/ \  I love long mails: http://email.is-not-s.ms/ | http://noone.org/abe/ (Web)

Attachment: signature.asc
Description: Digital signature


reply via email to

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