[FFe] sync gtk-gnutella 0.96.8-1 from debian unstable

[Daniel Letzeisen]

Binary package hint: gtk-gnutella

The current version of gtk-gnutella (0.96.6+dfsg-2) found in the Lucid and Debian sid repositories is rapidly approaching the point where users will be greeted with a flashing *** RUNNING AN OLD VERSION *** message when they start gtk-gnutella. gtk-gnutella 0.96.7 was recently released and fixes bugs that "severely impact" the health of the gnutella network. I know it is late in the Lucid development cycle, but it would be nice if the new version could find its way into Lucid (and sid).

Thanks

http://gtk-gnutella.sourceforge.net/en/?page=news

[Daniel Letzeisen (dtl131)]

I contacted the Debian maintainer and he plans to have 0.96.7 out soon.

"An upload is planned in a bunch of days. Previous version will expire
on 31st, latest one should land in unstable well before that time."

[Mtt.Castelli (mtt.castelli)]

21 March 2010, Version 0.96.8 Released

http://gtk-gnutella.sourceforge.net/en/?page=news

By the way, Joel Oliver mantain updated packages for a wide range of Ubuntu releases, here:
http://stinker.serveftp.net:8000/gtkgnutella/gtkgnutella.html
You could maybe contact him to make this work as a launchpad subscribers. If he wants...

regards.

[bbordwell (benbordwell)]

I am now getting the ancient version detected message, I built a .deb from source of version 0.96.8 and It is working great.

[Micah Gersten (micahg)]

Non-bug fix changes from 0.96.6:

21 March 2010, Version 0.96.8 Released

Version 0.96.8 is a stable release, containing critical bug fixes for the previous release 0.96.7

Under the Hood

    * Fixed Makefile to ensure proper compilation under Gentoo's build scripts.
    * Push-proxy client support now makes sure there is a space after a leading GET or HEAD word in the HTTP request.
    * The DIME parsing layer now given more information in case of an error.
    * Let THEX downloads peruse any supplied Content-Length information from the server (applicable only when the output is not chunked).
    * The VMM layer is now smarter at identifying foreign pages when allocating blocks larger than a single page.
    * When sending BYE messages, do not shutdown the socket layer if using TLS.
    * Compilation cleanup for SunOS, Sun Studio, and versions of GTK < 2.5.0.

07 March 2010, Version 0.96.7 Released

Improvements

    * Better resuming of downloads on session restarts.
    * Increased success rate on downloads from firewalled hosts.
    * Completed support for DHT publishing: DHT is now fully functional.
    * DHT bandwidth usage is now disaplayed in the GUI.
    * Added support for GGEP XQ (eXtended Query) as used by LimeWire for search terms longer than about 30 characters.
    * Try to decipher incorrectly encoded query flags from buggy Shareaza.
    * Parse incorrectly formatted X-Alt header from buggy Shareaza.
    * Implemented PFSP 1.0a: now emits and handles new X-Available header.
    * Added some MIME types and file extension mappings.
    * Added comand-line switch --compile-info to print some information that is hardcoded at compile-time.
    * Persist known TTH roots in magnets.
    * Swapped leaf and ultrapeer slots for GGEP UP in compliance with LimeWire which does not follow the published specification (so we don't either now but we are compliant with the de-facto standard).
    * Will now remove uncompressed ultra connections when full and getting a new compressed one.
    * Whitelist can now support TLS hosts and will refresh DNS mappings regularily.
    * Improved granularity of the background task used to scan the library in order to prevent noticeable processing freezes as much as possible.
    * Redesigned the output of the shell "status" command to show more information and display status flags (shown as icons in the GUI).
    * Added liberal parsing of urn:sha1: and urn:bitprint to cope with clueless sites using hexadecimal instead of base32 to encode the hashes.
    * Added -v option to the shell command "props" to also display values along.
    * Added real shell-like string parsing, and fixed escaping to work correctly.
    * Added the ';' separator to be used to separate mutiple shell commands.
    * Allow crashes and fast assertions "soft" failures to be timestamped.
    * Don't answer to UDP pings from bad nodes, which include "alien" hosts.
    * Loop through the TCP push-proxies more quickly until one can relay our push.
    * Passively revitalize PUSH routes and wakeup starving downloads.
    * Attempt to collect push-proxies passively from relayed query hits.
    * A completed file waiting for being moved a...

[StefanPotyra (sistpoty)]

ACK, FFe granted.

[Michael Bienia (geser)]

gtk-gnutella 0.96.8-1 got autosynced to maverick.

[Daniel Letzeisen (dtl131)]

Michael, this bug is for Lucid, where the Ubuntu repo version still reports "ancient version". Lucid still has 0.96.6, so this is still a bug.

[Steve Langasek (vorlon)]

Bugs for lucid are filed against "(Ubuntu Lucid)". This bug is filed against "(Ubuntu)", which means it tracks the current development release, where gtk-gnutella is current.

And we don't sync new upstream versions of software into released versions of Ubuntu. This bug looks as resolved as it's going to get.

[Daniel Letzeisen (dtl131)]

When I filed the bug, it was for Lucid development branch. The problem remains in Lucid, and an FFe for Lucid was granted in comment #5.

So do I have to re-file the bug or can I switch it to Lucid in some way?

[Steve Langasek (vorlon)]

I've opened a task against the Lucid version of the package. Please follow <https://wiki.ubuntu.com/StableReleaseUpdates#Procedure> to propose an update to lucid.

[Daniel Letzeisen (dtl131)]

Forget it. It is more practical to get updated .debs from: http://stinker.serveftp.net:8000/gtkgnutella/gtkgnutella.html
Marking invalid.

[Jim Kelly (kejianshi)]

Usually I feel that the Ubuntu guys are all over it. This isn't one of those times. This application is often used. It makes sense to me the the LTS Lucid would include this as an update, same as they would many others.
I think peer-to-peer file sharing probably ranks darn high on the "often used" list of features"
Do we really want Ubuntu users looking at a "Ancient version" message and having a version straight out of the Ubuntu store in the most recent LTS that won't connect without alot of coaxing because it wasn't update for 2 years running?
Seriously - I think a yearly update to the repo for the LTS is in order for this application.

[Micah Gersten (micahg)]

You're welcome to follow the Stable Release Update procedure as Steve
Langasek suggested. Someone else came along and marked the task
Invalid. If you feel it should be updated, by all means complete the
process for the update. If the upstream and Debian updates were only
bug fixes, then you have a good chance of getting in a stable release
update. If there were features, you should instead file for a backport:
https://help.ubuntu.com/community/UbuntuBackports

On 02/02/2011 05:45 PM, Jim Kelly wrote:
> Usually I feel that the Ubuntu guys are all over it. This isn't one of those times. This application is often used. It makes sense to me the the LTS Lucid would include this as an update, same as they would many others.
> I think peer-to-peer file sharing probably ranks darn high on the "often used" list of features"
> Do we really want Ubuntu users looking at a "Ancient version" message and having a version straight out of the Ubuntu store in the most recent LTS that won't connect without alot of coaxing because it wasn't update for 2 years running?
> Seriously - I think a yearly update to the repo for the LTS is in order for this application.
>

[Jim Kelly (kejianshi)]

OK - Will do. Actually, I'm fine. Already updated. I'm abit on the computer literate side (-; Its the not so computer savy that make up the other 99% of the world I'm concerned with. Good working and easily updated often used features go a long way towards making Ubuntu the OS of choice for the masses. GTK-Gnutella definitely qualifies, probably more so than most of the other included features of Ubuntu.

The stable Release Update procedure you mentioned works for me, you and possibly a few of my friends, which is great for fringe power-user stuff, but this function is wildly popular with the masses. I think we all want it to be 1 button update friendly, for the masses to get the latest stable release that will actually connect. That is what makes an OS popular.
Ease of use.

Take my dad, brother and almost everyone I know. They could care less about SSH, SSL and all the things we pay attention too. However, if gnutella updates are available, they will want it ASAP. Really - Thats most people. I don't want to explain to my dad that Ubuntu is the greatest thing since sliced bread... If you know exactly what you are doing. (-:

As for me personaly, I'm already taken care of.

Thanks for your time - I LOVE Ubuntu by the way. I convert as many users as I can to it.

[Micah Gersten (micahg)]

Getting this in via a Stable Release Update makes it automatic for most
users that have lucid-updates enabled.

On 02/02/2011 06:23 PM, Jim Kelly wrote:
> <snip />
> The stable Release Update procedure you mentioned works for me, you and possibly a few of my friends, which is great for fringe power-user stuff, but this function is wildly popular with the masses. I think we all want it to be 1 button update friendly, for the masses to get the latest stable release that will actually connect. That is what makes an OS popular.
> Ease of use.
> <snip />