Package Details: redeclipse 1.5.3-1

Git Clone URL: https://aur.archlinux.org/redeclipse.git (read-only)
Package Base: redeclipse
Description: free, casual arena shooter
Upstream URL: http://redeclipse.net
Licenses: custom, MIT, CCPL:by-sa, zlib, CCPL:by, CC0
Submitter: mtorromeo
Maintainer: carlwgeorge
Last Packager: carlwgeorge
Votes: 81
Popularity: 0.091384
First Submitted: 2011-03-15 22:28
Last Updated: 2015-07-20 02:49

Latest Comments

aexoxea commented on 2015-09-19 03:05

Hi @carlgeorge,

Ta :-)

As promised, I adjusted the '-git' sample to pull to that commit (11e754a) -- and then to the preceeding one (3e3eff5) -- in base repo. Both built and installed OK, but sadly I couldn't replicate the v1.5.3 segfaults in either, so that doesn't seem to be it.

Further, there are 262 commits across 127 files in base repo before we get to those; as noted before, I'm not in a position to go bisecting to find the point where it got fixed (and even if it is identified, it might well require a number of patches across a number of files -- I too don't have the requisite C/C++ knowledge to be able to tell, sorry).

In the meantime, I've encountered no crashes or errors in f7f2d38 (main issue is it being marked as incompatible with most servers, if you play online). I've noted the commits under the BBS thread to replicate the build I have for anyone that wants to use it.

Not sure what else to do at this point.

carlwgeorge commented on 2015-09-13 21:54

@aexoxea I tried to add that commit as a patch to 1.5.3, but it failed to apply cleanly. I don't know enough C++ grasp what that commit is changing.

aexoxea commented on 2015-09-12 12:13

I too had segfaults when trying to run the package, so as a test, I rigged up a sample 'redeclipse-git' package locally to see if the issue had been resolved upstream. At the time, 'f7f2d38' was the latest commit to red-eclipse/base.

The good news is the package built, installed and ran at that commit with no difficulties, so that provides another workaround (with the bonus of getting new/changed features since the tagged v1.5.3).

The bad news is that commit almost certainly had nothing to do with resolving the issue, and there are 382 preceeding commits across 134 files. It _might_ be commit '11e754a' as it mentions "segfault fixes" (and I'll check this if I get a chance before anyone else does), but otherwise a bisect would be needed (which I probably won't get the chance to do).

The package I built identifies as v1.5.4, though whether that means such a release will be tagged (and how soon) is unclear; I couldn't find anything that told either way on the Red Eclipse website.

Details of the sample 'redeclipse-git' files are available on BBS: < https://bbs.archlinux.org/viewtopic.php?id=202355 >.

Ulukai commented on 2015-08-28 15:53

Hello, I tried building it on 3 machines the past few days, and always get the same error:

libpng warning: Interlace handling should be turned on when using png_read_image
Fatal signal 11 (Segmentation Violation)

I'm running libpng 1.6.18-1.
What do I need to do to fix this?

Sibbe11 commented on 2015-07-23 12:06

Okay, thank you for this fast answer.

In the meantime i have written a much shorter PKGBUILD, which uses the official release package and not the git repositories. You can find it in this link:

https://www.dropbox.com/s/r4mo98am4f8ddfm/PKGBUILD?dl=0

I hope you like it ;-).

carlwgeorge commented on 2015-07-22 22:46

I can reproduce the error. The current version of libpng in Arch is 1.6.16-1. 1.6.17 is available upstream. I'm going to wait to investigate further and see if this fixes itself once the libpng package is updated to match upstream, as there is a decent chance this will resolve the issue.

Sibbe11 commented on 2015-07-22 18:32

The Package is not working. Everytime I start Red Eclipse I get the following:

"Fatal signal 11 (Segmentation Violation)"

I use the newest Arch Packages and the Nvidia Graphics Driver.

If I Download the Package from the Website and launch Red Eclipse after the Decompression of that Package, it works.

Any Ideas?

aexoxea commented on 2015-04-06 06:45

Fair enough; I appreciate what efforts you have been able to make :-)

As it is, I have tested 1.5.1-2 and can confirm the package can now be built, installed and run successfully (at my end at least) without any changes required, so it looks like all is now fixed (and I've marked the BBS thread as 'solved'). Much appreciated!

aexoxea commented on 2015-04-06 06:44

Fair enough; I appreciate what efforts you have been able to make :-)

As it is, I have tested 1.5.1-2 and can confirm the package can now be built, installed and run successfully (at my end at least) without any changes required, so it looks like all is now fixed. Much appreciated!

carlwgeorge commented on 2015-04-06 01:57

I've actually been traveling, so I haven't had a chance to work on this much or help upstream. I did get a moment last night to bump the version to 1.5.1. I just submitted 1.5.1-2 that includes a patch of the commit you mentioned had the fix.

All comments