Webalizer 2.23-08 is out since August 26, 2013 Not a fork like with 2.23-05 please. (a fork should never be called with the same name as original) app-admin/webalizer-2.21.02 ebuild work fine with 2.23-08 without gcc 4.4 patching. Reproducible: Always
Its in the tree. Please test and reopen this bug if there's an issue. The biggest change over 2.21.02 is that I download webalizer-geodb-20140201 rather than webalizer-geodb-latest. It is a QA violation to have the same ebuild install different files over time. I alert the user with elog messages on how they can manually update GeoDB.dat. We can also revbump over time. I'm not sure what you mean by fork. 2.23-05 simply had patrickfrei.ch's xtended patch that someone requested. Thanks for keeping an eye on this for me. I need the help watching upstream, so feel free to pester me if I miss another bump.
(In reply to Anthony Basile from comment #1) > Its in the tree. Please test and reopen this bug if there's an issue. Can't see it for the moment : # emerge --sync -q # equery l -p webalizer * Searching for webalizer ... [-P-] [ -] app-admin/webalizer-2.21.02:0 [-P-] [ ] app-admin/webalizer-2.23.05:0 I'll try ASAP. The reason why I was looking for a new version is that suddenly (but I don't know when) webalizer started to only show 2 last months in index.html. I didn't modify any config file. > I'm not sure what you mean by fork. 2.23-05 simply had patrickfrei.ch's > xtended patch that someone requested. http://patrickfrei.ch/webalizer/index.html "Webalizer Xtended is a fork of Webalizer" first line.
(In reply to Christophe PEREZ from comment #2) > > http://patrickfrei.ch/webalizer/index.html > "Webalizer Xtended is a fork of Webalizer" first line. Is it really a fork or a major patch? We have a request, bug #501616 to reintroduce USE=xtended. I'm not going to argue semantics, but is there a good technical reason why we can't have both webalizer and webalizer-xtends in the same package with USE=xtends as the trigger. I really don't want to add another package to the tree if this is easily fixed by just a USE flag.
Agreed, the USE flag has worked well for a long time.
(In reply to Anthony Basile from comment #3) > (In reply to Christophe PEREZ from comment #2) > > > > http://patrickfrei.ch/webalizer/index.html > > "Webalizer Xtended is a fork of Webalizer" first line. > > Is it really a fork or a major patch? We have a request, bug #501616 to > reintroduce USE=xtended. I'm not going to argue semantics, but is there a > good technical reason why we can't have both webalizer and webalizer-xtends > in the same package with USE=xtends as the trigger. I really don't want to > add another package to the tree if this is easily fixed by just a USE flag. Anthony, I see 2 possible issues with this: 1. Is it just a patch that is kept in sync with upstream releases or do we risk having to "redo" the patch on new releases? 2. Are licenses compatible and are we making it clear to the users what code they get? In my opinion, if the licenses don't match or the code base has diverged to the point users don't see it as just a patch, they should be separate packages. If required and anyone sees a point to it, we can go to the trouble of having a virtual. If we risk having to "redo" the patch for every release or delaying bumps until a new patch is out, I'd consider having them as separate packages, from a maintenance point of view.
(In reply to Jorge Manuel B. S. Vicetto from comment #5) > (In reply to Anthony Basile from comment #3) > > (In reply to Christophe PEREZ from comment #2) > > > > > > http://patrickfrei.ch/webalizer/index.html > > > "Webalizer Xtended is a fork of Webalizer" first line. > > > > Is it really a fork or a major patch? We have a request, bug #501616 to > > reintroduce USE=xtended. I'm not going to argue semantics, but is there a > > good technical reason why we can't have both webalizer and webalizer-xtends > > in the same package with USE=xtends as the trigger. I really don't want to > > add another package to the tree if this is easily fixed by just a USE flag. > > Anthony, > > I see 2 possible issues with this: > 1. Is it just a patch that is kept in sync with upstream releases or do we > risk having to "redo" the patch on new releases? The patch is 6000 lines long. Currently upstream is only targetting 2.23-05 rather than -08, but I have not tried to apply the -05 patch against -08. I am not interested in redoing such a patch. > 2. Are licenses compatible and are we making it clear to the users what code > they get? both are GPL-2 > > In my opinion, if the licenses don't match or the code base has diverged to > the point users don't see it as just a patch, they should be separate > packages. If required and anyone sees a point to it, we can go to the > trouble of having a virtual. If we risk having to "redo" the patch for every > release or delaying bumps until a new patch is out, I'd consider having them > as separate packages, from a maintenance point of view. If we went with a webailizer-xtends, then we could have each block the other. I'm not sure we need a virtual.
Created attachment 398930 [details] Add old xtended USE Lastest version of Xtender is now based on Webalizer 2.23.08 (http://www.patrickfrei.ch/webalizer/changelog/CHANGELOG). So this version can be used with Xtender again. I add the flag to this ebuild.
(In reply to Sébastien P. from comment #7) > Created attachment 398930 [details] > Add old xtended USE > > Lastest version of Xtender is now based on Webalizer 2.23.08 > (http://www.patrickfrei.ch/webalizer/changelog/CHANGELOG). > > So this version can be used with Xtender again. I add the flag to this > ebuild. This should not have been reopened since the bug was to bump to 2.23.08 and that was done long ago. Having said that, I'm going to carry the xtended patch with webalizer. Its convenient and I don't really care about the semantics of wether a 6000 line patch is a fork or no. Open a new bug when there's a bump needed to either webalizer or the xtended patch.
