Bug 173301
Summary: | ddskk needs apel 10.6 to build | ||
---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Jens Petersen <petersen> |
Component: | xemacs-sumo | Assignee: | Ville Skyttä <scop> |
Status: | CLOSED ERRATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | rawhide | CC: | extras-qa |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | 20051208-1 | Doc Type: | Bug Fix |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2005-12-23 21:25:57 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
Jens Petersen
2005-11-16 02:51:51 UTC
Ok, I've given this a bit of thought, and the two feasible alternative solutions I can come up with are: 1) Update the apel contained in xemacs-sumo to 10.6 (or later). 2) Update/patch only bits of apel contained in xemacs-sumo so that it works with ddskk and wl, and leave it at 10.2 and as is otherwise. If the number of packages in xemacs-sumo that depend on apel would be fewer than currently, I'd gladly remove it and use an external apel. But there are quite a few packages that depend/build-depend on it, and besides, apel itself build-depends on other things in xemacs-sumo -> we have a chicken-egg problem. In case I'm doing the work, I'm pretty strongly leaning towards option 2) above; I guess that would be fairly easily acceptable upstream too. Your thoughts? I forgot now why the newer apel was removed from the Fedora xemacs-sumo again... but naively (1) sounds like less work than (2). What advantage does (2) have? But I'm pragmatic about this, as long as wl and ddskk, etc build and work ok with xemacs-sumo I mind less how it is done. The version of apel distributed with xemacs-sumo is 10.2 plus some 15 changes committed by XEmacs folks after it, and it's possible that it was a patched one already when it was synced with upstream 10.2 5+ years ago. When done properly (ie. not just dumping 10.6 in the tree, ignoring all XEmacs-specific patches, but finding out what the changes are, and if they're still needed for something in xemacs-sumo or XEmacs in general), I think 1) could actually be more work than 2). 1) would be more "future proof", though. Anyway, I'll try to find time this week to take a look at 1) first. I have submitted a patch upstream that syncs apel in the XEmacs packages tree to 10.6 for review, and unless there are objections, I intend to commit it early next week. Then, when that patch is in and a new upstream apel package or updated sumos including it are available, it will be included in the next xemacs-sumo packages in FE. New 20051208 sumo including bundled apel 10.6 is now in CVS/devel. No builds yet, because apel 10.6 triggers a bug in latin-unity for which I believe an upstream fix will follow soon. If you have the time, you can build the new xemacs-sumo from CVS locally already now and test if wl and ddskk build and work with it. If you do so, be sure to post experiences here. Thanks, Ville: I tested building ddskk with apel-1.32-pkg.tar.gz in my .xemacs and it seems to be fine - so looks good to me. :) Thanks for testing. I heard that Santa has just delivered xemacs-sumo-20051208-1 to Extras FC-4 and devel ;) |