On Sun, Nov 06, 2005 at 10:37:53AM -0500, Jay Berkenbilt wrote: > Steve Langasek <vorlon@debian.org> wrote: > > Actually, depending on the details, uploading xerces27 during the KDE > > transition might have made things *easier*... oh well. :) > /me experiences brain gears turning > How's that? Because some packages could have switched from depending > upon libraries that were tangled in the transition to libraries that > weren't? Or more fundamentally, because transition xerces27 into > testing doesn't make anything uninstallable since xerces25 and > xerces26 are still there and then moving applications that depend upon > xerces25 of xerces26 over to xerces27 results in the transition of > xerces25 and xerces26 no longer making those applications > uninstallable? Perhaps having uploaded xerces27 and getting gdal and > friends to switch to it would have broken the link between icu/xerces > and unixodbc, thereby allowing them to transition on their own. gdal > wouldn't have been able to go, but it wouldn't have been keeping > xerces out. Is that right, or am I still missing something? Yes, that was my point in a nutshell. I didn't do any analysis to see *which* packages it would have been better to rebuild against xerces27 first in such a case, but splitting the unixodbc-using packages from the non-odbc packages probably would've been the way to go. Cheers, -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. vorlon@debian.org http://www.debian.org/
Attachment:
signature.asc
Description: Digital signature