On Sun, Sep 28, 2008 at 8:38 AM, David Relson <[EMAIL PROTECTED]> wrote: > On Sat, 27 Sep 2008 12:58:40 -0400 > Kirk Lowery wrote: > >> On Fri, Sep 26, 2008 at 2:47 AM, Neil Bothwick <[EMAIL PROTECTED]> >> wrote: >> > On Thu, 25 Sep 2008 17:59:35 -0400, Kirk Lowery wrote: >> > >> >> During upgrading today, I inadvertently allowed 2.3 to be deleted. >> >> I looked in portage, but it is apparently gone. >> > >> > Nothing is ever truly gone from portage >> > >> > http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/python/?hideattic=0 >> > >> > goes back as far as python-1/5/2 >> >> Thanks for the answers. I thought I'd report my experience here for >> the knowledgebase: >> >> I went back and downloaded not only the latest 2.3.6 ebuild into my >> local overlay, and all the attendant files (including important >> patches) in the "files" directory. In order to be able to emerge the >> ebuild file, I had to recreate the digest and manifest files. Don't >> know why the checksums were off, but they were. The solution was >> simple: >> >> # ebuild python-2.3.6-r6.ebuild digest >> >> took care of it. When I tried to restart my legacy software I got a >> python error: "ImportError: No module named thread". Sure enough, the >> equery use feature told me "-threads". So I overrode that in >> package.use. Nope, no change. Looking at the compile log, I saw this: >> >> ./configure --prefix=/usr --host=x86_64-pc-linux-gnu >> --mandir=/usr/share/man --infodir=/usr/share/info --datadir=/usr/share >> --sysconfdir=/etc --localstatedir=/var/lib --with-fpectl >> --enable-shared --enable-ipv6 --infodir=${prefix}/share/info >> --mandir=${prefix}/share/man --with-libc= --enable-unicode=ucs4 >> --with-threads --without-threads --with-threads --libdir=/usr/lib64 >> --build=x86_64-pc-linux-gnu >> >> Configure has both --with- and --without-threads!!?? >> >> Nothing I did seemed to make a difference. So I traced where in the >> ebuild this might be coming from: >> >> use threads \ >> && myconf="${myconf} --with-threads" \ >> && myconf="${myconf} --without-threads" >> >> I still don't know why (and would appreciate any knowledgeable person >> commenting here), but I simply commented out the third line above, >> re-emerged, and viola! my legacy software is up and running. >> >> Kirk > > An interesting defect to find, and a good job of detective work! > > Out of curiosity, I looked at the python ebuilds on my workstation. > > Up through python-2.4.4-r6, they have: > > use nothreads \ > && myconf="${myconf} --without-threads" \ > || myconf="${myconf} --with-threads" > > From python-2.4.4-r14 onwards they have: > > use threads \ > && myconf="${myconf} --with-threads" \ > || myconf="${myconf} --without-threads" > > Evidently the 2.3.6-r6 ebuild has the newer "use threads" test, but > with the minor (but fatal) operator flaw you found -- "&&" which > should be "||". > > Hope this is of interest,
Yes, indeed! I need to review my binary operators. ;-) It's too bad this is legacy stuff. We could file a patch and gain cyberspace kudo's and good karma! But I did learn a bit about ebuilds. And -- duh! -- comparing other similar ebuilds is an important strategy in tracing out these kinds of problems. Thanks for looking further into this. Kirk
