Re: 0.99 vs redhat
John Berninger <[email protected]>
| Newsgroups | gmane.network.up2date.current.devel |
|---|---|
| Message-ID | <[email protected]> |
Well, unfortunately, yes, since the bug in question is (evidently) with
rpm-python-4.0.2-8. What's really strange is that the labelCompare on
my 7.1 system worked as expected, and gave the correct results both
ways.
Hunter and I have looked at the RPM code in this narrow little
area, and as best we can determine, there is no material difference
between the versionCompare and labelComapre. The specifics are as
follows:
labelCompare takes two tuples as args, each tuple consisting of
(epoch, version, release). First, it looks at the epochs. If one epoch
is not NULL and the other is NULL, the non-NULL is more recent. If
they're both non-NULL, and not equal, the newer is newer. This is done
by converting to ints and doing a '<' or '>' test.
If the epochs are both NULL or are equal, a call to rpmvercmp is
ussues with the versions as arguments. This returns a -1, 0, or 1 to
determine which is equal. If it is -1 or 1, that value is returned, if
0, another call to rpmvercmp() is issued with the releases as arguments.
This return value is returned, whatever value it is.
For some reason, this works for me on a 7.1 platform, but
obviously not for everyone. Since the algorithm is identical for
versionCompare(), except that it takes the header and it extracts the
(epoch, version, release) for you, rather than expecting you to extract
it, there's no reason to switch to versionCompare, as tracking the full
header would be difficult at best. BLOB's are not known for performance
in database functions, which is what the header object would have to be
stored as.
On Fri, 08 Mar 2002, Rob Hagopian wrote:
> Not quite yet :-)
>
> This makes v7.2, and more specificly rpm-python-4.0.3-1.03 a prerequsite for
> current-0.99 and the spec file should be updated to reflect that...
>
> Personally, I'd prefer to see a workaround for 7.1, esp if there are other
> rpm-python calls that would correctly version the dbs...
> -Rob
>
> > -----Original Message-----
> > From: John Berninger [mailto:[email protected]]
> > Sent: Friday, March 08, 2002 11:45 AM
> > To: [email protected]
> > Subject: Re: [Current-server] 0.99 vs redhat
> >
> >
> > Well, with that, I'll assume that the problem was a bug in 7.1's rpm
> > module and was fixed in 7.2. Any objections to closing this
> > issue? :)
> >
> > On Fri, 08 Mar 2002, Rob Hagopian wrote:
> >
> > > [root@abbott i386]# for i in * ; do rpm -qp $i
> > --queryformat "%{NAME}
> > > %{EPOCH} %{VERSION} %{RELEASE}\n"; done
> > > php (none) 3.0.18 1.6.x
> > > php (none) 3.0.18 8
> > > php-imap (none) 3.0.18 1.6.x
> > > php-imap (none) 3.0.18 8
> > > php-ldap (none) 3.0.18 1.6.x
> > > php-ldap (none) 3.0.18 8
> > > php-manual (none) 3.0.18 1.6.x
> > > php-manual (none) 3.0.18 8
> > > php-pgsql (none) 3.0.18 1.6.x
> > > php-pgsql (none) 3.0.18 8
> > > [root@abbott i386]#
> > >
> > > You're right, the server above is v7.1, our other current
> > servers are v7.2,
> > > sorry... But 7.2 does work better (correctly, even):
> > >
> > > [root@x2hvd test]# rpm -q rpm-python
> > > rpm-python-4.0.3-1.03
> > > [root@x2hvd test]# cadmin create dh-test
> > > Mar 03 16:39:13 Adding php-3.0.18-8.i386.rpm
> > > Mar 03 16:39:13 Adding php-imap-3.0.18-8.i386.rpm
> > > Mar 03 16:39:13 Adding php-ldap-3.0.18-8.i386.rpm
> > > Mar 03 16:39:13 Adding php-manual-3.0.18-8.i386.rpm
> > > Mar 03 16:39:14 Adding php-pgsql-3.0.18-8.i386.rpm
> > > Mar 03 16:39:14 Adding php-3.0.18-1.6.x.i386.rpm
> > > Mar 03 16:39:14 type(hit) = <type 'string'>
> > > Mar 03 16:39:14 hit = php:3.0.18:8:
> > > Mar 03 16:39:14 Comparing packages: php-3.0.18-1.6.x and
> > php-3.0.18-8
> > > Mar 03 16:39:14 Package is older than package already in database.
> > > Mar 03 16:39:14 Adding php-imap-3.0.18-1.6.x.i386.rpm
> > > Mar 03 16:39:14 type(hit) = <type 'string'>
> > > Mar 03 16:39:14 hit = php-imap:3.0.18:8:
> > > Mar 03 16:39:14 Comparing packages: php-imap-3.0.18-1.6.x and
> > > php-imap-3.0.18-8
> > > Mar 03 16:39:14 Package is older than package already in database.
> > > Mar 03 16:39:14 Adding php-ldap-3.0.18-1.6.x.i386.rpm
> > > Mar 03 16:39:14 type(hit) = <type 'string'>
> > > Mar 03 16:39:14 hit = php-ldap:3.0.18:8:
> > > Mar 03 16:39:14 Comparing packages: php-ldap-3.0.18-1.6.x and
> > > php-ldap-3.0.18-8
> > > Mar 03 16:39:14 Package is older than package already in database.
> > > Mar 03 16:39:14 Adding php-manual-3.0.18-1.6.x.i386.rpm
> > > Mar 03 16:39:14 type(hit) = <type 'string'>
> > > Mar 03 16:39:14 hit = php-manual:3.0.18:8:
> > > Mar 03 16:39:14 Comparing packages: php-manual-3.0.18-1.6.x and
> > > php-manual-3.0.18-8
> > > Mar 03 16:39:14 Package is older than package already in database.
> > > Mar 03 16:39:14 Adding php-pgsql-3.0.18-1.6.x.i386.rpm
> > > Mar 03 16:39:14 type(hit) = <type 'string'>
> > > Mar 03 16:39:14 hit = php-pgsql:3.0.18:8:
> > > Mar 03 16:39:14 Comparing packages: php-pgsql-3.0.18-1.6.x and
> > > php-pgsql-3.0.18-8
> > > Mar 03 16:39:14 Package is older than package already in database.
> > > [root@x2hvd test]#
> > >
> > > -Rob
> > >
> > > > -----Original Message-----
> > > > From: John Berninger [mailto:[email protected]]
> > > > Sent: Friday, March 08, 2002 11:26 AM
> > > > To: [email protected]
> > > > Subject: Re: [Current-server] 0.99 vs redhat
> > > >
> > > >
> > > > Rob -
> > > >
> > > > Could you issue the following RPM query for each
> > of the files
> > > > involved (i.e. php-3.0.18-8.i386.rpm vs
> > php-3.0.18-1.6.x.i386.rpm):
> > > >
> > > > rpm -qp $filename --queryformat "%{NAME} %{EPOCH}
> > %{VERSION}
> > > > %{RELEASE}" and send us the output?
> > > >
> > > > Also, you mentioned you're running on 7.2, but the
> > > > rpm-python-4.0.2-8 package is a 7.1 RPM; the version
> > shipped with 7.2
> > > > was rpm-python-4.0.3-1.03. Do you have a spare machine you
> > > > can do a new
> > > > install of 7.2 on available, then up2date it through RHN,
> > then test a
> > > > fresh install of the current-0.99.0 RPM, to see what a
> > > > sanitized install
> > > > does for you?
> > > >
> > > > On Fri, 08 Mar 2002, Rob Hagopian wrote:
> > > >
> > > > > rpm-python-4.0.2-8
> > > > >
> > > > > I am able to reproduce it:
> > > > > Make the package directory
> > > > > Move in the most recent (php-3.0.18-8) packages
> > > > > [cadmin create <- shouldn't be necessary, but I
> > did do it]
> > > > > Move in the old (php-3.0.18-1.6.x) packages
> > > > > cadmin create
> > > > >
> > > > > Yields:
> > > > > Mar 03 10:25:22 Adding php-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:22 Adding php-imap-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:22 Adding php-ldap-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:22 Adding php-manual-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:23 Adding php-pgsql-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:23 Adding php-3.0.18-1.6.x.i386.rpm
> > > > > Mar 03 10:25:23 type(hit) = <type 'string'>
> > > > > Mar 03 10:25:23 hit = php:3.0.18:8:
> > > > > Mar 03 10:25:23 Comparing packages: php-3.0.18-1.6.x and
> > > > php-3.0.18-8
> > > > > Mar 03 10:25:23 Removing php-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:23 Older package deleted from db.
> > > > > Mar 03 10:25:23 Adding php-imap-3.0.18-1.6.x.i386.rpm
> > > > > Mar 03 10:25:23 type(hit) = <type 'string'>
> > > > > Mar 03 10:25:23 hit = php-imap:3.0.18:8:
> > > > > Mar 03 10:25:23 Comparing packages: php-imap-3.0.18-1.6.x and
> > > > > php-imap-3.0.18-8
> > > > > Mar 03 10:25:23 Removing php-imap-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:23 Older package deleted from db.
> > > > > Mar 03 10:25:23 Adding php-ldap-3.0.18-1.6.x.i386.rpm
> > > > > Mar 03 10:25:23 type(hit) = <type 'string'>
> > > > > Mar 03 10:25:23 hit = php-ldap:3.0.18:8:
> > > > > Mar 03 10:25:23 Comparing packages: php-ldap-3.0.18-1.6.x and
> > > > > php-ldap-3.0.18-8
> > > > > Mar 03 10:25:23 Removing php-ldap-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:23 Older package deleted from db.
> > > > > Mar 03 10:25:23 Adding php-manual-3.0.18-1.6.x.i386.rpm
> > > > > Mar 03 10:25:23 type(hit) = <type 'string'>
> > > > > Mar 03 10:25:23 hit = php-manual:3.0.18:8:
> > > > > Mar 03 10:25:23 Comparing packages: php-manual-3.0.18-1.6.x and
> > > > > php-manual-3.0.18-8
> > > > > Mar 03 10:25:23 Removing php-manual-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:23 Older package deleted from db.
> > > > > Mar 03 10:25:24 Adding php-pgsql-3.0.18-1.6.x.i386.rpm
> > > > > Mar 03 10:25:24 type(hit) = <type 'string'>
> > > > > Mar 03 10:25:24 hit = php-pgsql:3.0.18:8:
> > > > > Mar 03 10:25:24 Comparing packages: php-pgsql-3.0.18-1.6.x and
> > > > > php-pgsql-3.0.18-8
> > > > > Mar 03 10:25:24 Removing php-pgsql-3.0.18-8.i386.rpm
> > > > > Mar 03 10:25:24 Older package deleted from db.
> > > > >
> > > > >
> > > > > This does NOT happen if you add in the old (-1.6.x)
> > packages to the
> > > > > directory first (or both at the same time using 'mv
> > php-* i386/.'),
> > > > > suggesting that the label comparison isn't really working
> > > > and that the
> > > > > directory listings are really providing the ordering?
> > > > >
> > -Rob
> > > > >
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: John Berninger [mailto:[email protected]]
> > > > > > Sent: Friday, March 08, 2002 10:07 AM
> > > > > > To: [email protected]
> > > > > > Cc: Rob Hagopian
> > > > > > Subject: Re: [Current-server] 0.99 vs redhat
> > > > > >
> > > > > >
> > > > > > Rob -
> > > > > >
> > > > > > I've just tested the compare you indicated, and
> > > > I'm getting a
> > > > > > different result. The compare is issuing:
> > > > > > rpm.labelCompare( ('', '3.0.18', '1.6.x'), ('',
> > > > > > '3.0.18', '8'))
> > > > > > which should result in a -1, and which should then exit the
> > > > > > routine and
> > > > > > go on to the next RPM. When I manually issue this compare, I
> > > > > > do indeed
> > > > > > get a -1:
> > > > > >
> > > > > > >>> rpm.labelCompare( ('', '3.0.18', '1.6.x'), ('',
> > > > '3.0.18', '8'))
> > > > > > -1
> > > > > >
> > > > > > What version of the rpm-python package are you using?
> > > > > >
> > > > > > On Fri, 08 Mar 2002, Rob Hagopian wrote:
> > > > > >
> > > > > > > Yes, that was from the 6.2 tree (hence the question about
> > > > > > rpm v3 perhaps
> > > > > > > being different)...
> > > > > > >
> > > > > > > Server is RH 7.2, running current 0.99, clients are 6.2,
> > > > > > 7.0, 7.1, and
> > > > > > > 7.2...
> > > > > > >
> > > > -Rob
> > > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: John Berninger [mailto:[email protected]]
> > > > > > > > Sent: Friday, March 08, 2002 9:24 AM
> > > > > > > > To: [email protected]
> > > > > > > > Subject: Re: [Current-server] 0.99 vs redhat
> > > > > > > >
> > > > > > > >
> > > > > > > > Rob -
> > > > > > > >
> > > > > > > > I'm trying to confirm this is a bug in the
> > > > > > rpm.labelCompare()
> > > > > > > > function / code, but in my 7.2 tree I see only php 4.x
> > > > > > packages. What
> > > > > > > > release level are you attempting to server with Current?
> > > > > > Also, what
> > > > > > > > release level is your Current server (Red Hat
> > > > version, not Current
> > > > > > > > version)?
> > > > > > > >
> > > > > > > > On Thu, 07 Mar 2002, Rob Hagopian wrote:
> > > > > > > >
> > > > > > > > > inlined too...
> > > > > > > > >
> > > > > > > > > > Eww. This is not good - the code
> > used to compare
> > > > > > > > the package
> > > > > > > > > > ages is take directly from the Python RPM
> > bindings - we
> > > > > > > > actually make a
> > > > > > > > > > call to rpm.labelCompare() which is
> > implemented inside
> > > > > > > > the rpm-python
> > > > > > > > > > shared library. I'll check Red Hat's bugzilla and if
> > > > > > > > encessary submit a
> > > > > > > > > > bug about this. I don't think we can do
> > much, since we
> > > > > > > > don't want to be
> > > > > > > > > > in the business of comparing RPM ages...
> > that gets ugly
> > > > > > > > real quick.
> > > > > > > > >
> > > > > > > > > Of course not... strangely enough it did
> > correctly figure
> > > > > > > > out the other
> > > > > > > > > associated php packages (kicking out -1.6.x in favor of
> > > > > > > > -8)... could this be
> > > > > > > > > related to the fact that these are redhat 6.2
> > > > packages (rpm v3)?
> > > > > > > > >
> > > > > > > > > > This is definitely somewhere in the
> > Current code
> > > > > > > > - I'll try to
> > > > > > > > > > take a look at it tonight / tomorrow / this
> > weekend. I'm
> > > > > > > > not /too/
> > > > > > > > > > surprised at this, but having it warn me about
> > > > > > symlink issues on
> > > > > > > > > > different archs doesn't really seem quite right. :)
> > > > > > > > >
> > > > > > > > > And I wasn't too concerned about it (I've
> > gotten real used
> > > > > > > > to spurious
> > > > > > > > > warnings with gcc :-) ), but I did clean up the
> > tree rather
> > > > > > > > than gamble, so
> > > > > > > > > I can't tell you if it actually did
> > the-right-thing(tm)...
> > > > > > > > >
> > > > > > > > > > Except for the PHP thing above, you
> > > > really shouldn't
> > > > > > > > > > have to do that anymore - or, at least, we'd
> > hoped so...
> > > > > > > > >
> > > > > > > > > As long as v1.0 is perfect ;-)
> > > > > > > > >
> > > > > > -Rob
> > > > > > > > > _______________________________________________
> > > > > > > > > Current-server mailing list
> > > > > > > > > [email protected]
> > > > > > > > >
> > http://lists.dulug.duke.edu/mailman/listinfo/current-server
> > > > > > > >
> > > > > > > > --
> > > > > > > > Thank you,
> > > > > > > > John Berninger
> > > > > > > >
> > > > > > > > Systems Administrator [email protected]
> > > > > > > > Department of Mathematics Box 8205, Harrelson Hall
> > > > > > > > NC State University Raleigh, NC 27695
> > > > > > > > Phone: (919)515-6315 Fax: (919)515-3798
> > > > > > > >
> > > > > > > > GPG Key ID: A8C1D45C
> > > > > > > > Fingerprint: B1BB 90CB 5314 3113 CF22
> > 66AE 822D 42A8
> > > > > > > > A8C1 D45C
> > > > > > > > --
> > > > > > > > _______________________________________________
> > > > > > > > Current-server mailing list
> > > > > > > > [email protected]
> > > > > > > >
> > http://lists.dulug.duke.edu/mailman/listinfo/current-server
> > > > > > > >
> > > > > > > _______________________________________________
> > > > > > > Current-server mailing list
> > > > > > > [email protected]
> > > > > > > http://lists.dulug.duke.edu/mailman/listinfo/current-server
> > > > > >
> > > > > > --
> > > > > > Thank you,
> > > > > > John Berninger
> > > > > >
> > > > > > Systems Administrator [email protected]
> > > > > > Department of Mathematics Box 8205, Harrelson Hall
> > > > > > NC State University Raleigh, NC 27695
> > > > > > Phone: (919)515-6315 Fax: (919)515-3798
> > > > > >
> > > > > > GPG Key ID: A8C1D45C
> > > > > > Fingerprint: B1BB 90CB 5314 3113 CF22 66AE 822D 42A8
> > > > > > A8C1 D45C
> > > > > > --
> > > > > >
> > > > > _______________________________________________
> > > > > Current-server mailing list
> > > > > [email protected]
> > > > > http://lists.dulug.duke.edu/mailman/listinfo/current-server
> > > >
> > > > --
> > > > Thank you,
> > > > John Berninger
> > > >
> > > > Systems Administrator [email protected]
> > > > Department of Mathematics Box 8205, Harrelson Hall
> > > > NC State University Raleigh, NC 27695
> > > > Phone: (919)515-6315 Fax: (919)515-3798
> > > >
> > > > GPG Key ID: A8C1D45C
> > > > Fingerprint: B1BB 90CB 5314 3113 CF22 66AE 822D 42A8
> > > > A8C1 D45C
> > > > --
> > > > _______________________________________________
> > > > Current-server mailing list
> > > > [email protected]
> > > > http://lists.dulug.duke.edu/mailman/listinfo/current-server
> > > >
> > > _______________________________________________
> > > Current-server mailing list
> > > [email protected]
> > > http://lists.dulug.duke.edu/mailman/listinfo/current-server
> >
> > --
> > Thank you,
> > John Berninger
> >
> > Systems Administrator [email protected]
> > Department of Mathematics Box 8205, Harrelson Hall
> > NC State University Raleigh, NC 27695
> > Phone: (919)515-6315 Fax: (919)515-3798
> >
> > GPG Key ID: A8C1D45C
> > Fingerprint: B1BB 90CB 5314 3113 CF22 66AE 822D 42A8
> > A8C1 D45C
> > --
> > _______________________________________________
> > Current-server mailing list
> > [email protected]
> > http://lists.dulug.duke.edu/mailman/listinfo/current-server
> >
> _______________________________________________
> Current-server mailing list
> [email protected]
> http://lists.dulug.duke.edu/mailman/listinfo/current-server
--
Thank you,
John Berninger
Systems Administrator [email protected]
Department of Mathematics Box 8205, Harrelson Hall
NC State University Raleigh, NC 27695
Phone: (919)515-6315 Fax: (919)515-3798
GPG Key ID: A8C1D45C
Fingerprint: B1BB 90CB 5314 3113 CF22 66AE 822D 42A8 A8C1 D45C
--