I understand. Then let's approach this another way. Here's the patch we
made in createrepo/dumpMetadata.py:
def arch(self):
if self.tagByName('sourcepackage') == 1 or (
not self.options['relaxed-sourcerpm-check'] and not
self.tagByName('sourcerpm') ):
return 'src'
else:
return self.tagByName('arch')
Here's the original "if" expression:
if self.tagByName('sourcepackage') == 1 or not
self.tagByName('sourcerpm'):
What risks are we taking by allowing binary RPMs to omit a sourcerpm
tag? Can we assume that, for workaround purposes, the sourcepackage tag
was not adequate for detecting source RPMs and, hence, the additional
test for a missing sourcerpm tag was required? I lack your broad
exposure to RPMs, so I don't know how commonly this condition occurs.
Regards,
Jack
-----Original Message-----
From: yum-bounces@xxxxxxxxxxxxxxxxx
[mailto:yum-bounces@xxxxxxxxxxxxxxxxx] On Behalf Of Seth Vidal
Sent: Friday, February 26, 2010 11:41 PM
To: Yellowdog Updater, Modified
Subject: Re: createrep question
On Fri, 26 Feb 2010, Kidwell,jr, Jack wrote:
> I am advised to "lawyer up", so I cannot send the raw header dump you
> requested. (Pesky non-disclosure agreement.) I am sending an ascii
> header dump of the COTS RPM tags and their respective value data type.
A
> few generic tag values are exposed. Also attached is the script that
> dumps these headers.
>
>> From these tags observe that there is no 'sourcerpm' or
'sourcepackage'
> tag.
>
> I apologize for being obtuse.
>
> Thanks for your help!
I really cannot help you unless I can get more specific information on
this package.
I'm not trying to be an obstacle but this is like feeling around in the
dark.
-sv
_______________________________________________
Yum mailing list
Yum@xxxxxxxxxxxxxxxxx
http://lists.baseurl.org/mailman/listinfo/yum
_______________________________________________
Yum mailing list
Yum@xxxxxxxxxxxxxxxxx
http://lists.baseurl.org/mailman/listinfo/yum
