launchpad-dev team mailing list archive

[Jonathan Lange <jml@xxxxxxxxx>]

On Thu, Jun 7, 2012 at 2:57 PM, Stuart Bishop
<stuart.bishop@xxxxxxxxxxxxx> wrote:
> On Thu, Jun 7, 2012 at 8:38 PM, Jonathan Lange <jml@xxxxxxxxx> wrote:
>
>> I'm trying to follow the database patch process[1] to rename
>> Archive.commercial to Archive.suppress_subscription_notifications. I
>> hit problems when I get to 'make newsampledata'.
>>
>> First, I get a lot of spew: http://paste.ubuntu.com/1028563/. It's
>> mostly of the form:
>>  pg_dump: NOTICE: there are circular foreign-key constraints among
>> these table(s):
>>  pg_dump:   bug
>>  pg_dump: You may not be able to restore the dump without using
>> --disable-  triggers or temporarily dropping the constraints.
>>  pg_dump: Consider using a full dump instead of a --data-only dump to
>> avoid this problem.
>>
>> Can I safely ignore these?
>
> Yes. This is noise from the hack that is sampledata.
>

Thanks. I'll amend the DB patch page.

>
>> Second, if I run 'make newsampledata' straight off db-devel after
>> running 'make schema', I get changes to the sample data for
>> ArchivePermission and SourcePackagePublishingHistory.
>>
>> The ArchivePermission changes seem to be due to a new column, 'pocket'
>> on that table. I cannot make sense of the SPPH changes.
>>
>> My understanding is that 'make newsampledata' without any db patches
>> after a 'make schema' should result in 0 changes, and thus that
>> something went wrong before I started my branch.
>
> This would be the case if every database patch landed also updates the
> sampledata, but this step is often missed.
>

Ahh OK. I'll include those changes in my branch then.

>> I'm unclear how I should proceed, since I need to include sample data
>> changes in my database patch, but these changes are not related.
>
> I don't review sample data changes - it is a black box, and the sample
> data is good if the tests pass. Don't worry about it, and in a perfect
> world do the sample data update after review to keep the diff in the
> MP managable. As you have found, it gets pretty noisy sometimes.
>
>

Thanks,
jml