Reloading babase_test -- Was: Re: [Babase] progress on 07a update
lacey.maryott at duke.edu
lacey.maryott at duke.edu
Fri Jan 25 20:56:43 EST 2008
I did a little poking around, and I think I can probably fix these samples by
just looking back at the Psion files, which are luckily on the server,
so I can
do it from home :-D. Unless anyone objects, I think I'll just fix the times.
IT won't take me long... I can go ahead and start now. Then we can have the
validation both places, and don't need to lose the data
Quoting "Karl O. Pinc" <kop at meme.com>:
>
> On 01/25/2008 03:47:04 PM, Karl O. Pinc wrote:
>>
>> On 01/25/2008 02:12:59 PM, kfenn wrote:
>>
>>> So, my question for you is, can you give me a fresh copy of
>>> babase_test that mimics what is currently in babase?
>>
>> Ok. I'll be doing this now-ish, so anybody who's working in
>> babase_test will have to make allowances.
>
>
> I have a problem. There is a discrepancy between the validation
> in babase_test and babase. I just re-created the babase_test
> database, with what are _supposed_ to be the latest set of
> validation rules and found that while babase_test validation
> rules do not allow a ALLMISCS.Atime value to be later than
> 19:00, the babase database contains no such rule.
>
> This is probably a rule I added to babase after some discussion,
> but it did not go in because data cleanup was required.
>
> Here are the rows that violate this rule.
>
>
> select * from allmiscs where atime > '19:00';
> almid | sid | atime | txt
> -------+-------+----------+-------------------------
> 50 | 2937 | 20:10:26 | U,MISSED,OUT OF SIGHT.
> 374 | 5321 | 20:14:58 | U,1330/1430 NOT DONE
> 408 | 5596 | 20:32:59 | U,1700 SAMPLES NOT DONE
> 409 | 5596 | 20:33:18 | U,IDS WITH RSM.
> 416 | 5645 | 19:57:11 | C,ROC,C,WEN
> 417 | 5662 | 19:58:24 | C,ROC,C,WEN
> 464 | 6135 | 20:17:23 | U,1330 SAMPLE BATT. LOW
> 2041 | 36783 | 21:18:39 | C,YOB,C,MON
> (8 rows)
>
>
> As a result, the babase_test database will not load with
> the babase data. I could fudge things and remove this rule
> but it's probably better to clean up the data.
>
> Please let me know where to go with this.
>
>
> If you like, you can try copying the data yourself with the
> following commmand (all one line):
>
> pg_dump --format=c --compress=0 --schema=babase --data-only --user
> babase_admin babase | pg_restore --data-only --disable-triggers --user
> babase_admin --dbname=babase_test
>
> You will need the babase_admin administrator password.
>
> Note that the above command will only work once. After that
> all data must be deleted from the babase_test database before
> it can be run again.
>
>
> Karl <kop at meme.com>
> Free Software: "You don't pay back, you pay forward."
> -- Robert A. Heinlein
>
> _______________________________________________
> Babase mailing list
> Babase at www.eco.princeton.edu
> http://www.eco.princeton.edu/mailman/listinfo/babase
>
More information about the Babase
mailing list