[Babase] More PsionLoad Questions

Lacey Maryott lacey.maryott at duke.edu
Mon Jun 18 11:46:23 EDT 2007


Hey Karl,
    While uploading yesterday, Leah and I came up with a new list of 
questions, comments, concerns about Psion load. See below.

1. When we upload a file twice, there is no warning, or block against 
it. (We found out the hard way :-P) Is there a way to have the system 
prevent that?

2.  A follow up to 1. We entered a file twice and we need to get it back 
out of babase.  The corresponding SID's are 38529 through 38546.  Given 
what we found in that particular file, we expect that we will need to 
eliminate rows in samples, neighbors, parts, interact, allmiscs, and 
point_data.  Where should we start with getting rid of these rows? 

3. We were wondering what happens if in the middle of an upload, the 
internet glitches or cuts out or firefox crashes?  Will half the file go 
in, or none? Just curious.

4. Just FYI, it takes about 10 seconds to upload each clean file.

5. The new message you put in about what file is uploaded is very 
helpful. Interestingly though, it comes up even when an error message 
displays.  It says "uploaded file: 060701P3" then gives an error.  This 
is moderately misleading, could it be reworded if the file doesn't 
actually upload?

6. We found that each hour, to the minute I would get logged off, even 
if I was in the middle of uploading files (ie: without idle time).  Is 
there a way around this, or did you make this this way for a reason? 
It's not a bad idea if the comp is idle, just is kind of annoying if you 
are on a role :-P

7. In file 060921P1 there was a row that just said F2..., then the next 
2 rows were OS..., but the error which was picked up on this was 
"Kidsuckle must be S,N, etc".  We thought it odd that it didn't pick up 
on the missing infant position before jumping to the suckle code.  Also, 
is there a code for "UNK" if we have an F2 row but no information about 
the infant?

8. Another oops, kind of.  In 061003P4 at 15:50:11, OXF was sampled. OXF 
is not in biograph, so we changed it to 998, which has been the protocol 
for these types of errors. However, it turns out that one of Onyo's 
babies was born OXF, but its name seems to have been changed to ISO and 
there are no notes in the demog or anything to indicate this.  We hunted 
down both and we are confident this is what happened.  So, in the name 
of accuracy, the 998 is pntid 428168, nghid 476023 should be ISO and not 
998. Is there a way to change this manually?

9. Lastly, a WWWdiff question, in reference to the error we found where 
the one cell had a space after the sname, is it possible to tell the 
program to only read the first 3 spots in each cell, avoiding that 
problem? 

Ok, I think thats all!   (for now hehe)

thanks so much!

Lacey

-- 
Lacey Maryott
Alberts Lab
Department of Biology
Duke University
ph: 919-660-7306
fax: 919-660-7293
Lacey.Maryott at duke.edu 



More information about the Babase mailing list