1 (edited by adhale 13-07-2006 12:32:29)

Re: Trouble with lichens (R2k import to R6)

I've recently carried out (another!) import of my data from R2K to R6 (fresh installation of R6 & MSDE). One of the Surveys I have is "Wales Threatened Lichen Records". This is a dataset prepared by the British Lichen Society on contract to CCW. It was imported to my copy of R2K from a zipped Access db, with no problems. However, when I came to do the import to R6, this was the only one of several surveys (including others imported from elsewhere) which has not come over correctly. The Survey, Survey Events and Smples are there but not the Taxon Occurrences when viewed in the Observation window.

ImportRejects.txt reports no rejects on this import, and although there are some entries in the UpgradeErrors.mdb, I can't immediatley trace this problem back to any of them.

If I query the database directly to report the number of taxon occurrences by survey, the correct number of occurrences is reported for this survey.

The data was entered using the "British Isles List of Lichens and Lichenicolous Fungi", not the "official" FUNGI & LICHENS list. I wondered if this was the problem. However, I have a separate survey with lichen records entered by myself, mostly from the official list, but with a few from the "British Isles etc. " list. These seem to have come over OK.

R6 is version 6.6.8.59 with the dictionary update in Rec6Dictionary669.zip  applied.

Any suggestions please?

Thanks

Alan Hale
Countryside Council for Wales

Alan Hale
Aberystwyth

2

Re: Trouble with lichens (R2k import to R6)

No response on this so I assume everyone else is a baffled as I am.

I decided to try and import the data direct from the original zip file that I use to import to R2K2. This told me that the "missing" taxon occurrences were actually already in the database, since they were flagged up as duplicates. It also gave me the actual taxon occurrence keys, so I was able to double-check by a select directly on the database that they were all there - indeed they were.

So, for some reason Recorder 6 is choosing not to display these taxon occurrences.

This does remind me of a similar problem Sally Rankin reported with when taxon occurrences entered from the Environment Agency taxon dictionary in R2K2 were transferred to R6 - but if I recall correctly this is because that list was missing from R6. This is not the case here.

Again, I'd be very glad of any help.

Cheers

Alan

Alan Hale
Aberystwyth

3

Re: Trouble with lichens (R2k import to R6)

OK, I've done some more digging around and it appears that although the Taxon_Occurrence_Keys from this lichen dataset  are listed in the Taxon_Occurrence table, there are no matching entries in the Taxon_Determination table.

Any ideas on how (or if?) I can fix this?

Many thanks

Alan

Alan Hale
Aberystwyth

4

Re: Trouble with lichens (R2k import to R6)

Just a thought, but have you tried rebuilding your indexes?

Charles Roper
Digital Development Manager | Field Studies Council
http://www.field-studies-council.org | https://twitter.com/charlesroper | https://twitter.com/fsc_digital

5

Re: Trouble with lichens (R2k import to R6)

Good thought - but yes, I did, and the problem persists.
Cheers

Alan

Alan Hale
Aberystwyth

6

Re: Trouble with lichens (R2k import to R6)

Nothing immediately strikes me  on this, but I would be happy to look at it  if you can send me the data.  I wouldn't have thought that it was anything that complicated assuming  the  taxon determinations are there ? 

Mike Weideli

Mike Weideli

7

Re: Trouble with lichens (R2k import to R6)

If the taxon determination for an occurrence is not present or if it contains an invalid taxon list item key then the observation will not appear in the observation hierarchy because there is no species name to display.

The first thing I would try is checking whether there are entries in the corresponding tables for some of the foreign keys (TAXON_LIST_ITEM_KEY, DETERMINER, DETERMINATION_TYPE_KEY, DETERMINER_ROLE_KEY, ENTERED_BY, CHANGED_BY, SOURCE_KEY) that are in the missing determinations. The tables that are in UpgradeErrors.mdb should give you a clue as to which ones to start with.

Alternatively you can try cutting and pasting a missing determination into the Recorder 6 database. The error message that will appear should give you a clue to what is wrong.

If you have any problems with this please phone me on 01491 578633.

Sally Rankin
Recorder Approved Expert

Sally Rankin, JNCC Recorder Approved Expert
E-mail: s.rankin@btinternet.com
Telephone: 01491 578633
Mobile: 07941 207687

8 (edited by adhale 26-07-2006 17:06:46)

Re: Trouble with lichens (R2k import to R6)

Some progress with this.

I discussed on the phone with Sally this afternoon and we found, surprisingly, that pasting one of the missing rows in TAXON_DETERMINATION from R2k2 to the Rec6 linked Access database did NOT elicit any error. And when I the opened the Obs window in Rec6 the corresponding taxon occurrence was now visible!

Sally then suggested exporting the whole original TAXON_DETERMINATION to a new database, zipping it and importing into Rec6. I found however that I could not export the table - I kept getting a message from the JET engine saying another user was changing the data at the same time - which is very odd (I have stand-alone installations of both R2K2 and Rec6). I tried a Make Table query with no filter to get a copy of the table - same message. However, intriguingly, I was able to do a Make Tabe WITH a filter to just pick out the rows missing in Rec6. This looked promising, and I exported the new table to a new database, zipped it, and imported into Rec6 - apparently with no problem. However, the taxon occurrences STILL did not show up in the Obs window.

Then I thought of going back to the original zipped database that I imported the lichen data from into R2K2 in the first place, choosing All Imported to deal with duplicates. This seemed to work, with the caveat below:

At the very end of the import I got a message saying "Unable to insert the item since it references an item which does not exist". Very helpful! On brief inspection of the data, I found one taxon occurrence that was not now showing. Most were.

None of this explains why there was a problem with the original import. There were no references to the TAXON_DETERMINATION table in the UpgradeErrors.mdb. (It's worth noting here that there WERE a lot of errors reported relating to locations - these all seemed to derive from system data to do with Vice-Counties).

Now I'm going home!

Thanks for your help (and offers of help) everyone, especially Sally.

Cheers

Alan

Alan Hale
Aberystwyth

9

Re: Trouble with lichens (R2k import to R6)

The errors relating to vice-counties are probably due to the vice-county locations that get installed into R2K2 as a part of the import wizard add-in (I think). They're not in R6, hence the errors.

Charles Roper
Digital Development Manager | Field Studies Council
http://www.field-studies-council.org | https://twitter.com/charlesroper | https://twitter.com/fsc_digital