Hello Seth We did not change any Koha files. We did use Perl but to insert the full length final bracode number into the MARC records (along with a slew of other item-specific data such as call number, home branch, item type). The MARC records were then imported after the 952 fields were built. In a few cases there were a few identifiable categories of materials that did not use the expected barcode format, in which case we ran a script against the items table after import, to change the barcode numbers of those items. Lastly, if there were different barcode formats that could not be predicted, these were fixed by a custom script we spliced into the checkout process - it only runs if the scanned barcode is not found in the database. This is Petrus' work, and I have forwarded your messages to him and copy this one to him also. If you wish to work with him you could contact him direct, off list. Hope that helps, Agnes Seth Mayo wrote:

Can Petrus give any hints as to which perl files need to be changed to address barcode differences? I'm going to see if I can do it myself as I have some programming experience. However, is there a possibility of working with Petrus if I can't get it to work?

Seth Mayo IT Technical Manager Email: smayo@stpeters.nsw.edu.au ------------------------------------------------------- St Peter's Anglican Primary School Howe Street Campbelltown NSW 2560 PH: 02 4627 2990 Website: http://www.stpeters.nsw.edu.au

2009/11/18 Agnes Rivers-Moore <arm@hanover.ca>:

...

We have just migrated 6 libraries. Of those 6, three have more than one variant of barcode, two had 4 types. Really a nightmare!

It took us a lot of SQL and Perl script work by a consultant with those skills, but it is possible if the size of your collection makes it worthwhile. If you have a definable range of numbers that can be changed to one type, that can be done with SQL in the table after import, or by script editing the 952 fields before import. Better to use the first option if you want to import and match other circulation or item information after the MARC records are in. We also have a clever script that allows us to correct the Koha table by scanning the barcode, as if we were checking it out - saves the need to search for each title and edit the items individually. This was custom work, not part of Koha.

I have copied this message to the consultant who did the work for us...

Agnes

Seth Mayo wrote:

...

We are gradually getting everything together to move to Koha from Alice. Many of you will remember that back in July I was trying out various library systems and decided that Koha was a great system to get on board with.

At this time of the year, the nitty-gritty of the system is going to start getting looked at. I have a few questions to ask regarding the system but the most important one is:

Is there a way to force Koha to re-write barcodes on the fly?

Situation...

Our library happens to have two barcode formats on books (long story):

Type 1 Example: R36758W 2560 Type 2 Example: + 2560 013037 6

There are two problems with this - both having to do with our existing system (Alice for Windows) not actually cataloguing the whole barcode.

Apparently when the system was set up, the following is what it was set up to do (and thus what we need to achieve in Koha since changing the barcodes on all the books is not an option):

Type 1 conversion: R36758W 2560 has been stored as R36758 (string is concatenated) Type 2 conversion: + 2560 013037 6 has been stored as R13037 (string is rearranged and concatenated - ie. first 4 characters removed

The letter 'W' in the first type is randomised on the barcodes so there is no way of recovering it. The 'R' in the second type is simply added to everything.

Any help with any of this would be appreciated.

Seth Mayo IT Technical Manager Email: smayo@stpeters.nsw.edu.au ------------------------------------------------------- St Peter's Anglican Primary School Howe Street Campbelltown NSW 2560 PH: 02 4627 2990 Website: http://www.stpeters.nsw.edu.au _______________________________________________ Koha mailing list Koha@lists.katipo.co.nz http://lists.katipo.co.nz/mailman/listinfo/koha

--

Agnes Rivers-Moore Assistant Librarian Hanover Public Library

-- Agnes Rivers-Moore Assistant Librarian Hanover Public Library

On 2010/01/4, at 11:24 PM, Mason James wrote:

On 2009/11/17, at 5:10 PM, Seth Mayo wrote:

...

We are gradually getting everything together to move to Koha from Alice. Many of you will remember that back in July I was trying out various library systems and decided that Koha was a great system to get on board with.

At this time of the year, the nitty-gritty of the system is going to start getting looked at. I have a few questions to ask regarding the system but the most important one is:

hi Seth.

...

Is there a way to force Koha to re-write barcodes on the fly?

to me, thats the wrong way to approach your problem try to correct your barcodes *before* the import into Koha, and save yourself some future pain.

a small perl script to loop thru your bib file(s) and perform the barcode correction will do the trick just fine

preferably get your bibs from Alice as a big MARC file, then you can use perl's MARC::Record module to do the heavy lifting.

after the correction, import your bibs from the Koha->Tools->Import gui, as usual ;)

http://search.cpan.org/~mikery/MARC-Record-2.0.0/lib/MARC/Record.pm

Mason.

ah, i just re-read your email and clicked to your actual problem ;) you need to do something like... - identify which existing barcodes are corrupt - manually scan all the correct barcodes from the items in your collection, into a file (or db table) containing 2 fields for each item (a unique item-id and the item's corrected barcode) - loop thru your bib records and fetch the corrected barcode matched against that item's unique-id its a bit fiddly, but fixable ;) hope that helps, Mason.