Ebrary MARC Records

Ebrary records are regularly updated, with the library receiving files of MARC records to add to the collection and files of existing records to be deleted from the collection. 

Files of ebook records will not include item information. We have found that it is useful to create dummy item records for these ebrary records. Having item records will add flexibility to faceting in Koha improving the searchers’ ability to refine results to include/exclude the ebook collection. This is important when the ebook collection can be bigger than the print book collection in the Koha database. 

This document describes the following processes:

Page 2 Creating a MARC modification template to insert an item tag into the incoming ebrary records

Page 6 Creating the matchpoint so that incoming records can be checked against the database to see if 

they already exist

Page 9 Loading a file of new ebrary records

Page 14 Loading a file of ebrary deletes and processing them in Koha

MARC MODIFICATION TEMPLATE

The MARC modification template is a very powerful tool. It allows the library make changes to MARC records as they are imported into Koha. These changes can be the addition, deletion or alteration of MARC tags and subfields. The general process is described in detail here: http://es.koha-community.org/manual/master/publican/en/html/marcmodtemplates.html

In this document, we focus on adding a MARC tag to insert item information to the record.

In order to create an item record, the information needs to be added to a 952 tag. 

(The details of the use of the 952 tag can be found here: https://wiki.koha-community.org/wiki/Holdings_data_fields_(9xx) )

The following pieces of information need to be inserted into the 952 tag:

Note that you enter the code, not the description, in the branch, shelving location and itemtype values. 

There are other fields in the item record that you could choose to enter (eg Public Notes)

Fill in your selections above and retain for reference

Creating the Modification Template

Go to Tools and select MARC modification templates from the Catalog column

Select

Give it a name

(As you create more templates, you can use the “Duplicate” option to use a current template as the base for a new one.)

On the next screen

use the 

button to start building the transformation rules.

This opens the following screen

The system allows you to do a number of tasks:

The second drop-down on the Action allows you to define conditional processing

but that is for another day.

In this document we will only be adding new information to the MARC record. 

In order to add the 5 elements identified above, you need to create 5 separate actions, one per subfield.

Fill out the fields in the form 

Give each action a meaningful description

Repeat until you have completed the fields you want to add

MARC LOAD MATCHPOINT

In order to load the records into your Koha system, you will need to set up a match point. 

This is a rule where you tell the import process to look for existing records on the koha database based on some information in the incoming record. Typically this would be a match on the ISBN.

For ebrary records we are going to use the 001 tag as the match point as ebrary records have an unique identifier in this field. This is searched using the control-number index.

Creating the Matchpoint

Go to 

select 

from the Catalog menu

Press the 

button

Fill out the form like this:

Marching Rule Code Give this rule a name

Description Give a clear description of what it does

Match Threshold This is a number that the match rules have to meet for it to be deemed a match for 

an existing record 

Record Type The type of record being entered - bibliographic or authority.

The system allows multiple tests to be done (eg match on ISBN, and Title). They can be given individual weights and cumulatively they need to be equal to or more than the Threshold. In this example, we are just setting a single match point.

Search index This is the index that the matchpoint tag is included in

(See http://manual.koha-community.org/3.2/en/kohasearchindexes.html for details of what tags are indexed where)

Score This is the score a match on this tag gets

Tag This defines the tag and subfield to look for the matching information 

(The 001 does not have a subfield)

IMPORTING NEW MARC RECORDS

New Records

When new records are made available from ebrary,you will need to load them to Koha

The MARC load process has two steps: 

Stage MARC Records

Manage Staged Records

This allows you to import records and alter settings before committing them to the database. 

It also allows you to reverse an import is the results were not as expected.

Go to

Select 

Choose the file and upload it into Koha

Complete the form that displays after the load:

Comments about this file: Give it a meaningful comment to help identify the purpose of the load.

Record Type Bibliographic or Authority depending of data being loaded.

Character Encoding Should always be utf-8 unless provider indicates otherwise.

Some older systems use MARC8.

This will mess up diacritic displays if it is set incorrectly.

Format MARC or MARCXML depending on the source. Usually MARC

Modify Record… Select the template just created. 

(This will add the item information)

Record Matching Rule… Select the rule created above.

Action if found This tells the system whether or not to overlay records. Usually set to Replace 

existing records as the expectation is that the incoming record will be more up-to-date that the existing record 

Action if not found This tells the system what to do with new records.

Check for Embedded Items We will want to add item records so the settings are usually 

Yes / Always Add Items

Then press 

When the records are processed, you get a summary of the results:

Then press the 

button and prepare the records to be loaded into the system:

You will then see a detailed screen showing all the matches. 

The top of the screen allows you to change the match rules if you made a mistake, or want to try a different match point. 

You can also select the framework to be used for the incoming records. 

The matching titles are listed in the lower half of the screen. The “DIFF” column allows you to create a display that shows the difference between the incoming record and the existing one.

When you are happy to proceed, press the 

Button, and the records are included into the database.

When the load is complete, the system re-paints the screen showing a summary of the results and in the “Record” column at the right of the screen the updated or newly created record is shown.

If you want to check the record, simply right-click on the record number and select the  Open link in new window option.

Then you can check the process has done what you want. 

If not, you can use the

button to reverse the process.

DELETING EXISTING  MARC RECORDS

From Time to time Ebrary will send you files of records that are no longer part of the collection and should be deleted.

These records have a specific flag in the MARC leader

When you receive a file of MARC records to delete follow the same process as adding new records above. 

This will overwrite the existing record setting the delete flag on the MARC record.

Once the records are loaded there is then a report that will identify the affected records can be used to access the records to delete them.

The test file used above has been altered to have a “d” flag in the records. 

Loading the records again results in them being overlaid:

When you have uploaded the records, the settings will be slightly different:

These are the differences:

Modify Record… Do not select anything here - we do not want to create items

Record Matching Rule… Select the rule created above.

Action if found This tells the system whether or not to overlay records. Usually set to Replace 

existing records as the expectation is that the incoming record will be more up-to-date that the existing record 

Action if not found Ignore as they are not on your database.

Check for Embedded Items No

Then press

Manage the staged records and import as described above.

(You will need to request Interleaf to add the report described below to your system.)

Then go to 

Select 

Then find the report titled “Ebrary Records for Deletion” and press the

button

This will list all records on the system with the “d” flag. 

You can either delete these records individually by right-clicking on the title and opening the record in a new window and removing the record that way, or copying the biblionumbers in the first column and using the Batch deletion tool to remove them.

BATCH DELETION TOOL

Go to 

Select 

From the Catalog Menu

Fill out the form:

And press the 

button.

The system will then list the records to be deleted:

You can elect to keep individual records by un-ticking them, or use the “Select without…” options to refine the selection.

Press the

Button and the system will confirm the records are gone.