Batch Processing Springer Ebook Records

Purpose: Guidelines of using OCLC numbers supplied by Springer to do batch searches, exporting file to workstation, batch editing records in MarcEdit, loading records to Millennium through Data Exchange, and batch updating holdings on OCLC through Connexion Client.

Background:

Springer provides three ways for libraries to download Marc records for its ebook collection. The first way is through OCLC's WorldCat Collection Set Service. There is some cost involved. The second way is to use the Marc records supplied by Springer. These records were machine generated and of lower quality compared to OCLC records. The third way is to take advantage of the OCLC numbers supplied by Springer to do in-house batch processing. There is no additional fee for doing this if a library has a full subscription to OCLC. This procedure focuses on the third way.

Summary of File Processing Steps:

• Configure Connexion settings for batch load and create file to hold incoming OCLC number content.
• Cumulative file (.xlsx) retrieved from publisher web site.  New content culled for additions. 
• OCLC# column copied and pasted into .txt file for search in OCLC Batch Process mode
• Searches against OCLC numbers list conducted in batch mode and records stored online with OCLC in .dat format
• Exported .dat file is converted to .mrk for a series of data enhancement steps (call numbers, enhanced 856 field) using MarcEdit
• MarcEdit .mrk converted to.mrc for download to URSUS
• Millenium's DataExchange function is use for access to load profile I (capital i) for .mrc load of records
• Test full text access in webpac via authorized 856 links

Procedures:

I. New Content Retrieval and Batch Searching in OCLC Connexion

1. Create Local Bibliographic Files

    For easy management of the downloaded records, it's optimal to set up the specific Local File to save the Springer ebook records. Each Local File technically can hold up to 9,999 records, you can set up as many Local Files as you like. Steps to create new Local Files:
    a. Log on to Connexion. On the File menu, click Local File Manager.
    b. Check Bibliographic.
    c. Click Create File.
    d. In the Open window, enter the new file name. The current naming convention follow the form Springer2013-01-24.
    e. Click Open.





2. Configure Settings for Batch Process

Steps to set the client in batch mode:
    a. On the Tools menu, click Options>Batch.
    b. In the Record Actions area, under Perform local actions in batch, check Bibliographic Record Export.
    c. In the Batch Search Keys to Retain area, check Not Found or Error.
    d. Optionally, check the box next to Display batch reports automatically if you want to view a report each time you export a group of records.
    e. Click Apply.
    f. Click Close.   

 3. Run Batch Processing -- Online Search

    a. Go to Springer OCLC numbers list Web page.
    b. Download the cumulative spreadsheet to your computer.  
    c. Copy the OCLC numbers (see steps i-iii, below), paste them into Notepad, name and save the file in S:/Cataloging/SpringerMonthly. (For consistency, the file will generally have the same name as the local save file created in step 1, e.g., Springer2013-01-24.txt.)
       i. By convention, each month's batch load summary file (created in step III) includes the spreadsheet number of the last loaded file, e.g., Springer2012-12-18-to-32232LoadReport.txt saved in S:/Cataloging/DataExchange/Springer
       ii. To download the incoming month's batch load, begin with the OCLC number of the next record in the file (e.g., record number 32233, from the previous month).
       iii. As of 2012, we are not subscribing to the Springer Protocols collection, so the OCLC numbers for those records need to be skipped. The collection name is listed in column D under "Collection Set." It may be helpful to highlight or change the color of those records as a visual signal that they should be skipped.
    d. In Connexion, on the Batch menu, click Enter Bibliographic Search Keys.
    e. Select the Local File you created as the place to hold downloaded records.
    f.  Under Use default index, select OCLC Number (no:).
    g.  Click Import.
    h. Find the .txt file you created, click Open.
    i. When asked if you want to delete the original file, click No. (It's useful to have the txt file available as a backup in case something goes wrong with the batch process.)
    j.  Click Save.
    k.  Click Close.
    l. On the Batch menu, click Process Batch.
    m.  Check the box next to the Local File you want to hold the records.
    n. In Process area, check Online Searches.
    o.  Click OK to close Process Batch window. Wait for the batch to finish processing, which can take a few minutes, depending on the size of the batch.
    p.  After the batch processing is finished, Connexion will automatically log off. On the File menu, click Logon to continue.   

          

4. Create Data File (.dat) as Export Location

    a. On the Tools menu, click Options>Export.
    b. In the Destination area, click Create.
    c. Select File, click OK.
    d. Name the file (for consistency, generally the same name as the local bibliographical file created in step 1), click Open. Tip: Be sure to change default path when saving to S:/Cataloging/DataExchange/Springer else file is dumped to C:/ProgramFiles/OCLC/Connexion/Program.
    e. Click OK.
    f. Click Apply.
    g. Click Close.

5. Run Batch Processing -- Export

    a. On the Cataloging menu, click Search>Local Save File, select the file created in step 1, click OK.
    b. Select records in a list (To select adjacent records, click to highlight the first entry, and then press and hold <Shift> and click the last entry. Or to select nonadjacent entries, click to highlight the first entry, and then press <Ctrl> and click to highlight each additional entry.), on the Action menu, click Export.
    c.  On the Batch menu, click Process Batch.
    d.  Check the box next to the Local File you want to export the records from.
    e.  In Process area, check Export.
    f.  Click OK.

6. Update holdings on OCLC

    a. On the Cataloging menu, click Search>Local Save File, click OK.
    b.  Select records in a list (To select adjacent records, click to highlight the first entry, and then press and hold <Shift> and click the last entry. Or to select nonadjacent entries, click to highlight the first entry, and then press <Ctrl> and click to highlight each additional entry.).
    c. On the Action menu, click Holdings>Update Holdings.

7. Undo batch processing settings in Connexion.

Some of the Connexion batch processing settings will "stick" from session to session, so it's important to undo those settings before resuming single-record processing.
    a. On the Tools menu, click Options > Batch.
    b. In the Record Actions area, under Perform local actionsin batch, uncheck Bibliographic Record Export. (This changes the export setting back to exporting each record one at a time.)
    c. On the Tools menu, click Options > Export.
    d. In the Destination area, select Gateway Export: 130.111.64.3(5500). (This changes the export destination back to the URSUS server instead of a locally saved DAT file.)

II. Batch Editing in MarcEdit

1. Open the file in MarcEdit.
   • Open MarcEdit.
   • Under MARC Tools, select MarcBreaker.
   • Under Input File, click browse to navigate to the folder where the data file created in step I.4 is saved.
   • The Open File window searches for .mrc files by default, so the .dat file will not be visible. In the drop-down menu at the bottom right of the window, select All Files *.*.
   • Select the data (.dat) file created in step I.4 and click Open.
   • Under Output File, click browse to navigate to the folder where the data file is saved.
   • By default, the new file will have the same name as the old file, but with a .mrk suffix. Click Save.
   • Click Execute. Pay attention to the number of records listed in the Results box, which should match the number of records saved in the original .txt file.
   • Click Edit Records.
2. Check that all records have call numbers and all call numbers are complete.
   • Use the Find command to get a list of all call numbers by going to Edit > Find, typing =050, and clicking Find All.
   • Scroll through the list to check for incomplete or missing call numbers. Incomplete call numbers are usually missing a Cutter and year, which makes them noticeably shorter in length than complete call numbers.
   • To go to the record to complete the call number, highlight the incomplete call number and click on Jump to Page.
   • Complete the call number by adding the $b, the Cutter, and the year. Duplicate 050s can be deleted.
   • Continue until all call numbers are complete.
   • Repeat the Find All search to check that every record has a call number. If the number of 050s returned equals the total number of records, all records have a call number. If the number of 050s is less than the total number of records, one or more records is missing a call number. If there are missing call numbers, scroll through the records in the Find All Results window, to see which record has been skipped. Add new call numbers until all records have a call number.
3. Append "eb" to the end of call number
   • In MarcEdit, go to Tools > Edit Subfield Data
   • Enter the following information:
     Field: 050
     Subfield: b
     Field date: ^e (This is a special command that tells MarcEdit to append the new text to the end of the subfield instead of replacing the text currently in the subfield.)
     Replace with: eb
   • Some call numbers may have already ended in eb, in which case they will now end in ebeb. To correct this, go to Edit > Replace. In Find What, type ebeb. In Replace With, type eb. Click Replace All.
4. Filter out unwanted 856 fields
   • Change all 856 fields to 999 fields, replace =856 with =999
   • Change all 999 fields with the Springer URL back to 856 fields, replace =99940$3SpringerLink$uhttp://dx.doi.org/10.1007/ with =85640$uhttp://www.library.umaine.edu/auth/EZProxy/test/authej.asp?url=http://dx.doi.org/10.1007/
   • Add $zUM System-wide Access display notes:
     In Tools > Edit Subfield Data, enter the following information:
     Field: 856
     Subfield: u
     Field date: ^e
     Replace with: $zUM System-wide Access
5. Convert the file back to a MARC file.
   • Save the MarcEdit file (if you have not done so already) and close the file for convenience.
   • Under MARC Tools, click on MarcMaker.
   • Under Input File, click browse to navigate to the folder where the .mrk file is located and select the file.
   • Under Output File, click browse to navigate to the folder where the .mrk file is saved.
   • By default, the new .mrc file will have the same name as the old .mrk file, but with an .mrc suffix. Click Save.
   • Click Execute.

III. Load Records to Millennium through Data Exchange

1. Keep the prefix "ocm" as is in 001 fields
2. Load profile for Springer ebooks is Load I
3. For loading steps refer to Data Exchange Procedure

Important Notes

• Always run a couple of tests before processing a larger scale batch, especially before the batch exporting.
• Springer updates its OCLC Number List several times per month. For convenience, we load records monthly.
• Sample URL already in URSUS:  http://130.111.64.3/record=b4260756

Return to the Technical Services Page.

---

Created by: Kang Cao | Revised: 03/19/2013