Library Technical Services

Bulk Loading: Instructions (LTS Procedure #46)

Scope: This procedure provides instructions for running automated import, export, cleanup, reporting and various related jobs using shell scripts and other tools created by the Library Systems Office. The list of jobs includes the frequency of the work, the list of people responsible for performing this work, as well as the function of each job. In addition, this procedure includes guidelines for scheduling batch routines.

Contact: Jena Bakula

Unit: Database Management Services

Date last updated: 6/5/06

Date of next review: July 2006

General Instructions
List of Jobs
Job-Specific Instructions
Scheduling

A - B - C - E - H - I - L - M - N - O - P - R - S - Y

Job-Specific Instructions

• ACLS
  • Run this job whenever a new file of ACLS records becomes available.
  • Save the new file in \\Library30\input\vendorRecords\ACLS
  • Type "acls_load <filename>" where <filename> is the name of the file to be loaded.
    • Example: If the target file is called acls3b.marc, the command to load this file will be: acls_load acls3b.marc

• ANNEX LOCATION REPORT (Suspended)
  • Run this job monthly on the first business day.
  • Type "annex_reports".

• ASIA Q-V CALL NUMBER CLEANUP
  • Run this job monthly on or about the 20th.
  • Type "asiaqv run ".
  • FTP extracted files to PC and use NWU's Vger Location Change program to update MFHD and item records in each of the three files.

• AUTHORITY RECORDS: NAMES AND SUBJECTS
  • Run these jobs weekly or whenever new NAF and SAF files are available. Watch for the "LC files" memos from Peter Ward on CTSBULK-L.
  • Type "auth <filename>" where:
    • <filename> is the name of the file to be loaded. Files are numbered sequentially using the designations "unname" or "unsub" followed by a two-digit year and a two-digit week number.
    • Examples: If the previous bulkloaded NAF file was called unname00.17 (i.e. unspanned name authorities for the year 2000, week 17), the command to load the next file will be: auth unname00.18. If the previous bulkloaded SAF file was called unsub00.19, the command to load the next file will be: auth unsub00.20
  • To view discarded or rejected records:
    • Type "cd <directory name>" to open the appropriate file directory on Library9
    • Type "ls" to list the files in that directory
    • Type "more <file name>" to retrieve the records
    • Example: If the file containing the "Discards & Rejects DATA" is called: /usr/local/batch_data/Auth.014.0808/not_loaded.mrc, the commands to view discards and rejects will be:
      • cd /usr/local/batch_data/Auth.014.0808
      • ls
      • more not_loaded.mrc
  • Monitor the Global Headings Change queue. As needed, run PCat Jobs 11 & 12 to expand entries in the queue and PCat Job 13 to perform global headings changes (names only). The commands to run these jobs are, respectively: ghc11.pl, ghc12.pl, and ghc13.pl
  • Perform manual cleanup on remaining changed and deleted headings. Give Chinese, Japanese, and Korean work to the PCS CJK Assistant or CJK student.

• BATCH DELETE
  • Run this job on the first Tuesday of the month. Do not run on the same day as OCLC Export.
  • Type “batch_delete <Date 1> <Date 2> <RLIN file name>” where:
    
    • <Date 1> is the first date covered by the file, entered in the form YYYYMMDD. Example: 20021125
    • <Date 2> is the last date covered by the file, corresponding to the last Sunday preceding the run, entered in the form YYYYMMDD. Example: 20030105
    • <RLIN file name> is the name of the file to be sent to RLIN. These file names contain a "V" prefix and are numbered sequentially within the same scheme used for RLIN Export. Example: V258
    • Example: If the previous Batch Delete job covered records through Sunday, Jan. 5, 2003 and the previous RLIN export file was numbered V258, the command to run Batch Delete on Tuesday, Feb. 4 will be: batch_delete 20030106 20030202 V259
  • CTS will use BatchCat to delete Voyager records, when applicable, if there are 50 or more. If there are less than 50, DBQ&E will delete these records manually. To access via F-Secure the Voyager BIB record IDs for manual deletion:
    • Type "cd <directory name>" to open the appropriate file directory on Library9
    • Type "ls" to list the files in that directory
    • Type "more <file name>" to retrieve the BIB record IDs
    • Example: If the file containing the "Bib keys for CTS deletion using BatchCat" is called: /usr/local/batch_data/BDelete.008.1352/leader_d.L, the commands to retrieve the BIB record IDs will be:
      • cd /usr/local/batch_data/BDelete.008.1352
      • ls
      • more leader_d.L

• CASALINI ENHANCED / CASALINI FULL
  • Run this job whenever new files become available.
  • For CASALINI ENHANCED: Type “vendor_load casplus <filename> <filename> <filename><filename> <filename>" where filename is the specific name of the file to be loaded.Use lower-case letters in all file names. Note that up to five files can be handled with one command.
  • For CASALINI FULL“vendor_load casfull<filename> <filename> <filename><filename> <filename>"
    • Example: If the target files are called NYRA40b.083 and NYRA40g.055, the command to load these files will be: vendor_load casplus nyra40b.083 nyra40g.055 (use casfull for CASALINI FULL loads)
  • Retrieve the "Titles Not Loaded", if any, from the Casalini folder on the Library 30.
• COPY NUMBER CLEANUP
  • Run this job monthly on or about the 15th.
  • Type "copynum".
  • FTP extracted files to PC and use BatchCat ChangeItemNumber.exe program to update item records.

• EBRARY
  • Run this job whenever a new file of EBRARY records becomes available.
  • Save the new file in \\Library30\input\vendorRecords\Ebrary
  • Type "ebrary_load <filename>" where <filename> is the name of the file to be loaded.
    • Example: If the target file is called ebrary5f.marc, the command to load this file will be: ebrary_load ebrary5f.marc

• HARRASSOWITZ
  • Run this job whenever new files become available. There are two kinds of files: approvals and standing orders.
  • Type “vendor_load harrassowitz <filename> <filename> <filename><filename> <filename>" to load approvals and “vendor_load harrass_so <filename> <filename> <filename><filename> <filename>" to load standing orders, where filename is the specific name of the file to be loaded. Use lower-case letters in all file names. Note that up to five files can be handled with one command.
    • Example: If the target files are called 20020920053152.24.invoices.marc21.abc and 20020924414617.24.invoices.marc21.abc, the command to load these files will be either: vendor_load harrassowitz 20020920053152.24.invoices.marc21.abc 20020924414617.24.invoices.marc21.abc or vendor_load harrass_so 20020920053152.24.invoices.marc21.abc 20020924414617.24.invoices.marc21.abc
  • Retrieve the "Titles Not Loaded", if any, from the Harrassowitz folder on the Library 30.
• IBERBOOK / ITURRIAGA
  • Run these jobs whenever new files become available. Iberbook and Iturriaga activity is divided into two loading sequences. The first sequence takes a file with many invoices and splits the records into separate files indicating the invoice number as part of the file name. The second sequence is executed using the "iberload" or "itturload" command.
  • Step 1: type "vendor_load iberbook <filename>" or “vendor_load itturiaga <filename>"where filename is the specific name of the file to be split into separate files, one for deach invoice. These split files are placed in the Iberbook or Itturiaga folder, ready for the second loading sequence.
    • Example: If the Iberbook target file is called marc15821.001, the command to load this file will be: vendor_load iberbook marc15821.001
    • Example: If the Iturriaga target file is called pe030811.mrc, the command to load this file will be: vendor_load itturiaga pe030811.mrc
  • Step 2: type “vendor_load iberload <filename>" or "vendor_load itturload <filename>"where filename is the specific name of the file to be loaded into Voyager.
    • Example: If the Iberbook target file is called marc15821.split_marc.35868, the command to load this file will be: vendor_load iberload marc15821.split_marc.35868
    • Example: If the Iturriaga target file is called pe030811.mrc.split_marc.14341, the command to load this file will be: vendor_load itturload pe030811.mrc.split_marc.14341

• INITIAL ARTICLE / TITLE TAG CLEANUP
  • Run this job on the first Wednesday of the month.
  • Type "initial_article <Date 1> <Date 2>" where:
    
    • <Date 1> is the earliest date in the target range, entered in the form YYYYMMDD. Example: 20030801
    • <Date 2> is the latest date in the target range, entered in the form YYYYMMDD. Example: 20030831
    • Example: To run the job for August 2003, the command will be: initial_article 20030801 20030831
  • Download the reports for RLIN and individual language cleanup into Excel files and distribute to appropriate staff for manual cleanup.
    
    
• ISBN EXTRACT
  • To create a file of ISBN numbers from a file on the Library 30, type: “isbn_extract <filename> <jobname>" where filename is the specific name of the target file and jobname is the type of job you wish to run.
  • Example: To extract IBSNs from an RPS file called Corn0715.mrc, the command to extract the ISBNs will be: isbn_extract Corn0715.mrc rps

• LC RESOURCE FILE
  • Run this job weekly after receiving the two "LC files" memos from Peter Ward on CTSBULK-L.
  • Load the following five types of files from Peter Ward into the LC Resource File: cnbook, unmap, unmus, unser, unvis. Do not load unname and unsub files (these are authority records).
    
  • Type "lcload <filename> <filename> <filename><filename> <filename>" where filename is the specific name of the file to be loaded. Note that up to five files can be handled in the weekly dataload.
    
    • Example: If the target files are called cnbook03.01 and unser03.01, the command to load these files will be: lcload cnbook03.01 unser01.01
    • Example: If the target files are called cnbook03.04, unmap03.01, unmus03.01, unser.01.04, and unvis.01.01, the command to load these files will be: lcload cnbook03.04 unmap03.01 unmus03.01 unser03.04 unvis03.01
• LOST/MISSING WITHDRAWALS REPORT
  • Run this job on the first workday of the month.
  • Type "lostmissing <Date 1> <Date 2>" where:
    
    • <Date 1> is the earliest date in the target range, entered in the form YYYYMMDD. Example: 20030301
    • <Date 2> is the latest date in the target range, entered in the form YYYYMMDD. Example: 20030331
    • Example: To run the report for withdrawals done in March 2003, the command will be: lostmissing 20030301 20030331
  • Use MS Access and Snap software to reformat the report, then forward to Howard Brentlinger in IRIS, Collection Services.

• MARCADIA
  • Run this job monthly.
  • The MARCadia process consists of eight steps: (1-4) extracting records for the current month -3, the current month -18, the current month -12, and the current month -24, and (5-8) reloading records for the current month -3, the current month -18, the current month -12, and the current month -24.
  • For the extracts, type "marc_export <Date 1> <Date 2> <MARCadia filename> <Extract Type>" where:
    • <Date 1> is the first create/update date of the monthly extract, entered in the form yyyymmdd. Example: 20021001
    • <Date 2> is the last create/update date of the monthly extract, entered in the form yyyymmdd. Example: 20021031
    • <MARCadia filename> is the name of the file to be sent to MARC Link. Name the file based on the date of the extract (YYMMDD) followed by the suffix "A". If more than one file is extracted on the same day, use B, C, D, etc. as the suffix. Example: If a file is extracted on Dec. 13, 2001, call it: 011213A. If a second file is extracted on the same day, call it: 011213B.
    • <Extract Type> is a switch to indicate which set of Voyager records are targeted. Type "reg" to retrieve backlog, CIP, and COR records for the current month -3 and current month -18 ranges. Type "supp" to retrieve COR records only for the current month -12 and current month -24 ranges.
    • Example (extract commands): In February, 2003, four extracts are run. The target months are February and August 2001, and February and November 2002. If the first and second extracts are run on February 12 and the third and fourth extracts are run February 13, the commands for the four jobs will be:
      • marc_export 20010201 20010228 030212A supp
      • marc_export 20010801 20010831 030212B reg
      • marc_export 20020201 20020228 030213A supp
      • marc_export 20021101 20021130 030213B reg
  • Watch for the "Marcadia NYCX Results Notification" memo on CTSBULK-L, which will tell us when files are available for reload. Reload the "Match file."
  • For the reloads, type "marc_reload <local path/filename>" where:
    • <local path/filename> is the name of the folder in the "Cornell" directory of the MARC Link server in which the returned "Match file" has been placed, followed by the filename. The name of the folder will normally be the same as the name of the match file, but without the suffix. If the FTP fails, check the MARC Link server to verify the folder and file names.
    • Example (reload commands): When the files exported in the extract example above are returned, they will probably be called "030212A.M", "030212B.M", "030213A.M", and "030213B.M". The commands for the four reloads will be:
      • marc_reload 030212a/030212A.m
      • marc_reload 030212b/030212B.m
      • marc_reload 030213a/030213A.m
      • marc_reload 030213b/030213B.m

• MARCIVE
  • Watch for the "Data Ready for CITH" memo on CTSBULK-L, informing us that a new monthly file is available for loading. Run the load on either Tuesday or Thursday afternoon.
  • Following the instructions in the e-mail, FTP the file from the Marcive server to the \input\vendorRecords\Marcive\monthly_input folder on the Library 30. The data file must have a unique name.
  • To load the records, type "marcive_import <filename>" where filename is the specific name of the file to be loaded.
  • Use BatchCat to create MFHDs.
  • After the load is complete, FTP the following discard and reject files for manual cleanup:
    • discard_pre.mrc
    • setB.mrc
    • serD.mrc
    • marcive-rejects
  • FTP the dead_serials.L file and use the list of BIB IDs to identify records to be updated.
  • After manual cleanup and record updating is complete, run the cleanup program that identifies unwanted duplicate records. Type "marcive_cleanup run". Start this job in the early morning.
  • Retrieve the file delete.list and use BatchCat to delete the records.
  • Retrieve the file problem.list and resolve ambiguous matches manually.

• NETLIBRARY
  • Run this job monthly or whenever OCLC makes new files available. There are two kinds of files: owned and unowned. Records "purchased by our PDA" are owned. Records "added to our PDA" are unowned.
  • Follow the instructions in the e-mail announcement to retrieve the file from OCLC. FTP the file to the Vendor Records / NetLibrary folder on the Library 30.
  • In the F-Secure client, type “netowned <filename>” to load owned records and “netunowned <filename>” to load unowned records, where <filename> is the name of the file to be loaded.
    • Example): If the target file is called D030409.B7670.RECORDS.bin, the command to load this file will be either: netowned D030409.B7670.RECORDS.bin or netunowned D030409.B7670.RECORDS.bin
  • Clean up new records added during netowned loads (most will be duplicates). Resolve discards and rejects manually.

• OACIS
  • Run this job quarterly (February, May, August, November) on the first Wednesday of the month.
  • Type "oacis_ex run".

• OCLC CJK (Suspended Nov. 2004)
  • Run this job monthly or whenever OCLC CJK files are available. Watch for the "TechPro File available" memos from OCLC on CTSBULK-l.
  • Type “cjk <OCLC filename> <RLIN filename> <OCLC date range>” where:
    
    • <OCLC filename> is the name of the filename to be loaded. Example: coo1227 [will load coo1227.mrc]
    • <RLIN filename> is the name of the file to be transferred to RLIN. Monthly loads are numbered sequentially using a "V" prefix. Example: V23
    • <OCLC date range> is a 16 digit number that is passed to RLIN by the program. The first eight digits represent the first date covered by the file in the format YYYYMMDD; the second eight digits represent the last date covered by the file using the same format. Example: 2001011320010214
    • Example: If the previous bulkloaded OCLC CJK file was called coo0214.mrc and its assigned RLIN file name was V22, the command to load a new file called coo0228.mrc will be: cjk coo0228 V23 2001021520010228
• OCLC EXPORT
  • Run this job on Monday mornings. Do not run on the same day as Batch Delete or a day after Batch Delete.
  • Type "oclc_ex <Date 1> <Date 2>" where:
    
    • <Date 1> is the first create/update date of the weekly extract, entered in the form dd-mmm-yy.
      Example: 19-jun-00
    • <Date 2> is the last create/update date of the weekly extract, entered in the form dd-mmm-yy.
      Example: 25-jun-00
    • Example: If the previous exported file covered the period from Monday, June 19, 2000 thru Sunday June 25, 2000, the command to load the next week's file will be: oclc_ex 26-jun-00 02-jul-00
• OCLC PASSWORD
  • Change our OCLC EDX password every 8 weeks on Wednesday morning (1/7/04, 3/31/04, 6/23/04, etc.).
  • Open an FTP session for the OCLC EDX client:
    Host Name/Address: edx.oclc.org
    Host Type: IBM MVS
    User ID: TCOO1
  • In the Password box, type: "[current password]/[new password]/[new password]".
  • Example: If the the current password is D4E5F5 and the new password will be D4E5F6, type: D4E5F5/D4E5F6/D4E5F6
  • Verify that new password works properly.
  • Send new password to Peter Hoyt. He will update the FTP tables for our OCLC batch jobs (OCLC Export, Batch Delete, Recon, etc.)

• OKU WITHDRAWALS COUNT
  • Run this job on the first Thursday of the month.
  • Type “withdrawn <Date 1> <Date 2>” where:
    
    • <Date 1> is the first date covered by the file, entered in the form YYYYMMDD. Example: 20030501
    • <Date 2> is the last date covered by the file, entered in the form YYYYMMDD. Example: 20030531
    • Example: To run the report for May 2003, the command will be: withdrawn 20030501 20030531

• OPAC TITLE COUNT
  • Run this job quarterly (February, May, August, November) on the first Wednesday of the month.
  • Go to Voyager Harvest Request
  • In the Requester NetID box type sb342, and in the Job Name box type: BIBCount
  • Click on Copy, type your ID and the new name of the job, click on Copy and on the main screen click on the Resubmit button.
  • Forward the E-mail with the results to Ida Martinez (im58).

• PROQUEST UPDATES(Suspended)
  • Run this job whenever a new file of ProQuest records becomes available.
  • Type “proquest <filename>” where filename is the specific name of the file to be loaded:
    • Example: If the ProQuest file is called ProQuest20040322.rec, the command will be: proquest ProQuest20040322.rec
  • Use BatchCat to create and update MFHDs and to delete obsolete records:
    • pqdupdate.pl add_mfhds.txt add username password prod
    • pqdupdate.pl update_mfhds.txt update username password prod
    • delete_records delete_pq_obsolete.txt maxnumber username password
  • After the BatchCat work is complete, update the PURLs for the loaded BIB records on the PURL server.

• PURL UPDATES (Suspended)
  • Note: this program is currently used only in conjunction with ProQuest Updates.
  • Using F-Secure SSH or Telnet, logon to the batch account on Copia:
    Host Name: copia.library.cornell.edu
    Userid: ctsbulk
    Password: ******
    
  • Type "sudo batch_purl replace".
  • When job is complete, type "sudo batch_purl add".

• RECON
  • The Recon process consists of three steps: (1) loading the monthly file of new records from OCLC; (2) extracting the smart barcode data, including the creating a file of empty item record IDs and barcode numbers, approximately 6 weeks after the dataload; and (3) sorting, formatting, and shipping the smart barcode data to the barcode vendor.
  • Phase 1: loading the records.
    • Watch for the "Cornell University (COO) Retrocon" memo on CTSBULK-L, informing us that a new monthly file is available for loading. Run the load on the first late Thursday afternoon following receipt of this notification.
    • To load the records, type "recon_load <File Name> <Barcode Number>" where:
      • <File Name> is the unique portion of the name of the file to be loaded, followed by the "cc" suffix. Example: v02m10.cc
      • <Barcode Number> is the next 7-digit barcode number in the sequence reserved for Bulk Import and BatchCat. Example: 2892313
      • Example: If the file name that appears in the notification is ftp.v02m10.cc / BIB and the last barcode used for Recon Bulk Import was [31924]2892312[X] , the command to load the records will be: recon_load v02m10.cc 2892313
    • After the dataloads are complete, verify the record count and send e-mail confirmation to OCLC that we have retrieved the files.
    • Use BatchCat to create MFHDs and item records for the Set B BIB records.

  • Phase 2: extracting the smart barcode data.
    • Run this job monthly, approximately 6 weeks after the Phase 1 dataloads are complete.
    • Type "smart_barcode_extract <Date> <Barcode Number>" where:
      • <Date> is the date of the corresponding Phase 1 Bulk Import dataload, entered in the form YYYYMMDD. Example: 20021014
      • <Barcode Number> is the next available barcode number reserved for Bulk Import and BatchCat. Example: 3231010
      • Example: If the Phase 1 Bulk Import dataload took place on Oct. 14, 2002 and the next available barcode number reserved for Bulk Import and BatchCat is 3231010, the command to extract the smart barcode data will be: smart_barcode_extract 20021014 3231010
    • Use BatchCat to map barcode numbers into the empty item records in Set D.

  • Phase 3: exporting the smart barcode data.
    • Run this job monthly after the Phase 2 extract and BatchCat processing of Set D records are complete.
    • Type "smart_barcode_format <Path> <Barcode 1> <Barcode 2> <Barcode 3> <Barcode 4>" where:
      • <Path> is the variable portion of the SetABD file name from the Phase 2 extract. Example: SmartBC.340.1233
      • <Barcode 1> is the first barcode number in the Set AB dataload range. Example: 2892313
      • <Barcode 2> is the last barcode number in the Set AB dataload range. Example: 2903959
      • <Barcode 3> is the first barcode number in the Set D extract range. Example: 3231010
      • <Barcode 4> is the last barcode number in the Set D extract range. Example: 3232855
      • Example: If the SetABD file name from the Phase 2 extract is SmartBC.340.1233/SetABD.srt, and the barcode ranges from the Phase 1 dataload and the Phase 2 Set D extract are, respectively, 2892313-2903959 and 3231010-3232855, the command to export the smart barcode data will be: smart_barcode_format SmartBC.340.1233 2892313 2903959 3231010 3232855
    • Within a day or two after they receive the data, the vendor (Data2) will fax barcode samples and an "Authorization for Production" to Jim LeBlanc. Give the samples to Sally Lockwood for approval. Zoe Stewart-Marshall will normally sign the authorization form, but Jim LeBlanc may sign in her absence. Fax the authorization back to Data2 to initiate the barcode run. Barcodes will be shipped to Sally Lockwood.

• RLIN EXPORT
  • Run this job on Monday afternoons.
  • Type "rlin <Date 1> <Date 2> <RLIN filename>" where:
    
    • <Date 1> is the first create/update date of the weekly extract, entered in the form dd-mmm-yy.
      Example: 19-jun-00
    • <Date 2> is the last create/update date of the weekly extract, entered in the form dd-mmm-yy.
      Example: 25-jun-00
    • <RLIN filename> is the name of the file to be sent to RLIN. These file names contain a "V" prefix and are numbered sequentially within the same scheme used for Batch Delete. Example: V22
    • Example: If the previous exported file covered the period from Monday, June 19, 2000 thru Sunday June 25, 2000 and the last assigned RLIN file name was V22, the command to load the next week's file will be: rlin 26-jun-00 02-jul-00 V23
  • Clean up problem records in Voyager based on automated "RLIN Export Summary" and "Bad Character" reports and on weekly error reports from RLG.

• RPS (RUSSIAN PRESS SERVICE)
  • Run this job whenever new files become available.
  • Extract ISBNs from the RPS records.
  • Search the resulting "isbn.txt" file against OCLC using CatME.
  • Merge the OCLC records into the original RPS file and split the merged file according to invoice number by typing: "rps <RPS filename> <OCLC filename>" where RPS filename is the name of the original file and OCLC filename is the name of the .dat file retrieved by the CatME search.
    • Example: If the RPS file is called Corn0715.mrc and the OCLC file is called Corn0715.dat the command to merge the files and split the merged file by invoice number will be: rps Corn0715.mrc Corn0715.dat
  • Load the files into Voyager by typing: "vendor_load rpsload <filename> <filename> <filename> <filename> <filename>" where filename is the specific name of the split file to be loaded. Note that up to five split files can be handled with a single command.
    • Example: If the merge process resulted in split files called Corn0715.mrc.split_marc.mb032123, Corn0715.mrc.split_marc.mb032124, and Corn0715.mrc.split_marc.B03A1161, the command to load the files into Voyager will be: vendor_load rpsload Corn0715.mrc.split_marc.mb032123 Corn0715.mrc.split_marc.mb032124 Corn0715.mrc.split_marc.B03A1161
  • Retrieve the "not-loaded" titles, if any, from the RPS folder on the Library 30 for manual processing.
  • Clean up the RPS folder to reduce confusion.
• SEARCHING ISBNs AGAINST OCLC

• SERIAL SOLUTIONS UPDATES
  • Run this job bimonthly.
  • Generate sleek records from Serial Solutions data feed --
    • Import the latest Serial Solutions data into: \\Library30\elicense\trackingDatabase.mdb
    • Update the tblTrack administrative metadata within trackingDatabase.mdb
    • Run ssbulk.pl
    • Copy output file to: \\Library30\input\vendorRecords\SerialSolutions
  • On Library9, type: “serial_solutions <filename>” where filename is the specific name of the file to be loaded:
    • Example: If the Serial Solutions file is called 20040830.mrc, the command will be: serial_solutions 20040830.mrc
  • Use BatchCat to create and update MFHDs and to delete obsolete records:
    • Add new MFHDs to new records based on IDs in output in the "new_bibkeys.txt" report using ss_hlds.pl
    • Update MFHDs in existing records based on IDs in output in the "mfhddupd.txt" report using ss_hlds.pl
    • Pass "dupes.mrc" and "delete.txt" to Nate Rupp for manual updates. When work is complete, Nate will return the delete.txt report for automated deletion.
    • Pass file of eJournals with double 245 fields to David Banush for manual cleanup by a student.

• SERIES NOT STARTED FLIP
  • Run this job on the first Wednesday of the month.
  • Type "490_flip <Date 1> <Date 2>" where:
    • <Date 1> is the earliest date in the target range, entered in the form YYYYMMDD. Example: 20030301
    • <Date 2> is the latest date in the target range, entered in the form YYYYMMDD. Example: 20030331
    • Example: To flip the 490 0 fields for all cataloging done in March 2003, the command will be: 490_flip 20030301 20030331
  • FTP the "035.L" and "490.L" discard files (if any records appear there) and the "Report of MARC 830 fields." Open using Excel and perform manual cleanup, as appropriate.

• YANKEE
  • Run this job on Monday mornings.
  • Type “yank <mmddyy>” where <mmddyy> is the 2nd half of the filenames to be loaded.
    • Example: If the target file is called YANK050801.dat, the command to load this file will be: yank 050801