Electronic Theses and Dissertations

From UA Libraries Digital Services Planning and Documentation
Revision as of 15:17, 26 April 2010 by Jlderidder (Talk | contribs)

Jump to: navigation, search

ETDs 20100426.png

Workflow Overview

A. ProQuest uploads deposits of zip files to the content.lib.ua.edu server via ftp into the ftpaccess home directory, and notifies Janet Lee-Smeltzer of the upload. Janet notifies the Metadata Librarian to process them.

B. the Metadata Librarian runs the PERL script "moveContent" which is located in the scripts directory of her home area on content.lib.ua.edu. This script picks up all zip files sitting in the etd_deposits directory (which corresponds to teh ftpaccess home area), identifies the date the date the files were deposited, and relocates the files into a directory named for this date of deposit (yyyymmdd) in etd_deposits.

C. The Metadata Librarian then runs the PERL script "processEtds" which is located in her scripts directory. This script will ask which directory to process, and whether the files are for May, December, or September graduation, so it knows when to start the embargoes. It will create a subdirectory in the "working" directory which matches the name of the selected directory. Within this it will create 3 subdirectories: OPEN, PRQ, and CONTENT. OPEN is where it will open the zip directories. PRQ is where it will put the renamed and altered XML for processing. CONTENT is where it will put the renamed content files.

The "processEtds" script performs these tasks:

  1. extracts from each metadata file the following information:
    1. title,
    2. author,
    3. year the manuscript was completed
    4. year the degree was awarded
    5. embargo code (if any)
  2. calls the InfoTrack.bornDigital mysql database table on libcontent1.lib.ua.edu to find the next filenumber to assign and the InfoTrack.lookup table to determine the next persistent URL;
  3. records the item number assigned, the PURL assigned, the author and title in these tables
  4. inserts the assigned item number (filename, minus the extension) into a UA_identifier attribute and the assigned PURL into a UA_purl attribute within the DISS_submission field in a copy of the metadata
  5. places this altered copy of the metadata into an PRQ subdirectory; the copy will be named with the assigned filename followed by ".prq.xml" (thus a correctly named file would be: u0015_0000001_0000023.prq.xml) to indicate this is still ProQuest XML.
  6. copies all the bitstreams and renames them appropriately, placing them in a CONTENT subdirectory. The primary PDF will be named with the assigned filename followed by ".pdf"; subsidiary files will be numbered sequentially, with a 4-digit left-padded number attached to the assigned filename, followed by the extension. So the first subsidiary file for this file (if a jpeg) would properly be named u0015_0000001_0000023_0001.jpg, and the second (if a text file) would be named u0015_0000001_0000023_0002.txt.
  7. creates an entry for each record in a tab-delimited xmlList.xml file which contains the following fields:
    1. assigned filename
    2. original filename
    3. author
    4. title
    5. directory (created out of zip file name)
    6. year the manuscript was completed
    7. year the degree was awarded
    8. an indicator of the existence of subsidiary files (a count)
    9. the embargo code
    10. date item is made available via the web
    11. the assigned PURL

D. The Metadata Librarian works with the deposited content to create valid MODS files meeting our local profile, which include the assigned identifier and PURL, and are named for the assigned identifier with a ".mods.xml" extension.

E. The Metadata Librarian also creates valid MARC files for upload into our OPAC system, which reference the included assigned identifier and PURL.

F. She places the finished MODS in a MODS directory next to the PRQ and CONTENT directories.

G. The Metadata Librarian notifies Jody that the records are ready.

H. Jody copies the deposits to the Deposit subdirectory on the storage server, and runs a script which will move the files into the correct subdirectories for long-term storage, linking them into the LOCKSS manifests.

I. Jody extracts the filenames and embargo codes from the xmlList.xml, adds in the date the embargo starts, and calculates the end of the embargo dates. She then adds this information to a list or database entry that is checked by the periodic refreshing script.

J. The periodic refreshing script crawls through the storage directory, picks up new and modified files, checks for embargo dates not yet past, and if this raises no flags, copies the content and MODS to the web directories for online delivery. The periodic refreshing script will also note if an embargo date is past.

I. The Metadata Librarian submits a batch upload of the MARC records into our catalog system.

J. The Metadata Librarian checks the final display and access via both OPAC and digital library system to verify that no problems exist. If any problems are encountered, she contacts Jody and we work out how to fix them.  :-)

K. The Metadata Librarian will batch upload the MARC to WorldCat.

L. Jody will also set up a script that will check the database on the first of each month for embargoes which are due to lift that month; this script will email Jody and the Metadata Librarians with the filename, title, author, and date that the embargo is to lift.

M. The Metadata Librarian will then prepare and upload the no-longer-embargoed-content into the local OPAC and into WorldCat.

N. Should the metadata require remediation, the Metadata Librarian add a recordChangeDate field, and will upload the altered MODS files to libcontent1 and run relocate_all.pl. This will copy the files to the live web directory (except for those under embargo) and also to a deposits directory for archival storage.

O. Jody will transfer these into archival storage.

Reference: Find_our_content_online

Jody's process: Preprocessing ETDs

Personal tools