Self-PLP, MCU, and duplicate records §

The primary purpose of Self-PLP is to resolve duplicates in the incoming file before launching PLP to search the union catalog for matching records. However, it is also possible to use Self-PLP … 

  • to organize a file of duplicate records for viewing in MCU, or  
  • to use the powerful PLP Match Rules to identify duplicate records in a standalone file 

Follow the steps on this page to view a file of duplicate records in MCU.
 

If you want to know what 'PLP' is, please read the 'Overview' (first link on the left).
If you want to know a little more about 'MCU', use the 'MCU' link on the left.
 


By design, MCU has no logic for identifying matching records. In addition, MCU requires every record of the input file to contain a special synchronization field. Fortunately, Self-PLP is able to provide both of these requirements. 

MCU can be used to view a file of duplicate records if the following pre-requisites are met: 

  1. the records to be compared must be MARC21 
  2. the records to be compared must be in the same physical file1) 
  3. the records to be compared must be linked by a common identifier, or, by what PLP calls a primary 'match key' 
  4. all records in the file must contain a holdings code 
  5. all records in the file must contain a system Id 

Note: the last two pre-requisites are usually performed by PLP's pre-processing module (the instructions for setting this up follow in 'Step. 1' below). 

If two (or more) separate files are to be compared, and the match key is in a different field in each file, the match key location will need to be standardized before concatenating the files and starting Self-PLP 2) 

The primary match keys supported by PLP and Self-PLP are: LCCN, ISBN, ISSN, OCLC number, Title, and SysId. The SysId key is derived from the 035 $a. Note that PLP includes a pre-processing option that makes it possible to copy a match key from another field into an 035 $a, effectively making the SysId key whatever you want it to be. 

1. Create a Library Load Profile §

Since PLP requires a library profile for all projects, the first step will be to open PLP, and select 'Library Load Profiles' from the Configuration menu. 

Follow the steps on this page to set-up the System ID section of a Library Load Profile. 

Follow the steps on this page to set-up the Holdings code section of a Library Load Profile. 

The third section of the library load profile ('Misc Options') can be left at its defaults for the task at hand. 

Note: The purpose of the library load profile is to enable PLP Pre-processing, which is normally the first task performed in all projects, to pass through the file and identify/resolve any problems with SysIds and Holdings codes before PLP begins. 

2. Create a set of Match Rules §

The next step is to configure a set of Match Rules. Start PLP, select 'Match Rules' from the Configuration menu; then, from the 'File' menu, select 'New'. 

Although the number of options on this page may seem bewildering, only three steps are needed for the task at hand. 

First, determine the match key ('Primary key') that will be used to organize the duplicate records. For example, if working with a file of records with duplicate 020$a's, select 'ISBN'. If working with a file that contains a match key that is not one of the standard library match keys, select 'CUSTOM'. Drag and drop the selected match key from the 'Primary Keys: ALL' box onto the box directly beneath it.  

Second, add the 'XXXXX' crosscheck to the config3). Click on 'XXXXX' in the 'Crosschecks: ALL' box (it should be the last item), and drag and drop it onto the 'Crosschecks' box directly beneath it. 

Finally, if necessary, adjust the 'Max Hits' limit for the primary key that you added above. The maximum possible value is '99'; this means there can be up to 99 records in each group of duplicate records before PLP will give up on the grouping process for the current record. 

Select the 'Save As' button and enter a new, descriptive name for this configuration. Something like 'Send all records to MCU' might be appropriate here. Respond 'Yes' to the prompts that follow (use the 'Yes to All' option when it appears).  

Then press 'Save' (and confirm the prompts again). 

3. Setup PLP §

Once the two configuration files have been created, start PLP (if necessary), and select 'Run Options' from the main menu. Make sure both of the options 'Skip Pre-processing' and 'Skip Self-PLP' are not checked, then select the 'Skip PLP' option.  

 

If you are using Self-PLP to setup a file for MCU, do not select either of the special 'Self-PLP only' run options (for details on these options, which were added in subversion 217, please see the note here 'Avoiding XC Fails in Self-PLP').  

No other options on this menu apply. 

Returning to the main PLP form, set-up the four required filenames. Use the '…' button to launch a navigation dialog and select each one.  

Note that each time you select a file, PLP will validate it. 

The four require files are: 

  1. the MARC ('Library') file that you wish to process for viewing in MCU,  
  2. the Library Load profile that you created for this file in Step #1 
  3. the Self-PLP Match Rules that you created for this task in Step #2 
  4. the 'PLP Match Rules'; for this option, select 'default tmq matchrules'4) 

4. Run the Job §

Press the 'Begin' button on the main PLP form. 

If there are errors in the configuration, you will be prompted to correct them. For example, if you forget to enter one of the four required filenames (or if one of the filenames has become invalid), a message will pop-up. Once the problem has been corrected, press 'Begin' again. 

PLP will now begin the Pre-Processing stage. When it is complete, a brief report will pop-up.  

 

Press OK to begin the next stage, which will be Self-PLP. 

If instead, you see a message like this– 

 

–click OK, click Cancel, and then from the Reports menu (on the main PLP form), select 'Pre-processing–Details'. This report should provide a hint as to what section of the library load profile needs to be changed in otder to avoid the problem. If you need help setting up a profile, please contact TMQ. 

When Self-PLP completes, another brief report will pop-up. Hopefully, the report will say that no records matched, no records went to 'No Match', and that all of the records failed crosschecks: 

 

If that's the case, then press OK, and PLP will ask you if you want to open the results in MCU. If you respond 'Ok', PLP will launch MCU, and then shut itself down.  

If you respond 'No', PLP will add an entry to its MCU Queue (available from the 'Run Options' menu), and advise that you can start the MCU session at a later time by accessing it from this menu. 


Once you have loaded the records into MCU, switch over to the MCU section of this site for further help. 

Here is a link to a sample project for Self-PLP. The steps on this page provide a fairly quick way to get an MCU session up and running after downloading a small MARC sample file, and all the required PLP config files. 


Repeating the PLP Setup steps 

After you have mastered the PLP setup tasks in section 3, above, you may avoid repeating them a second time by using the 'Quickset', or quick setup, list: 

  • Press <F8> to save the setup (Run Options|Save Quickset) 
  • Retrieve this setup in the future by pressing <F7> (Run Options|Load Quickset) 

The 'Quickset' saves the four filename parameters on the PLP form, plus the three 'Skip …' Run Options, to a plain text file5)

1) if necessary, use the MARC Report 'Concatenate Files' to join two or more MARC files together
2) this can be accomplished with a utility like MARC Global
3) This is a special crosscheck which forces all records that match on a primary key to 'fail crosschecks'; i.e., be sent to MCU
4) these are the match rules that would be used when PLP searches the union catalog database
5) the file is named '_quickset.txt' and is located in the program installation folder
self-plp/viewing_dupes.txt · Last modified: 2013/04/27 09:09 (external edit)
Back to top
CC Attribution-Noncommercial-Share Alike 3.0 Unported
Driven by DokuWiki Recent changes RSS feed