Migration from CEPAL to MARC

Author: Marcos Mirabal Garcia (Universidad de Pinar del Río) System: ABCD (Automation of Libraries and Documentation Centers)

This guide describes the procedure to migrate a serials database from the CEPAL format (common in Latin America) to the MARC format, which is widely used in ABCD.

1. Analysis and Field Mapping

The first and most critical step is to map the fields from the source format (CEPAL) to the destination format (MARC). Below is the mapping table used in this example.

CEPAL Field	Field Name	MARC Field
2	Control Number	1
3	Topographic Signature	84
16	Personal Author (M)	100
17	Corporate Author (M)	110
18	Title (M)	245
19	Translated Title (M)	242
20	Pages	300
38	Publisher	260
39	City	260
40	Country	43
41	Edition	250
43	Year	260
47	ISBN	20
64	Text Language	41
72	Notes	500
76	Topical term	650
104	Location	852

2. Creating the Migration Script

Based on the mapping above, you need to create a Procedure Script (.prc). This script uses the CISIS formatting language to transform the data during the process.

Create a file named migration.prc with the following content:

/* Delete all existing fields first */
'd*',

/* Add fixed MARC Leader fields (Tags 3000+) */
'<3005>n</3005>',
'<3006>a</3006>',
'<3007>m</3007>',
'<3008>0</3008>',
'<3017>5</3017>',
'<3018>a</3018>',
'<3019>0</3019>',

/* Migrate Data Fields */
/* If field exists in CEPAL (p), wrap content in MARC tags */

if p(v2)   then '<1>',v2,'</1>' else '<1>',mfn,'</1>' fi,
if p(v3)   then '<84>^a',v3,'</84>' fi,
if p(v16)  then '<100>^a',v16,'</100>' fi,
if p(v17)  then '<110>^a',v17,'</110>' fi,
if p(v18)  then '<245>^a',v18,'</245>' fi,
if p(v19)  then '<242>^a',v19,'</242>' fi,
if p(v20)  then '<300>^a',v20,'</300>' fi,
if p(v38)  then '<260>^b',v38,'</260>' fi,
if p(v39)  then '<260>^a',v39,'</260>' fi,
if p(v40)  then '<43>^a',v40,'</43>' fi,
if p(v41)  then '<250>^a',v41,'</250>' fi,
if p(v43)  then '<260>^c',v43,'</260>' fi,
if p(v47)  then '<20>^a',v47,'</20>' fi,
if p(v64)  then '<41>^a',v64,'</41>' fi,
if p(v72)  then '<500>^a',v72,'</500>' fi,
if p(v76)  then '<650>^a',v76,'</650>' fi,
if p(v104) then '<852>^a',v104,'</852>' fi

3. Executing the Migration

The migration is performed using the mx utility via the command line.

Preparation

1. Create a working folder.
2. Copy your database files (e.g., cepal.mst, cepal.xrf) to this folder.
3. Copy the migration.prc file created in Step 2 to this folder.
4. Ensure you have the mx executable (compatible with your database version) accessible.

Step-by-Step Commands

A. Prepare the Source Database

If your source is an ISO file, import it first.

Windows:

mx iso=source.iso create=cepal -all now

Linux:

./mx iso=source.iso create=cepal -all now

B. Run the Migration

Apply the procedure script to convert records from CEPAL to MARC. This creates a temporary database (e.g., temp_marc).

Windows:

mx cepal proc=@migration.prc create=temp_marc -all now

Linux:

./mx cepal proc=@migration.prc create=temp_marc -all now

C. Export to ISO for ABCD Import

Export the converted database to an ISO file. We use outisotag1=3000 to preserve the standard MARC leader fields generated in the script.

Windows:

mx temp_marc iso=marc_import.iso outisotag1=3000 -all now

Linux:

./mx temp_marc iso=marc_import.iso outisotag1=3000 -all now

4. Final Import into ABCD

1. Access your ABCD interface.
2. Create a New Database (e.g., Books) using the standard MARC structure.
3. Go to Utilities > Import ISO.
4. Select the marc_import.iso file generated in Step 3C.
5. Run the import.

You have now successfully converted a CEPAL database structure into a MARC database structure. This logic can be adapted to migrate between any other formats by modifying the mapping table and the .prc script.