Data Migration

[eaquigley]

---

Author Name: Gustavo Durand (@scolapasta)
Original Redmine Issue: 3504, https://redmine.hmdc.harvard.edu/issues/3504
Original Date: 2014-02-03
Original Assignee: Kevin Condon

---

This ticket covers two primary use cases:

1. Providing real test data during development and testing.
2. Generating a final migration script for production cutover.

Some implementation issues that will need to be considered and most likely will become separate tickets are:
-differing validation rules between v3.6 and 4.0 where something not required in v3.6 is required in 4.0.
-different field sizes between v3.6 and 4.0 where the initial sizes in 4.0 are too small and need to be increased to accomodate v3.6 data, eg. dataset description.

---

Child issue(s): #121
Redmine child issue(s): 3530

---

[posixeleni]

@scolapasta @eaquigley For the Dataset Level Metadata Migration from 3.6-4.0 I have created the following table with mappings from 3.6-4.0. Anything highlighted in yellow needs to be reviewed and discussed before deciding on final migration https://docs.google.com/spreadsheets/d/1XYhiS3gFPs49LC2IhcHuHso7FuQ-ePPxP5nOOsleMks/edit?usp=sharing

CC/ @kcondon

P.S. New metadata fields that do not have a corresponding match in 3.6 are not included in this table.

[raprasad]

For the initial migration, should the dataverse 4.0 tables include extra fields such as:

("original" indicates Dataverse 3.6)

1. original_collection_id
2. original_study_id
3. original_study_verision_id
4. original_study_version_url

These types of fields may be used for checking data integrity, going back to check records, etc.

cc/ @kcondon @scolapasta