dump_and_anonymise script retains patient year of birth (unless rand date would be in future); shell script needs to pass src db name to anonymise.pl for skipping Genomics db unit_number anonymisation; output every 1000 patients rather than 100
dump_and_anonymise script retains patient year of birth (unless rand date would be in future); shell script needs to pass src db name to anonymise.pl for skipping Genomics db unit_number anonymisation; output every 1000 patients rather than 100
dump_and_anonymise script retains patient year of birth (unless rand date would be in future); shell script needs to pass src db name to anonymise.pl for skipping Genomics db unit_number anonymisation; output every 1000 patients rather than 100
new db anonymisation function; screen categories sorted lc (for new PNH/AA entry); fixed syntax error in mpla deployment script
9 years ago
README.txt
* does a mysqldump of hilis4/genomics/uclh/etc database and anonymises the patients + patient_case tables
* excludes some tables with patient info (eg patient_demographics - difficult/unnecessary to anonymise)
* generates a temporary db so required partition space will depend on size of uncompressed source db data dump (HILIS4 around 1GB)
* outreach & hmrn dbs can be directly exported via mysqldump as they have no patient data
patient_id is FK to hilis4.patients.id and request_id is FK to hilis4.requests.id
* expected usage is on a test/staging server with current clone of required db (LTH149890 or 'official' test HMDS server)
* usage: sh dump_and_anonymise_db.sh <cloned_db_to_anonymise> <db username> <db password>
