-- Browser(s) name and version -- : Chrome 156.0
Hi,
I'm setting up a mass action on the User table to remove inactive customer accounts for GDPR data-retention purposes: users who haven't logged in for more than 7 years.
The filter "User columns > lastvisitDate < {time}-220903200" fails with:
Incorrect DATETIME value: '1573952771'
{time} is replaced by a Unix timestamp, which works fine for HikaShop's own timestamp columns (user_created, order_created, ...), but lastvisitDate comes from the Joomla #__users table and is a DATETIME column, so MySQL rejects the integer.
Workarounds I tried:
- A fixed value like 20191116000000 works, but has to be updated manually.
- A date string like "2019-11-16 00:00:00" is rejected with "Wrong value" because only numeric values are accepted for <, >, <=, >=.
- REGEXP works but is also static.
Feature request:
When a filter value contains {time} and the target column is a DATETIME/DATE column (such as the Joomla user columns lastvisitDate and registerDate), please convert the calculated timestamp to 'Y-m-d H:i:s' before building the query. Alternatively, a dedicated filter such as "Last visit older than X days" would also solve it.
This would allow fully automated, cron-triggered retention clean-ups without manual maintenance, which is useful for many shops given GDPR requirements.
Thanks in advance!
Kind regards,
Lumiga