04 December 2020, 8:22

Dear members of the Mahara Community

For an upcoming scientific publication we (Prof. Stefan Oppl, Gregor Pirker and I from Danube University in Austria) are currently trying to analyse the old Mahara database from Danube University. We want to investigate how people have used the platform.

One of my major problems is that some users in the database have NULL entries for their ctime field. Do you have any idea what may be the cause for this? I need the ctime in order to analyse for how long a user has been active on the mahara instance.

This is also closely related to my second problem: When analysing the creation dates of artefacts I end up with some (a very little percentage) negative numbers. For this analysis I analyse the time difference between the ctime of the artefact and the ctime of its author. Again, do you have any idea what might be the cause for this? Am I doing something stupid or is this an issue with the data?

For both cases it could be that I have a lack of understanding the database or the problems are just 'glitches' in the data so any advice on how to solve these problems is greatly appreciated.

Thank you for your time and all the best,
  Martin Dobiasch

04 December 2020, 10:28

Hi Martin

For the problem with NULL 'ctime' values - this looks to be an issue that was created in 29 Jan 2010 with commit e5367399795f521bd7f3742f1ef1dedca15eeb2c where the ctime column was added allowing NULL.

So this means that if any code that adds lines to 'usr' table didn't have ctime set it would be false or any usr added before this time would also have that value as NULL

For the second problem that looks to be a bug in Mahara - thank you for reporting it - I've made a bug report about it here https://bugs.launchpad.net/mahara/+bug/1906723

If you find any relevant information during your analysis feel free to add it there

Cheers
Robert