RFE: Consider using an ORM to support databases other than Postgres

[jwatt]

Happy New Year! And thanks for getting Commento development moving again! 😄

Tying Commento to Postgres limits how many people can use it. In principle, an ORM like GORM could allow a far wider audience to use Commento by making other databases an option (MySQL and SQLite in particular).

I've no idea whether anyone might be interested in doing the work to implement this, but I thought I'd file this issue on the off-chance.

[souramoo]

Happy new year! 😄

I completely agree we should move to an ORM, this would definitely help increase where commento can run!

Currently, the only things holding this back from the original commento are:

• Reliance on pgsql features like triggers
• A folder full of migrations to help upgrade older versions of commento
• Hardcoded SQL statements in the /api folder

I will think about how best to address all of these!

[jwatt]

Awesome, thank you for outlining the work and the willingness to look into the various pieces. I understand these are all significant pieces of work however, so no pressure!

[0mni]

I too would love to help in some way to support mysql. All my other services use MySQL, they also support postgresql, all except one. And the chances of Ghost supporting postgresql are .... not likely, especially since it looks like they dropped support for it a few versions back.

[dflock]

I would love to see this. SQLite support in particular would open up hosting options a lot; would be much simpler to host serverless, etc... I know this is only a small part of the work, but SQLite does support triggers.

[dflock]

I guess step one would be to choose an ORM?

[jwatt]

Possibly, although it may make more sense to wait until there's someone willing to work on this, since they may have a strong opinion on which ORM they want to use. :-)