Squashing migrations?

[Brian Rutledge]

Disclaimer: I know very little about this topic (but I want to learn). 

I've successfully (so far) integrated django CMS into an existing large project. One big caveat is that the migrations take a long time, which can make testing and deployment more cumbersome. At the moment, we've got 242 migrations, and 80 of those are related to django CMS and its plugins.

Of course, we can work on squashing our existing migrations. But I'm curious if the django CMS developers have explored that possibility.

[Iacopo Spalletti]

I actually had this discussion with Markus Holtermann at DUTH, and I'm
actually +1 on doing a squashmigrations with any major/minor release of
django CMS
Django migrations handle this case very well and we can make the life
easier for new projects while retaining compatibilities with upgrading ones.
Django project is undergoing a major effort to drastically reduce the
migration calculation time, even for larger sets. I don't know it's
going to be in 1.11 or not, though


--
Iacopo Spalletti

Nephila s.a.s. - Firenze
Telefono: +39 055 5357189
Assistenza Tecnica: +39 055 3985730
http://nephi.la

[Brian Rutledge]

Cool. Is this something that could happen in the 3.4.x series?

[czpython]

Because of the nature of this change, this would be done in a major (4.x.x) or minor release (3.5.x).