Thanks for the explanation. I am using Inventree for a lot of task you mentioned even though I am a single individual designer and I do understand the point of having things organized in one place. And that is the part where I did not understand the point of needing another storage (i.e. VCS) separate from the database.
There is no benefit of putting data that does not fit the database schema into a database. I mean you could, but then you could use database with sql as a raytracer too.
But i interpret the current solution as the simplest thing that can work.
This is becomming more and more off topic. But
a) you can modify the schema to fit the data you need - even footprints and symbols if that is what you need. After all the database is there to serve you and store “data” you need. That is why it is called "data"base. For example BLOBs are a way how to store such kind of data.
b) I did not mean necessarily to store the data directly in the database but under the control of the database. That means that there are no two separate storage system that need to be synchronized.
For example I store all my projects, libraries, 3D models, etc… in a local git repo. All data have have to be contained in that repo with the exception of link into my inventree and datasheets. And I also try to cache the datasheet localy as much as possible. This means that I can checkout any project in any stage anywhere in the past. This also means that I can have a CI/CD tools running creating all my outputs and it can create outputs from any project anywhere in the past. And all is because I have all the important data in one storage system without the need of synchronization betweem multiple systems… Except that Inventree of course.
I understand that you can change your schema. But a drawing is not a table, so blob would be more appropriate. But again kiCad is doing the simplest that can work, which is pull a table off the database. To be able to put other things in the database it would need to be able to push into the database. Which is not the goal for simplest thing that may work.
Blobs also just replicate a file so… You could also just map your file system to the database too.
No comment on your workflow.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.