Hello,
django timescaledb seems to work fine. just 2 questions.
Timescaledb mentions no unique / primary key constraint is needed.
using your code i see that the primary key gets removed (e.g. no unique constraints, no primary keys) and again: that all seems to work fine.
However: django is quite explicit it needs a unique identifier based on a single column.
see for example here: https://docs.djangoproject.com/en/5.0/topics/db/models/#automatic-primary-key-fields
How is this uniqueness covered when using timescaledb ?
other question:
in order to have some speed letting django identify the instances(i tested on django admin with ~ 40 million rows) i created an index manually (raw sql) on the id column. That brings good performance. However I did not see id index management in your code. What are your thoughts, recommendations to get a good speed finding instances (for example in django admin) ?
Hello,
django timescaledb seems to work fine. just 2 questions.
Timescaledb mentions no unique / primary key constraint is needed.
using your code i see that the primary key gets removed (e.g. no unique constraints, no primary keys) and again: that all seems to work fine.
However: django is quite explicit it needs a unique identifier based on a single column.
see for example here: https://docs.djangoproject.com/en/5.0/topics/db/models/#automatic-primary-key-fields
How is this uniqueness covered when using timescaledb ?
other question:
in order to have some speed letting django identify the instances(i tested on django admin with ~ 40 million rows) i created an index manually (raw sql) on the id column. That brings good performance. However I did not see id index management in your code. What are your thoughts, recommendations to get a good speed finding instances (for example in django admin) ?