Postgres has a super readable codebase and the functions in the article seem to come directly from the code. When you understand the general model of how postgr

Postgres has a super readable codebase and the functions in the article seem to come directly from the code. When you understand the general model of how postgres works it's fairly easy to browse the code. Having said that I haven't look at the deeper structures at all before and have mostly just perused the query planner / executor code. We were looking through it earlier to see why our nested loop nodes where performing badly — which ultimately made us realise we were just misreading the EXPLAIN output :-)