only parse one function per file??

[Remi-C]

Hey dear maintainer, I really like the idea of this package, it's a neat way to solve a common doc problem.
One thing is causing me massive headache : this package seems to pick only one function per SQL file.
This is a big problem for me, I have ~100 files containing functions, and most files contain several functions that I want to keep bundled together (like do / undo functions, or helper functions + set of wrappers, or functions that need to be created in a given order, or ...).
Am I missing something here / any possible workaround?
Thanks ! 🙇

[jackscodemonkey]

It's built around one object per file and looks for the first block comment in the file to parse for the expected fields.

The easiest solution would be if i added a start/stop delimiter to look for in the form of a SQL comment. That could be configured in the sphinx conf.py with a toggle to enable the extra parsing, keeping it optional.

What I'd be able to do in that case is split the delimited blocks into in memory files and process them normally from there.

You'd have to decorate the start / stop of your objects in file something like this:

-- sphinx_start
/*
sphinx-sql comment block
*/
create or replace function test()
returns void;
....
-- sphinx_stop

-- sphinx_start
/*
sphinx-sql comment block
*/
create table test_table (id int);
-- sphinx_stop