asyncpg VS httpx

- SqlAlchemy/asyncpg => you can’t use it if you’re using PgBouncer (necessary most of the time with Postgres) in transaction mode? What?? https://github.com/MagicStack/asyncpg/issues/1058

OK I stand corrected, asyncpg has these two C files:

https://github.com/MagicStack/asyncpg/blob/master/asyncpg/pr...

https://github.com/MagicStack/asyncpg/blob/master/asyncpg/pr...

If you are interested here is a post by the psycopg author about psycopg2 and 3 and performance versus asyncpg.

https://www.varrazzo.com/blog/2020/05/19/a-trip-into-optimis...

This pure Python library claims quite fabulous performance: https://github.com/MagicStack/asyncpg

I believe it because that team have done lots of great stuff but I haven't used it, I just remembered thinking it was interesting the performance was so good. Not sure how related it is to running on the asyncio loop (or which loop they used for benchmarks).

what a great post, we have had a ton of issues with users using pgbouncer and it's not because things are "broken" per se, it's just the situation is very complicated, and pgbouncer's docs are also IMO in need of updating to be more detailed and in a few critical cases less misleading, specifically the prepared statements docs.

This blog post refers to this misleading nature at https://jpcamara.com/2023/04/12/pgbouncer-is-useful.html#pre... .

> PgBouncer says it doesn’t support prepared statements in either PREPARE or protocol-level format. What it actually doesn’t support are named prepared statements in any form.

That's also not really accurate. You can use a named prepared statement just fine in transaction mode. start a transaction (so you aren't in autocommit), use a named statement, works fine. you just can't use it again in another transaction, because it will be "gone" (more accurately, "unmoored" - might be in your session, might be in someone else's session). Making things worse, when the prepared statement is "unmoored", its name can then conflict with another client attempting to use the same name.

so to use named prepared statements, you can less ideally name them with random strings to avoid conflicts, or you can DEALLOCATE the prepared statement(s) you used at the end of your transaction. for our users that use asyncpg, we have them use a uuid for prepared statements to avoid these name conflicts (asyncpg added this feature for us here: https://github.com/MagicStack/asyncpg/issues/837). however, they can just as well use DEALLOCATE ALL, set this as their `server_reset_query`, and then so that happens in transaction mode, also set `server_reset_query_always`, so that it's called at the end of transactions. Where pgbouncer here IMO entirely misleadingly documents this as "This setting is for working around broken setups that run applications that use session features over a transaction-pooled PgBouncer." - which is why nobody uses it, because pgbouncer claims this is "broken". It's not any more broken than it is to switch out the PostgreSQL session underneath a connection that uses multiple transactions. Pgbouncer can do better here and make this clearer and more accommodating of real world database drivers.

asyncpg is another great driver if you're using asyncio and want maximum performance (although they also break with DBAPI, but the tradeoff may be worth it).

asyncpg: 5.5k starts, last commit recently, ~150 issues, some incompatibility, few open PRs, extensive README. Includes benchmark showing it's supposedly 3x faster than aiopg and psycopg2, psycopg3 is not mentioned in the benchmark.

Quart-DB uses asyncpg to manage the connections and buildpg to parse the named parameter bindings.

A major performance boost specifically on inserts with timescaledb was actually starting to use https://github.com/MagicStack/asyncpg.

Simple query runs long when DB schema contains thousands of tables #186

For reference, it's a butterfly, not a moth.

Source: https://github.com/encode/httpx/issues/834

Very cool, first I'm hearing of httpx https://www.python-httpx.org/

I think most people would start with trying out requests or something for this kind of work, I'm guessing that didn't work out? You've got a star from me.

To access 10 different commands at the same time, that is tricky but definitely doable.

First thing that comes to mind, you can use aliases.

To keep it simple, lets use 3 examples instead of 10: harlequin (this project), pgcli (https://www.pgcli.com/) and httpx (https://www.python-httpx.org/)

Setup a main home for all your venvs:

    cd ~

There are already some implementations for Python HTTP clients. One of them is aiometer. But it's not suitable for my use case. Since httpx already has the internal pool, it would be better to reuse the design.

Besides, flama also provides support for SQL databases via SQLAlchemy, an SQL toolkit and Object Relational Mapper that gives application developers the full power and flexibility of SQL. Finally, flama also provides support for HTTP clients to perform requests via httpx, a next generation HTTP client for Python.

We can use the async HTTP client provided by httpx, a fully featured HTTP client for Python with an API broadly compatible with requests, so it can be used in pretty much the same way in most cases.

httpx in place of requests library

The main value of Requests is that it provided an abstract interface on top of HTTP, which was designed well-enough to become a standard. But today it has fallen way behind in its field, and there are much better alternatives such as HTTPX [0].

[0] https://www.python-httpx.org/

HTTPX is a popular Python library that provides an asynchronous HTTP client, and it can be beneficial for enabling async support in Django. While Django itself does not require HTTPX for async support, using HTTPX in combination with Django's async views can bring several advantages:

The underlying library for both sync and async is httpx (https://www.python-httpx.org/) which may be limited from the HTTP Client perspective but it may be possible to add rate limiting at a Session level.