Vadim Markovtsev
Vadim Markovtsev
As any other query method of `Database`. It just happens that `iterate()` allows calling another method in parallel easier. The problem is deeper. For example, this is a real production...
The column defaults are higher level sqla API ("Query") than the API used in `databases` ("Core"). Emulating them properly is quite hard. For example, the functions can access the context...
Hey @Feijo you are probably using SQLAlchemy in an unsupported way. Try repeating that construct with a regular, synchronous API: ```python engine = create_engine("postgresql://user:password@localhost/dbname", echo=True) print(engine.connect().execute(query).fetchone()) ```
Please test with the synchronous API and report the result.
Hey @sukratkashyap sorry for this regression. I just did not know it was possible, another example of https://xkcd.com/1172/ :laughing: Seriously though, I clearly see that the surrounding code assumes a...
`execute_many` does not currently invoke `executemany` of the Postgres driver. I don't know why :smile: I have this in my code: ```python class ExecuteManyConnection(databases.core.Connection): """Connection with a better execute_many().""" async...
They just disappeared haha. I had to spend a titanic effort to make a new release last time. My [own PR](https://github.com/encode/databases/pull/211) is hanging since June 2020, for example. My guess...
I also thought about this in [my blog post](https://towardsdatascience.com/how-we-optimized-python-api-server-code-100x-9da94aa883c5) and will try to hack an alternative production-ready deserializer to numpy recarrays. My requirements are: - Drop-in replacement for asyncpg interface....
Done. [Blog post](https://betterprogramming.pub/i-forked-asyncpg-and-it-parses-database-records-to-numpy-20x-faster-e71024a84bff) Repository: https://github.com/athenianco/asyncpg-rkt As I wrote in the post, merging back would require some effort, so I want to check my opportunities with the maintainers.