Presto (old site, new site, see comment below) is an open source SQL engine used in big data scenarios. We use it with the community edition of jOOQ and would like to suggest you support it officially. In our preliminary testing, its SQL "dialect" is largely identical to PostgreSQL (because both products are very close to the standard).
Thank you very much for your suggestion.
Currently, we're not looking into officially supporting any additional databases that are far below rank 20 on this list: http://db-engines.com/en/ranking. Presto is currently at 162
The traction we would be getting from such support would just be too low to justify the effort that goes into maintaining an officially supported driver.
Hello, is Presto dialect on the roadmap?
@ssharma Thanks for your message. It is on the long term roadmap (thus this issue), but we currently do not prioritise it until we see more traction
Is there any way to vote for this feature? Honestly, at the moment two dialects I care about are Hive and Presto. I wonder if there is anything I can do to see them being on your radar? Thanks
@HasanAmmori We do not have a formal process, although we do consider upvotes and comment feedback.
In the meantime, since I first checked in 2016, Presto did gain a lot of traction according to https://db-engines.com/en/ranking, so it definitely makes sense to review an integration.
We're about to release 3.12, and have a few SQL databases on our radar for ~3.13~ (future versions), including:
We'll reconsider looking into supporting Presto as well - no promises of course.
FYI, as of January 2019, the presto project lives at https://prestosql.io/ and is stewarded by the newly created _Presto Software Foundation_ which is headed by the three creators of Presto. See https://www.prweb.com/releases/presto_software_foundation_launches_to_advance_presto_open_source_community/prweb16070792.htm for more information.
I'm linking this here as the link in the issue description still points to the old site where remaining project members continue to create (fewer) releases.
Thanks, @bannmann
Presto maintainer here. If you have any questions I will be happy to answer and help.
As far as testing is concerned, I can recommend dockerized Presto (directly
or with testcontainers https://www.testcontainers.org/modules/databases/presto/).
Thanks, @findepi, I'll be in touch
The roadmap for 3.15 is already full with new dialects. This one won't fit anymore.
Update from the project. Presto SQL has been renamed to Trino. This is the version maintained by Presto creators and long-time maintainers.
And Facebook continues to maintain their version under PrestoDB name.
The two projects have already diverged heavily, with differing support for language features, (e.g. Trino supports OFFSET whereas PrestoDB does not; you can read more in 2019 and 2020 reviews), and with separate JDBC libraries.
@lukaseder, I fully understand this isn't a convenient situation to jOOQ project maintainers, since you need to choose which one to add support for, unless you choose to support both.
Since the original request wasn't clear about which version they're using, it probably doesn't make the decision easier.
Happy to answer questions, should there be any. Here, over email or Trino community slack.
Update from the project. Presto SQL has been renamed to Trino
Thanks for the update. I've created a new issue to track this:
https://github.com/jOOQ/jOOQ/issues/11485
I fully understand this isn't a convenient situation to jOOQ project maintainers
Au contraire, this is very convenient, as it lowers priorities for both dialects 馃榿
@findepi wrote:
Since the original request wasn't clear about which version they're using (...)
Well, my original request was from 2016 when there was only one version. However, in 2019 my company chose Presto SQL. We will continue to use the releases of that newly rebranded project. Hence, if we had to chose, our vote goes to Trino (#11485) instead of Presto (#5414).
Hint: Depending on the SQL queries one uses, people may have success using jOOQ's PostgreSQL dialects to connect to Presto or Trino. We have been using this workaround for more than 4 years now without any trouble.
It appears that Presto is used also in Amazon Athena: https://aws.amazon.com/athena/features/?nc=sn&loc=2
Regardless of dialect, support for Amazon Athena would certainly be welcome here
Understood. Now if anyone has budget to speed up this development, we might ... speed up this development :)
Most helpful comment
@findepi wrote:
Well, my original request was from 2016 when there was only one version. However, in 2019 my company chose Presto SQL. We will continue to use the releases of that newly rebranded project. Hence, if we had to chose, our vote goes to Trino (#11485) instead of Presto (#5414).
Hint: Depending on the SQL queries one uses, people may have success using jOOQ's PostgreSQL dialects to connect to Presto or Trino. We have been using this workaround for more than 4 years now without any trouble.