I'm very curious how well this is working for you in practice, since I've been thinking about what it would look like to share a single Rust UI implementation across a webapp and native apps.
If you have a table that undergoes frequent updates and this is a concern of yours, look into setting a fillfactor on it. This will help Postgres keep writes in the same page, which means it doesn't need to touch indexes for non-indexed columns.
Looking at your PR for it (https://github.com/LuaJIT/LuaJIT/pull/149) it sounds as if there are a few issues remaining, but I'm not enough of an expert to know how problematic they are in practice. Do you think that this feature is stable enough for production use?
That is, indeed, what Sequel does. You can query the database object directly (for example, `DB[:posts].where{comments_count > 5}.exclude(poster_name: "Bob").order(:comments_count.desc).limit(10).all`) and get back an array of hashes.
Sequel's model layer is also faster than ActiveRecord's, because much of the additional functionality that ActiveRecord piles onto all records (dirty tracking, single table inheritance, etc.) are available in Sequel via a plugin system. You can enable the plugins you want to use and not pay the overhead of all the others. You can even enable specific plugins for only the models where you'll actually want to use them.
It also has a lot less magic (no association proxies, unless you enable the plugin for them :)), ridiculously customizable (many more options for associations, custom eager loading logic, and so on), and has an implementation of its Postgres adapter written in C for performance.
Highly, highly recommended.
Edit: Oh, and an issue tracker that is almost always at zero, with a very fast response time. As someone who has contributed a patch to ActiveRecord, I can't tell you how nice that is.
(I'm the author of Que, the job queue discussed in the post most extensively)
This isn't terribly surprising to me, since I have an appreciation for what long-running transactions will do to a system, and I try to design systems to use transactions that are as short-lived as possible on OLTP systems. I realize that this should be explicitly mentioned in the docs, though, I'll fix that.
I'll also note that since the beginning Que has gone out of its way to use session-level locks, not transaction-level ones, to ensure that you can execute long-running jobs without the need to hold open a transaction while they work. So I don't see this so much as a flaw inherent in the library as something that people should keep in mind when they use it.
(It's also something that I expect will be much less of an issue in version 1.0, which is set up to use LISTEN/NOTIFY rather than a polling query to distribute most jobs. That said, 1.0 has been a relatively low priority for much of the last year, due to a lack of free time on my part and since I've never had any complaints with the locking performance before. I hope I'll be able to get it out in the next few months.)
My point is that the only way to wrap your jobs in the same transactions as the rest of your data is to have your job queue in your RDBMS. If you don't have that, you can't guarantee that they are consistent, that your backups have snapshots of each at the same time, etc.
Inconsistency between jobs and the rest of your data isn't a problem for many (or even most) use cases, but there are certainly times when you need it.
The major benefit of putting your queue in your RDBMS, which isn't commonly brought up in these discussions, is that it lets you protect your jobs with the same ACID guarantees as the rest of your data. This is very valuable for some use cases.
I have a Postgres-based job queue that uses advisory locks to get around some of the drawbacks he mentions (job lock queries don't incur writes or block one another like SELECT FOR UPDATE would). Feedback is welcome: https://github.com/chanks/que
Hi, I'm the author of Que. It's true that you can't really completely solve the idempotence problem for jobs that write to external web services (unless those web services provide ways for you to check whether you've already performed a write - see the guide to writing reliable jobs in the /docs directory), but that's a limitation that'll apply to any queuing system. I'd definitely say that Que, being transactional and backed by Postgres' durability guarantees, does give you better tooling for writing reliable jobs than a Redis-backed queue would in general.
I'm happy to answer any questions you or anyone else might have.
You have some valid criticisms of the implementation, though pgsql-hackers has their reasons (http://www.postgresql.org/message-id/CAHyXU0ybwZZUbuQQVFQMK3...) for the way things are. Regardless, my point is that the situation is not so bad as "Every time that function is executed, it is re-parsed and re-planned."
PL/pgSQL function plans are cached. Functions written in other languages may not be, I don't know, but in my experience PL/pgSQL is the most widely used.
What I would find helpful is some sort of operator to merge two JSON values. I use a JSON field to collect various statistics relating to a table's rows, and it'd be nice to not have to worry about overwriting old values. Something like:
One thing people don't mention that often with Postgres, but is worth some attention, is the Heap-Only Tuples (HOT) optimization that was added in 8.3. During an UPDATE, Postgres will try to fit the new version of the row on the same page as the old one - if it's able to do this, it means the indices don't have to be updated at all.
So if you're expecting a table to have a heavy UPDATE load it may be a good idea to lower its fillfactor from the default of 100 (for example, ALTER TABLE foo SET (fillfactor = 95);). This means that once a page is 95% full no new rows will be inserted into it, and the remaining 5% will be available for updated rows. HOT will also do "mini-vacuums" on individual pages as needed to free up space.
Of course, it's up to you to decide whether this tradeoff is worth increasing the effective size of your table by ~5%. But even tables with fillfactors at the default of 100 will get some benefit from HOT.
It provides an ActiveModel-compliant interface, so it works fine with Rails. It also has a lot less magic (associations don't use proxy objects, for example) and has a lot of features that ActiveRecord is sorely missing (such as complete support for composite primary keys). I don't know what your other problems with ActiveRecord are, but Sequel is definitely worth a look. I'm very happy with it.