[NEW💥] Node.js 18 LTS is now available. What’s new? 🤔 In honor of the number 11 (#Funfact Undici means ‘eleven’ in Italian), we decided to make our Top 11 Node.js 18 features. We invite you to read them here
👉https://t.co/sqI7IXPjnj
@ibuildthecloud “We don’t give a crap about developer experience” was one of the statements I was told by one of the team’s leaders while interviewing for a job there. 😇 Stuck with me for whatever reason.
@jalonen_lauri@PostgreSQL@planetpostgres 1. Use one big insert or CTEs to compose the query
2. Pipelining is in the works, which will reduce such issues https://t.co/ImovFPyvmh
Super excited for the Custom Highlight API to come soon!
Spec ➡️ https://t.co/hVjzHhHbMx
This will enable many libraries and apps to do find-on-page, spellchecking and other such use cases faster and much easier.
@brycebaril I'm also using a zset for scheduled jobs with the hash key in it. And a hash to save the job itself, pushing it on expiration👍
https://t.co/MGCI7gcojX
Interesting redis stats
The memory of a Hash is lower than List
HSET foo id foo bar baz
MEMORY USAGE foo
78
XADD bar * id foo bar baz
MEMORY USAGE bar
594
XADD bar * id foo bar baz
614
LPUSH quz id foo bar baz
MEMORY USAGE quz
150
LPUSH quz id foo bar baz
MEMORY USAGE quz
169
@brycebaril In my case I wanted a list of entries I could pass to xadd later in a lua script without doing a round-trip application-side. True, memory-wise it doesn't really matter. Api-wise it would be neat to have some similar support for immutable lists like in streams.
@Redisinc Ok, if I want to persist a list on a hset value, messagepack would work that's bundled with redis.
cmsgpack.pack and cmspack.unpack
https://t.co/74jiGd4ffR
@Redisinc it would be interesting to have an immutable stream like structure (key value key value) without the overhead of lists and hashes. A hash currently can't replicate the stream-like structure as keys are deduplicated. But the list has double the initial memory usage.