c_raza retweeted
A first version of our Datahike cli tool is now available.
github.com/replikativ/datahi…
It provides SQLite features, but with fully decoupled reading on snapshots and full Datalog. git-like branching and merging is also in the pipeline.
c_raza retweeted
In case you all missed it. Datahike now works in the browser and persists to IndexedDB. We got so much other good stuff coming as well!
lambdaforge.io/2021/03/03/da…
c_raza retweeted
After several months of work by our team with the collaboration of the community, it is possible to bring you the 0.3.3 release of #Datahike
Get it here: github.com/replikativ/datahi…
#Datalog #Clojure #ClujureScript #datascript
Datahike + Heroku = Instant Datalog backed Clojure Web App – Nextjournal #datahike #heroku #Clojure nextjournal.com/kommen/datah…
c_raza retweeted
In collaboration with @homebase__io we provide a preview of IndexedDB browser support for @datahike_io github.com/replikativ/datahi… This is a big milestone to a globally joinable database ecosystem, the basis for cybernetic social infrastructure we are working on at the LF coop.
c_raza retweeted
My marathon length workshop from @clojurenorth finally dropped! Learn how to build a functional, data-driven backend app step-by-step in a couple hours the #Clojure way with #datahike, #integrant, #reitit, and a REPL at youtu.be/eink6oICdOQ.
c_raza retweeted
.@konradkuehne is working on Datahike, a Datalog database. He is looking at adding tuple support, tweaking the API to be more similar to Datomic, attribute predicates and entity specs, and lazy seqs for transactions.
c_raza retweeted
This is just sad. Shame on you @reddit, you're making baby Tim Berners-Lee cry.
c_raza retweeted
30x faster?
Here's how I did it:
bsless.github.io/datahike-da…
#Clojure
We have landed some improvements in the Datalog parser which also made it into clj-kondo (@borkdude). Thanks to @_bsless the parser is now ~30x faster. github.com/lambdaforge/datal…
It originally written by @nikitonsky as part of DataScript. Thank you 🙏! #clojure #clojurescript
c_raza retweeted
We have landed some improvements in the Datalog parser which also made it into clj-kondo (@borkdude). Thanks to @_bsless the parser is now ~30x faster. github.com/lambdaforge/datal…
It originally written by @nikitonsky as part of DataScript. Thank you 🙏! #clojure #clojurescript
c_raza retweeted
Our team with the collaboration of the community is very happy to announce a new release of #Datahike. Find all the details here:
github.com/replikativ/datahi…
#clojure #clojurescript #replication #replikativ
c_raza retweeted
We are working on a front-end to interact with a Datahike database...
github.com/replikativ/datahi…
#Datahike #Clojure #Fulcro
The article about Datahike's Java API has been published by DZone: dzone.com/articles/how-to-us…
#Datahike #Java #Clojure #Springboot #lambdaforge
c_raza retweeted
Take a look at Datahike, a project with a growing community
#Datahike #Clojure #functionalprogramming
New release: After several months of work we are happy to bring you the 0.3.0 version of Datahike! You can find the new features here:
github.com/replikativ/datahi…
#Clojure #Datahike #replication
c_raza retweeted
New release: After several months of work we are happy to bring you the 0.3.0 version of Datahike! You can find the new features here:
github.com/replikativ/datahi…
#Clojure #Datahike #replication
If you are interested in using an immutable #datalog based database in Java, with the latest release of #datahike, I wrote an article about how to use its new #java API.
lambdaforge.io/2020/05/25/ja…
c_raza retweeted
Datahike released the new version 0.3.0
The new features can be found here github.com/replikativ/datahi…
Have fun trying it out. 🙂🚀
#Datahike #clojure #datomic #functionalprogramming @lambdaforge













