This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
- # aws-lambda (1)
- # babashka (3)
- # beginners (97)
- # cider (20)
- # clojure (9)
- # clojure-spec (8)
- # clojure-uk (9)
- # clojurescript (25)
- # core-async (2)
- # cryogen (7)
- # cursive (4)
- # fulcro (9)
- # graalvm (14)
- # graphql (8)
- # luminus (2)
- # off-topic (18)
- # pathom (3)
- # re-frame (1)
- # reagent (4)
- # shadow-cljs (36)
- # spacemacs (1)
- # vscode (2)
I’m pushing 3.1.5-RC1 to clojars. I did a bug fix to form-state/dirty-fields: There was an obscure bug where if you pre-created a sub-form with a tempid, then the diff would include the ident to it in the :before of the delta. The new version correctly elides tempid refs in the
:before when it is the ID of an ident.
This new RC also updates the websocket code for the Fulcro Inspect Electron app to use Sente (instead of node’s socket-io, which crashes cljdoc), and as such will not work with prior versions of the inspect electron app. I’m releasing a new version of Electron inspect that will use sente as well.
IMPORTANT: React Native users will need to set the env variable (or jvm prop)
SENTE_ELIDE_JS_REQUIRE=true for it to work in mobile devices.
If you use sente with your application for websockets, then that could cause version problems. I’m using sente 1.15 (latest)…if you’re using an older version it could break depending on the mismatch.
Fulcro Inspect Electron 2.3.0-RC1 is here: https://github.com/fulcrologic/fulcro-inspect/releases/tag/2.3.0-RC1
It would be nice to hear from anyone working in react native. If you have an app that uses a large client db, I’d also love feedback on the new db-explorer feature (which is in this electron inspect app). It should make navigating the db much faster/easier.
Thanks to @adambros for his hard work on this version. He did a lot of work porting to sente, and has also done most of the work on the new db explorer tab.
When you have a few minutes @mdhaney would love your feedback, since I know you use this with mobile.
I’ll see what I can do tomorrow. I’ve been holding off upgrading libs for the last month because we’re nearing beta release, but I can probably do a quick smoke test at least.
If no one else has needed to supply multiple inputs from Fulcro to Pathom on the server then perhaps my thinking is a bit wrong?? Anyway I implemented a workaround - as part of
df/load I put the key
:pathom/context , which is noticed on the server, causing the data structure to be changed so that
:pathom/context is re-positioned to where Pathom needs it to be, prior to the parser being called. I imagine there's a better way, but this will do for now...