v0.10.4

Changed

• Interceptors, effects, and coeffects are now captured in re-frame's tracing.

v0.10.3

New

• add purge-event-queue to the API. See day8/re-frame-test#13 for motivation.
• added a new FAQ entry Reagent looks terrific. Why do I need re-frame?
• added a new Infographic explaining how dominoes play out over time in the browser

Changed

• Debounce trace callbacks to handle larger batches of traces at once, to improve efficiency.
• Improved error messages to not have multiple spaces before variables.

Fixed

• Handle js/performance not being defined in NodeJS. #439
• Improve cache eviction behaviour of subscription caches. In more complex applications a subscription may have been unnecessarily created and destroyed several times after a Figwheel re-render.

Historical changes are at CHANGES.md.

v0.10.3-rc2

This is the release candidate for v0.10.3. Try it out in your applications, and let us know if there are any issues.

You can see the CHANGELOG for all of the changes.

v0.10.2 (2017.10.07)

New Features

• On dispose callbacks now run on JVM re-frame. #386.
• Log warnings when tracing is not enabled and you try to add a trace callback #395.

New Docs

• added a new FAQ entry How do I turn on/off polling a database every 60 secs (hint: effects)
• added a new FAQ entry How do I do full-stack re-frame?
• Added a gitbook for re-frame docs #414.
• Lots of spelling fixes, for which we are eternally grateful.

See the full changelog at CHANGES.md

v0.10.1

0.10.1 (2017.08.17)

• fix a showstopper bug introduced at the last minute in v0.10.0

0.10.0 (2017.08.16)

New Docs

• added API documentation
• added testing docs
• added a new mental model
• added a new FAQ entry on dispatch processing
• added a new FAQ entry on representing normalised data in app-db
• added a new FAQ entry on how to register a global interceptor

Breaking

• #357
  I'd be amazed if this actually broke any apps. Shocked! But, better safe than sorry.
  The effect handler for :db has changed: if the new value provided tests
  identical? to the existing value within app-db, then app-db is not reset!.
  Previously, app-db was always reset! irrespective,
  which potentially caused Layer 2 subscriptions to run unnecessarily. So this is a tiny
  efficiency change in this edge case, and it results in behaviour that better matches
  programmer intuitions.

Minor Fixes and Improvements

• #400 Improve error message when a registered cofx can't be found
• The effect handler for :dispatch-n will now ignore nils. See checkin
• #340
• #341 Fix re-frame.core/on-changes to work even if event handler does not set :db.
• #395 Warn users if they try to enable tracing without setting the closure define re-frame.trace.trace-enabled?.

v0.10.0

This release has a show-stopper bug in it. You shouldn't use it, and instead should upgrade to 0.10.1.

v0.9.4

This is another minor release, adding a CITATION.md file, and changes for self hosting.

You can see the full changelog here.

v0.9.3

This is another minor release, fixing a bad ns form, some undefined behaviour around reg-sub, and the update-coeffect interceptor.

N.B. The changes to reg-sub breaks code that was wrong but previously worked. The reg-sub function was too liberal in the separator it accepted, now only :<- is accepted.

You can see the full changelog here.

v0.9.2

This is a small release, mainly switching re-frame's lein-npm dependencies to use :devDependencies. You can see the full changelog here.

v0.9.1

This is a hotfix release which fixes dynamic subscriptions. The full changelog is available here.