summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorLi Haoyi <haoyi.sg@gmail.com>2019-11-03 23:17:13 +0800
committerLi Haoyi <haoyi.sg@gmail.com>2019-11-03 23:17:13 +0800
commitf7b2a7b84a6ca561fad2a534091db93dc9486754 (patch)
tree4060a60246760bebbda607a6a603ea5b1ea7be99
parentb8a67854660887398cd1247b0248c3ad134029be (diff)
downloadcask-f7b2a7b84a6ca561fad2a534091db93dc9486754.tar.gz
cask-f7b2a7b84a6ca561fad2a534091db93dc9486754.tar.bz2
cask-f7b2a7b84a6ca561fad2a534091db93dc9486754.zip
actor readme
-rw-r--r--docs/pages/4 - Cask Actors.md6
1 files changed, 6 insertions, 0 deletions
diff --git a/docs/pages/4 - Cask Actors.md b/docs/pages/4 - Cask Actors.md
index 9c739ad..dca3afe 100644
--- a/docs/pages/4 - Cask Actors.md
+++ b/docs/pages/4 - Cask Actors.md
@@ -72,6 +72,12 @@ elsewhere e.g. if you have a remote error aggregating service. The actor
continues processing messages after the failure in the state that it was left
in.
+Cask Actors are meant to manage mutable state internal to the Actor. Note that
+it is up to you to mark the state `private` to avoid accidental external access.
+Each actor may run on a different thread, and the same actor may run on
+different threads at different times, so you should ensure you do not mutate
+shared mutable state otherwise you risk race conditions.
+
## Example: Asynchronous Logging using an Actor
Here is a small demonstration of using a `cask.actor.SimpleActor` to perform