You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 84e03d8
Browse filesBrowse the repository at this point in the historyBrowse files
fix: keep server stores until destroy() and compare object query hashes by content (#48)
On the server every store armed a 500 ms init timer and a 10 s touched
timer although a request renders once and calls destroy(); the timers only
kept ~1.7 MB of state and 20 handles alive per request. The manager now
detects the server and arms no destroy timers there.
SuspenseQuery compared hashes with strict equality, so an object or array
hash created on every render never matched the stored one and the query
was re-run and re-thrown forever. Object hashes are now serialised before
they are stored and compared.
Copy file name to clipboardExpand all lines: docs/guide/ssr.md
+2Lines changed: 2 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -43,6 +43,8 @@ Examples that need cleanup:
43
43
- cleanup returned from `init`
44
44
- internal store cleanup
45
45
46
+
On the server the manager arms no destroy timers (`destroyTimers` only apply in the browser): stores live exactly as long as the request, and `destroy()` is the only teardown.
0 commit comments