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 }}
This repository has been archived by the owner on Dec 9, 2017. It is now read-only.
Say we record the above script in the browser, and then replay under node.js. In the browser, process is undefined, so this gets recorded to the trace. During replay, we read in the trace and overwrite the real process global with undefined, which seems bad. While replay itself seems to proceed, replay.js reads process at the end, which causes a crash.
Not sure of the right fix for this one. Maybe replay.js should stash away a pointer to the process object before starting actual replay?
The text was updated successfully, but these errors were encountered:
msridhar
changed the title
Don't overwrite process when recording in browser and replaying under node
Overwriting process global when recording in browser and replaying under node
Apr 15, 2014
Actually, stashing away the pointer in replay.js is insufficient; all analyses (and dependent libraries) that refer to process would have to do something different, which seems impossible to enforce.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Consider the following code:
Say we record the above script in the browser, and then replay under node.js. In the browser,
process
isundefined
, so this gets recorded to the trace. During replay, we read in the trace and overwrite the realprocess
global withundefined
, which seems bad. While replay itself seems to proceed,replay.js
readsprocess
at the end, which causes a crash.Not sure of the right fix for this one. Maybe
replay.js
should stash away a pointer to theprocess
object before starting actual replay?The text was updated successfully, but these errors were encountered: