suddenly inexplicable errors in v8 since last update
Hello all,
since the latest update to Max I am experiencing some errors in complex blocks of code where I am not at all pointed to its origin, let alone what can be done about it.
object_findregistered: looking for single sym registered object obj:dictionary:u841003805 in namespace:array, but found in dictionaryI can extract from this that somewhere an array with specific name is expected but instead a dictionary under that name is received. Now, this error never came before and I am literally clueless as to where the problem hails from. The code thereafter does not do what it is supposed to so my program is no longer working.
Anyone any ideas?
Cheers.
Very hard to say without an example. It sounds like that something was previously returning or outputting an array containing a dictionary that now is simply returning or outputting a dictionary, and either calling an array expecting function in JS or passing to another max object's array method.
Another stab in the dark for v8 specifically is if you have a msg_array method defined and are passing in the message array obj:dictionary:u841003805 somehow. But that could also be true for any other array supporting objects in your patch.
Ha, finding the responsible culprit to provide an example is exactly what I struggled with since I was not being pointed anywhere near the error in the environment. And it is a complex project with many js-dependies spread everywhere.
Anyhow, in trying to extract the problem and by going through the code line by line I have in fact ended up somewhere remote in the patch where — as you had guessed — an obj:dictionary:u/$(&%=§§)=? array was submitted to an array object which then interpreted the first element of the dictionary as the array name and thus threw the error. I am not sure why this had not happened before and whether some parsing behaviour has changed. Either way, here at least this issue got fixed. So thank you for your help!
Glad to hear! If you can isolate that part of your patch any and are able to identify a change in behavior, we could then evaluate if there is a bug that has been introduced here.
Thanks!