ID:1919743
 
BYOND Version:508
Operating System:Windows 10 Pro 64-bit
Web Browser:Chrome 44.0.2403.155
Applies to:Dream Daemon
Status: Open

Issue hasn't been assigned a status value.
Descriptive Problem Summary:
A runtime error has reported a seemingly random text string from the code as source file, instead of the actual source file.

Numbered Steps to Reproduce Problem:
No idea. Have a huge codebase maybe?

Code Snippet (if applicable) to Reproduce Problem: A couple examples:
runtime error: Cannot read null.vampire
proc name: attack (/obj/item/weapon/nullrod/attack)
source file: The emergency shuttle is already on its way.,41
usr: Yahir Hunt (/mob/living/carbon/human)
src: the null rod (/obj/item/weapon/nullrod)
call stack:
...

runtime error: Cannot execute null.afterattack().
proc name: preattack (/obj/item/offhand/preattack)
source file: <span class='sinister'>You remember the Realm of Nar-Sie, The Geometer of Blood. You now see how flimsy the world is, you see that it should be open to the knowledge of Nar-Sie.</span>,38
usr: Igor Slovenich (/mob/living/carbon/human)
src: the offhand (/obj/item/offhand)
call stack:
...


The second one is found Here, line 11405


Expected Results:
To show the source file

Actual Results:
Actually shows a seemingly random text string somewhere in the code.

Does the problem occur:
Every time? Or how often? No idea, seen it in a couple of locations now.
It also appears that once a broken runtime like this appears all runtimes that are the same start having the same issue.
In other games? No idea, probably not.
In other user accounts? Yes.
On other computers? Yes.

When does the problem NOT occur?
No idea

Did the problem NOT occur in any earlier versions? If so, what was the last version that worked?
No idea, however I've only started seeing this recently.

Workarounds:
No clue.
Looks like something is causing problems for the string tree. Either that, or the proc data itself is corrupted. Without more info I don't have a lot I can investigate.

Is it at all possible that this is related to the recent change to stop counting recursion limits as critical errors? I didn't see such an error in the code, though if a try/catch is handling it then it wouldn't show up. The change I made removed a deliberate safety feature, so the fact of something bad coming from it wouldn't shock me.
19:11:54 Runtime detected
Cannot execute null.copy to(). at MOUSE_ALT_KEY:78
proc name: breathe (/mob/living/carbon/human/proc/breathe)
source file: MOUSE_ALT_KEY,78
usr: null
src: Sean Stroble (/mob/living/carbon/human)
call stack:
Sean Stroble (/mob/living/carbon/human): breathe()

have seen several other instances of this, it seems random refs are being passed to world/Error in place of the actual source file. here's another http://ss13.pomf.se/uploads/2015-08-26_01-47-53.txt