What is the funniest or strangest error message you have ever seen from a compiler?
My personal favorite is one I came across with java: cannot cast boolean to Boolean
Any others?
What is the funniest or strangest error message you have ever seen from a compiler?
My personal favorite is one I came across with java: cannot cast boolean to Boolean
Any others?
PROGRAMMER IS INSUFFICIENTLY POLITE
From ick, the INTERCAL compiler.
Not sure if it's the 'best', but I always find C++ template errors fun; especially the 25-line long ones where you spend longer trying to understand the error message than you did writing the code in the first place!
I dont' remember the actual wording, but Intel Fortran compiler had a clear opinion when he found that the horrible code (NB: not mine) I was feeding him modified the cycle variable inside a cycle. His opinion was sort of "screw you, I refuse to compile this blasphemy"
Edit:
Ok, so either they changed the error message, or that was my personal interpretation:
fortcom: Error: test.f, line 4: An assignment to a DO variable within a DO body is invalid. [I]
i=8
---------^
Even g77 refuses to compile it.
test.f: In program `foo':
test.f:2:
do i=1,7
1
test.f:4: (continued):
i=8
2
Attempt to modify variable `i' at (2) while it serves as DO-loop iterator at (1)
Not so funny anyway. Feel free to downvote me.
A few from here:
"Too many errors on one line (make fewer)"
"We already did this function"
"This label is the target of a goto from outside of the block containing this label AND this block has an automatic variable with an initializer AND your window wasn't wide enough to read this whole error message"
But really, all of them.
Not really a compiler error, but I like IE's Javascript error message "null is null or not an object".
Visual studio's "C1001: INTERNAL COMPILER ERROR
" always gets me laughing for some reason.
Ocaml has a wealth of head-scratching error messages. One of my favorites is:
This expression has type int but is here used with type int
which you can get if you accidentally override an existing type in the top level.
My favorite error message comes from the Glasgow Haskell Compiler:
qcgadt.hs:115:7:
My brain just exploded.
I can't handle pattern bindings for existentially-quantified constructors.
In the binding group
(Rep x) = uncompressRep s
In the definition of `uncompressRep':
uncompressRep (0 : 1 : 0 : s) = let (Rep x) = uncompressReps in Rep (RList x)
Digital Equipment Basic-Plus and Basic-Plus-2 in RSTS/E OS, running on PDP-11's:
Program Lost -- Sorry.
Heh, :-)
"long long long is too long for gcc"
Xcode was cranky at me.... (C++)
You are in a maze of twisty compiler features, all different
From GCC 1.34's "implementation-defined" handling of #pragma
, displayed if your system was unable to launch emacs -f hanoi
, hack
, or rogue
(which were its preferred implementations of #pragma
).
This message is from Squeak.
Whenever I realized I had started to run code that was destined to produced a stack overflow error, I quickly hit CTRL+C (squeak for cancel operation). Sometimes it worked but when it didn't it would popup a dialog which said
Insufficient resources to interupt this operation
But yes, enough resources to popup the dialog.
PHP
Parse error: syntax error, unexpected T_PAAMAYIM_NEKUDOTAYIM in Command line code on line 1
Got this today and thought it was funny:
Warning 8 warning C4800: 'BOOL' : forcing value to bool 'true' or 'false'
Error: This DO variable has already been used as an outer DO variable in the same nesting structure.
I just found this old screenshot in my hard drive. I can't quite remember what had gone wrong... and I am afraid to know:
From .NET, my favourite has always been:
Nullable object must have a value.