91 Error and warning messages

This chapter provides detailed information about the meaning of some error messages or on how to recover from errors.

91.1 Error messages

91.1.1 apply: no such "list" element

One common cause for this error message is that square brackets operator ([ ]) was used trying to access a list element that whose element number was < 1 or > length(list).

91.1.2 argument must be a non-atomic expression

This normally means that a list, a set or something else that consists of more than one element was expected. One possible cause for this error message is a construct of the following type:

(%i1) l:[1,2,3];
(%o1)                       [1, 2, 3]

(%i2) append(l,4);
append: argument must be a non-atomic expression; found 4
 -- an error. To debug this try: debugmode(true);

The correct way to append variables or numbers to a list is to wrap them in a single-element list first:

(%i1) l:[1,2,3];
(%o1)                       [1, 2, 3]

(%i2) append(l,[4]);
(%o2)                     [1, 2, 3, 4]

91.1.3 assignment: cannot assign to <function name>

Maxima supports several assignment operators. When trying to define a function := has to be used.

91.1.4 expt: undefined: 0 to a negative exponent.

This message notifies about a classical division by zero error.

91.1.5 incorrect syntax: , is not a prefix operator

This might be caused by a command starting with a comma (,) or by one comma being directly followed by another one..

91.1.6 incorrect syntax: Illegal use of delimiter )

Common reasons for this error appearing are a closing parenthesis without an opening one or a closing parenthesis directly preceded by a comma.

91.1.7 loadfile: failed to load <filename>

This error message normally indicates that the file exists, but can not be read. If the file is present and readable there is another possible for this error message: Maxima can compile packages to native binary files in order to make them run faster. If after compiling the file something in the system has changed in a way that makes it incompatible with the binary the binary the file cannot be loaded any more. Maxima normally puts binary files it creates from its own packages in a folder named binary within the folder whose name it is printed after typing:

(%i1) maxima_userdir;
(%o1)                 /home/gunter/.maxima

If this directory is missing maxima will recreate it again as soon as it has to compile a package.

91.1.8 makelist: second argument must evaluate to a number

makelist expects the second argument to be the name of the variable whose value is to be stepped. This time instead of the name of a still-undefined variable maxima has found something else, possibly a list or the name of a list.

91.1.9 Only symbols can be bound

The most probable cause for this error is that there was an attempt to either use a number or a variable whose numerical value is known as a loop counter.

91.1.10 operators of arguments must all be the same

One possible reason for this error message to appear is a try to use append in order to add an equation to a list:

(%i1) l:[a=1,b=2,c=3];
(%o1)                 [a = 1, b = 2, c = 3]

(%i2) append(l,d=5);
append: operators of arguments must all be the same.
 -- an error. To debug this try: debugmode(true);

In order to add an equation to a list it has to be wrapped in a single-element list first:

(%i1) l:[a=1,b=2,c=3];
(%o1)                 [a = 1, b = 2, c = 3]

(%i2) append(l,[d=5]);
(%o2)             [a = 1, b = 2, c = 3, d = 5]

91.1.11 Out of memory

Lisp typically handles several types of memory containing at least one stack and a heap that contains user objects. To avoid running out of memory several approaches might be useful:

• If possible, the best solution normally is to use an algorithm that is more memory-efficient.
• Compiling a function might drastically reduce the amount of memory it needs.
• Arrays of a fixed type might be more memory-efficient than lists.
• If maxima is run by sbcl sbcl’s memory limit might be set to a value that is too low to solve the current problem. In this case the command-line option --dynamic-space-size <n> allows to tell sbcl to reserve n megabytes for the heap. It is to note, though, that sbcl has to handle several distinct types of memory and therefore might be able to only reserve about half of the available physical memory. Also note that 32-bit processes might only be able to access 2GB of physical memory.

91.1.12 part: fell off the end

part() was used to access the nth item in something that has less than n items.

91.1.13 undefined variable (draw or plot)

A function could not be plotted since it still contained a variable maxima doesn’t know the value of.

In order to find out which variable this could be it is sometimes helpful to temporarily replace the name of the drawing command (draw2d, plot2d or similar) by a random name (for example ddraw2d) that doesn’t coincide with the name of an existing function to make maxima print out what parameters the drawing command sees.

(%i1) load("draw")$
(%i2) f(x):=sin(omega*t);
(%o2) f(x) := sin(omega t)
(%i3) draw2d(
        explicit(
          f(x),
          x,1,10
        )
      );
draw2d (explicit): non defined variable
 -- an error. To debug this try: debugmode(true);
(%i4) ddraw2d(
        explicit(
          f(x),
          x,1,10
        )
      );
(%o4) ddraw2d(explicit(sin(omega t), x, 1, 10))

91.1.14 VTK is not installed, which is required for Scene

This might either mean that VTK is actually not installed - or cannot be found by maxima - or that maxima has no write access to the directory whose name is output if the following maxima command is entered:

(%i1) maxima_tempdir;
(%o1)                         /tmp

91.2 Warning messages

91.2.1 Encountered undefined variable <x> in translation

A function was compiled but the type of the variable x was not known. This means that the compiled command contains additional code that makes it retain all the flexibility maxima provides in respect to this variable. If x isn’t meant as a variable name but just a named option to a command prepending the named option by a single quote (') should resolve this issue.

91.2.2 Rat: replaced <x> by <y> = <z>

Floating-point numbers provide a maximum number of digits that is typically high, but still limited. Good examples that this limitation might be too low even for harmless-looking examples include Wilkinson’s Polynomial, The Rump polynomial and the fact that an exact 1/10 cannot be expressed as a binary floating-point number. In places where the floating-point error might add up or hinder terms from cancelling each other out maxima therefore by default replaces them with exact fractions. See also ratprint, ratepsilon, bftorat, fpprintprec and rationalize.