mathics3 / mathics-development-guide Goto Github PK

Hi, I am new to mathics. When reading the document of how to extend mathics, I found the description of usage of Builtin and Predefined maybe incorrect.
In chapter of evaluation.rst, it says

Functions which take no parameters are generally subclassed off of the Builtin class.
However when a function takes parameters it method’s Object class is derived either directly indirectly from the Predefined class rather than the Builtin. To figure out which apply method in the class object to call, each method’s document string (or docstring) is consulted. The lookup process is kicked off using the evaluate() method found in the Predefined class.

And when I contiue to read the document of Hello, World! – using Predefined and Introducing the Builtin Class also with the source code of Builtin and Predefined class, I found the description above is quite opposite.

https://github.com/Mathics3/mathics-development-guide/blob/6b5e3c92089eeed69528ba005afe04e5a4ae94d0/docs/extending/developing-code/code-overview/evaluation.rst

In calling-mathics-library-from-within-python-code, section Mathics3 Results as Python Objects

It seems that the example code should be:

# This is the same as before
from mathics.session import MathicsSession
session = MathicsSession(catch_interrupt=True)

# These are Mathics3 classes we are going to use.
from mathics.core.expression import Expression, Integer
from mathics.core.systemsymbols import SymbolFactorial

# Compute 20!
x = Expression(SymbolFactorial, Integer(10)
              ).evaluate(session.evaluation).to_python()
print(x) # 2432902008176640000

and

# This is the same as before
from mathics.session import MathicsSession
session = MathicsSession(catch_interrupt=True)

# These are Mathics3 classes we are going to use.
from mathics.core.expression import Expression, Integer
from mathics.core.symbols import SymbolPlus, SymbolTimes

# Compute 5 * (6 + 3)
x = Expression(SymbolTimes, Integer(5),
      Expression(SymbolPlus, Integer(6), Integer(3))
      ).evaluate(session.evaluation).to_python()
print(x) # 45