FMSLogo Code

Refactoring: Slightly simplify the logic for handling the return value of a macro.

This moves:
if (unev == Unbound)
{
assign(unev, NIL);
}

To the branch where this change matters.

It also precomputes its effect instead of assigning env twice.

Note that I haven't confirmed that unev can be Unbound.

Fix Bug #609; Procedures are now re-treeified if they are:
- redefined with DEFINE
- redefined with TO
- erased with ERASE
- overwritten with COPYDEF

The problem is that these workspace procedures all untreeified the previous procedure.
This was historically needed because the tree had a circular reference and changing
the case object's procnode while it still had a tree would leak the tree. However,
the evalulator can't handle when a running procedure is untreeified and the generation
changes, as this sets the tree of bodylist to NIL, and the evaluator can't re-treeify NIL.

A recent change reworked the structure of a tree to no longer creates circular references.
Therefore, explicit untreeifying is no longer needed, so that logic could all be removed.

This is a rare fix where the fix is entirely done by removing code.

Add a comment that eval.lgo also tests the treeification process in paren.cpp.

Fix a comment

Fix a typo in a comment

Add LOADTEST.REDEFINEWHILETREEIFYING, a regression test for bug #249.

Bug #249 is a memory leak when treeiying a procedure (which happens when it's first run) causes a file to load that redefines that procedure.

This is similar to EVALTEST.SILENTLOADFROMTREEIFY, except that this test is designed only to test the memory leak.

Add assert(!is_freed(nd)) to nodetype().
This required making is_freed return false when given NIL.
This was done to make freed memory use bugs easier to find.

Fix a memory leak in LPUT when it had bad input and the :erract handler returned a newly allocated list.
The leak was introduced in recent refactoring when llput was made to use validate_list_argument().
The problem is that the list NODE* that should have been gcref'd was also used to walk the list,
so NIL ended up being gcref'd instead.

Refactoring: inline untreeify()
After recent changes, it was only called from untreeify_body() and that function only called untreeify().
Therefore, the existence of both functions wasn't justified.
I retained the function with the stronger name.

Cleanup: Remove unused function untreeify_line.
This because unneeded when treeifying a proc's bodylist stopped changing each body line into a TREE.