1 This is a living document - delete things when done. Avoid discussion.
4 Current version (Cataract Creek)
5 - handle free_type more cleanly??
6 - change 'labels' to only be created for 'use'
7 - handle syntax errors better - recover quickly.
8 - use precedence levels for expressions
11 - anonymous field - array or struct (or pointer to these)
12 multiple anon struct ar allowed if they don't conflict
13 - [] can apply to anon array field
14 - anon struct field gets fields interpolated
16 - manifest values for arrays and structs [a,b,c]
21 - split values so I can have an array of just the value (1 byte for u8)
22 - integers, unsigned, bitfield, float
25 - pure, loaded, overloaded, augmented
26 - owned: once, counted, collected
28 - array buffer - can be added to and grows.
29 - char, string search, regexp search
31 - allow "do stuff" as a stand-alone statement (scope)
33 Next version (Govetts Creek):
34 - functions and procedures
35 - Finalize what a "main" program looks like.
38 - per-field attributes
39 constant, stable, add-only, read-only, mutable, owned, borrowed, dependant, pure
43 - constructors and destructors - or "after" ??
44 - vtables, fat pointers, list of approaches
45 - operators as interface methods
46 - interfaces, inheritance
47 - modules, imports and exports
48 - closures, threads, co-routines, generators
49 - introspection / reflection ?
50 e.g. support serialization
51 find function given a string holding the name
52 measure coverage, adjust based on performance metrics
53 auto-create mock objects
54 Is this just parsing the details in the obj file?
56 - string / slice / strbuf
57 - parsing library - sscanf equiv, regexp, LALR
60 - sockets / http / HTML
61 - parameterised types, and dependant types
62 - message passing primitives
63 - overloading for numbers
65 - ensure list_head type concept can work
67 - pattern matching for destructuring??
70 - iso suffixes for number?
72 - case fall-through or go-to
73 - break/continue or "next","last"