celtic / rubyex
- Source
- Commits
- Network (0)
- Issues (0)
- Downloads (0)
- Wiki (1)
- Graphs
-
Tree:
f4b12ae
rubyex / TODO
| 969b8255 » | celtic | 2008-08-19 | 1 | Game plan: | |
| a2ecca38 » | celtic | 2008-10-01 | 2 | * finish regexps (interpol in r.e. literal) | |
| bc90d009 » | celtic | 2008-08-29 | 3 | * attr_* | |
| 4 | * *_eval | ||||
| a8fcf63c » | celtic | 2008-09-12 | 5 | * MRI compatibility layer | |
| 4a88901c » | celtic | 2008-09-13 | 6 | * working GC | |
| 2837bf59 » | celtic | 2008-08-22 | 7 | ||
| a8fcf63c » | celtic | 2008-09-12 | 8 | Shorter term problems: | |
| 9 | * catch/throw | ||||
| 10 | * Float range is small | ||||
| 11 | * RubyClass::new_instance(): these should not be centralised | ||||
| 12 | * assignments should always return rvalue | ||||
| 13 | * Fixnum, Symbol don't have instance variables | ||||
| 14 | * flex encounters push-black overflows due to brace nesting occasionally | ||||
| 15 | * Module::method syntax | ||||
| 16 | * '`' method impossible to declare due to lexer switching modes | ||||
| 17 | * Dir::glob needs to work to MRI's wildcard syntax | ||||
| 18 | * and/&&, or/|| need to not evaulate second condition if first fails/succeeds | ||||
| 19 | * `p 1 and 2' should parse to `(p 1) and 2' | ||||
| aa5943a4 » | celtic | 2008-09-13 | 20 | * if we `def a(b); end', then just `a' should raise an ArgumentError of not enough arguments | |
| 66103e8b » | celtic | 2008-09-20 | 21 | * expand hash-style on end to hash last argument | |
| 7a67a46f » | celtic | 2008-09-20 | 22 | * optional argument def a(b=(anything; can(); go = here)) syntax needs to work | |
| 59182ce4 » | celtic | 2008-09-21 | 23 | * finish fixing the edge cases when parser encounters syntax error causes memory leak (valgrind) | |
