1 _____________________________________________________________________________
4 - Sensible tree-printout
5 - make Tib.Block extend Tree<>
9 - more natural phrasing of metagrammar?
11 - finalize metagrammar and rdp-op's
13 - Lay down the law on the different kinds of Sequence productions
19 - Deal with the problem of zero-rep productions and whitespace insertion
21 - switch maximal to not-followed-by (~/~)
23 - should Union.add() be there?
24 - should Atom.top() be there?
26 - fix the location stuff, it's broken
27 - decent/better error messages
33 - RFC2822 (email message/headers)
36 ______________________________________________________________________________
39 - clean up the whole Walk situation
41 - cleaner solution to "maximal"?
44 - right now I can only lift the last child in a forest... begs
45 the question of what the right representation for Forests is
46 if we need to be able to do lift operations on it.
49 - "Regular Right Part" grammars (NP Chapman, etc)
50 - Attribute unification
52 - serialization of parse tables
53 - inference of rejections for literals
54 - "prefer whitespace higher up" (?)
55 - "ambiguity modulo dropped fragments"?
56 - can this be checked statically?
57 - eliminated statically?
59 ______________________________________________________________________________
62 - Implement a k-token peek buffer (for each state, see if it "dead
63 ends" during the next k Phases based solely on state -- ignoring
66 - Arrange for the SPPF corresponding to dropped subtrees to never be
67 generated (or merged, etc)
69 - Is there any way we can avoid creating a GSS.Node instance for
70 nodes which are transient in the sense that they have only one
73 - Implement "GLR syntactic predicates" -- the ability to do
74 arbitrary lookahead (ie "followed-by" and "not-followed-by" for
75 arbitrary patterns). This enables generalized longest-match and
76 lets us drop the Maximal hack.
78 - Re-read Rekers, particularly the stuff on optimal sharing
80 - Isolate the Element objects from Parse.Table/GSS so we can move
83 - consider allowing a Forest.Body to represent some other Tree whose
84 Body's should be [recursively] considered part of this Forest.
86 - perhaps not: right now we have a nice situation where
87 Forest.Ref instances become immutable once iterator()ed. This
88 also gives us a strong place to to culling with the certainty
89 that we won't throw out a Body which would later be salvaged
90 by some yet-to-be-added dependency.
92 - Figure out if there is a way to:
94 - allow unwrapping of children other than the very last one.
96 - fold repetitions into an array form in Forest, before
97 conversion to Tree. The major problem here is that multiple
98 tree-arrays are possible, all of different lengths. Worse,
99 even if they're all the same length, not all elements belong
100 in the same "possibility vector" as all others. You
101 essentially need a GSS to represent the array, which perhaps
102 is what the unfolded form was in the first place.
104 - Wikipedia grammar (needs to be both lexerless and boolean)
107 => Ordered Choice (";" operator)
109 - bring back in parse-table phase resolution of precedence (just
110 like associativity). This can be inferred from the use of ">"
111 when the rules are in one of these special forms:
122 where "_" is anything and "E" is the defining nonterminal.
123 Essentially what we're looking for is the situation where the
124 leftmost portion of one rule produces another rule, and the
125 rightmost portion of the latter produces the former.
127 I'm not 100% certain that this is as "strong" as the prefer/avoid
128 form (try to prove this, you probably can), but it's "what people
129 intend" most of the time.
131 - implement Johnstone's algorithm for "reduced, resolved LR
132 tables" to eliminate superfluous reductions on
135 ______________________________________________________________________________
138 - Rekers & Koorn note that GLR Substring Parsing can be used to do
139 really elegant and generalized "autocompletion".