1 _____________________________________________________________________________
4 - evil problems with: (x y? z /ws)
5 - it gets even more evil than that
9 - MUST HAVE BETTER ERROR MESSAGES
10 - use for developing java15.g
14 - topology no longer needed as an arg to parser
15 - expose parser's protected method?
17 - do Forest/Tree still need a Region?
21 ______________________________________________________________________________
24 - More topology untangling [later]
25 - tib: use the lexer only for indentation increases/decreases
26 - grammar highlighting?
28 - Forest needs a "manual access" API
29 - the unwrap bit in Forest makes it really hard to expose an API for forests
33 ______________________________________________________________________________
36 - finalize metagrammar and rdp-op's
41 - RFC2822 (email message/headers)
42 - clean up the whole Walk situation (?)
45 ______________________________________________________________________________
48 - serialization of parse tables
50 - "ambiguity modulo dropped fragments"?
51 - can this be checked statically?
52 - eliminated statically?
54 - substring parsing for better error messages
57 - "Regular Right Part" grammars (NP Chapman, etc)
58 - Attribute unification
60 - inference of rejections for literals
61 - "prefer whitespace higher up" (?)
63 - Labeled edges on trees (associate a label with each slot in the
64 child array in Forest.Body? might make equality tough) --
65 equivalent to Feature Structures. Colon-labeling.
67 ______________________________________________________________________________
70 - Partly-Linear-PATR? (O(n^6) unification grammar)
72 - Implement a k-token peek buffer (for each state, see if it "dead
73 ends" during the next k Phases based solely on state -- ignoring
76 - Arrange for the SPPF corresponding to dropped subtrees to never be
77 generated (or merged, etc)
79 - Is there any way we can avoid creating a GSS.Node instance for
80 nodes which are transient in the sense that they have only one
83 - Re-read Rekers, particularly the stuff on optimal sharing
85 - Isolate the Element objects from Parse.Table/GSS so we can move
88 - consider allowing a Forest.Body to represent some other Tree whose
89 Body's should be [recursively] considered part of this Forest.
91 - perhaps not: right now we have a nice situation where
92 Forest.Ref instances become immutable once iterator()ed. This
93 also gives us a strong place to to culling with the certainty
94 that we won't throw out a Body which would later be salvaged
95 by some yet-to-be-added dependency.
97 - Figure out if there is a way to:
99 - allow unwrapping of children other than the very last one.
101 - fold repetitions into an array form in Forest, before
102 conversion to Tree. The major problem here is that multiple
103 tree-arrays are possible, all of different lengths. Worse,
104 even if they're all the same length, not all elements belong
105 in the same "possibility vector" as all others. You
106 essentially need a GSS to represent the array, which perhaps
107 is what the unfolded form was in the first place.
109 - Wikipedia grammar (needs to be both lexerless and boolean)
112 => Ordered Choice (";" operator)
114 - bring back in parse-table phase resolution of precedence (just
115 like associativity). This can be inferred from the use of ">"
116 when the rules are in one of these special forms:
127 where "_" is anything and "E" is the defining nonterminal.
128 Essentially what we're looking for is the situation where the
129 leftmost portion of one rule produces another rule, and the
130 rightmost portion of the latter produces the former.
132 I'm not 100% certain that this is as "strong" as the prefer/avoid
133 form (try to prove this, you probably can), but it's "what people
134 intend" most of the time.
136 - implement Johnstone's algorithm for "reduced, resolved LR
137 tables" to eliminate superfluous reductions on
140 ______________________________________________________________________________
143 - Rekers & Koorn note that GLR Substring Parsing can be used to do
144 really elegant and generalized "autocompletion".
147 ______________________________________________________________________________
150 - Incremental parse table construction
151 - "lazy GLR" and "lazy trees" -> language with first-class CF matching
152 - perhaps linear boolean grammars instead? (linear time, quad space)
153 - Forest parsing => chained parsers
154 - unification parsing, attributes, etc
156 - Take another stab at maximal-match? Nonterminal not-followed-by is
158 - Error recovery based on substring parsing