1 _____________________________________________________________________________
4 - needs/hates/follow API ugliness
6 - Topology crap is kinda messed up
8 - do Forest/Tree still need a Region?
9 - reconsider the degree of genericization
10 - GraphViz stuff pollutes the API...
11 - Forest needs a "manual access" API
12 - the unwrap bit in Forest makes it really hard to expose an API for forests
14 - evil problems with (x y? z /ws)
15 - ParseFailed, GSS, Walk, Parser, Sequence, Forest
19 - grammar highlighting?
20 - comment indentation vs block indentation?
22 - recursive { { foo } }
24 ______________________________________________________________________________
27 - finalize metagrammar and rdp-op's
32 - RFC2822 (email message/headers)
33 - clean up the whole Walk situation (?)
36 ______________________________________________________________________________
39 - serialization of parse tables
41 - "ambiguity modulo dropped fragments"?
42 - can this be checked statically?
43 - eliminated statically?
45 - substring parsing for better error messages
48 - "Regular Right Part" grammars (NP Chapman, etc)
49 - Attribute unification
51 - inference of rejections for literals
52 - "prefer whitespace higher up" (?)
54 - Labeled edges on trees (associate a label with each slot in the
55 child array in Forest.Body? might make equality tough) --
56 equivalent to Feature Structures. Colon-labeling.
58 ______________________________________________________________________________
61 - Partly-Linear-PATR? (O(n^6) unification grammar)
63 - Implement a k-token peek buffer (for each state, see if it "dead
64 ends" during the next k Phases based solely on state -- ignoring
67 - Arrange for the SPPF corresponding to dropped subtrees to never be
68 generated (or merged, etc)
70 - Is there any way we can avoid creating a GSS.Node instance for
71 nodes which are transient in the sense that they have only one
74 - Re-read Rekers, particularly the stuff on optimal sharing
76 - Isolate the Element objects from Parse.Table/GSS so we can move
79 - consider allowing a Forest.Body to represent some other Tree whose
80 Body's should be [recursively] considered part of this Forest.
82 - perhaps not: right now we have a nice situation where
83 Forest.Ref instances become immutable once iterator()ed. This
84 also gives us a strong place to to culling with the certainty
85 that we won't throw out a Body which would later be salvaged
86 by some yet-to-be-added dependency.
88 - Figure out if there is a way to:
90 - allow unwrapping of children other than the very last one.
92 - fold repetitions into an array form in Forest, before
93 conversion to Tree. The major problem here is that multiple
94 tree-arrays are possible, all of different lengths. Worse,
95 even if they're all the same length, not all elements belong
96 in the same "possibility vector" as all others. You
97 essentially need a GSS to represent the array, which perhaps
98 is what the unfolded form was in the first place.
100 - Wikipedia grammar (needs to be both lexerless and boolean)
103 => Ordered Choice (";" operator)
105 - bring back in parse-table phase resolution of precedence (just
106 like associativity). This can be inferred from the use of ">"
107 when the rules are in one of these special forms:
118 where "_" is anything and "E" is the defining nonterminal.
119 Essentially what we're looking for is the situation where the
120 leftmost portion of one rule produces another rule, and the
121 rightmost portion of the latter produces the former.
123 I'm not 100% certain that this is as "strong" as the prefer/avoid
124 form (try to prove this, you probably can), but it's "what people
125 intend" most of the time.
127 - implement Johnstone's algorithm for "reduced, resolved LR
128 tables" to eliminate superfluous reductions on
131 ______________________________________________________________________________
134 - Rekers & Koorn note that GLR Substring Parsing can be used to do
135 really elegant and generalized "autocompletion".
138 ______________________________________________________________________________
141 - Incremental parse table construction
142 - "lazy GLR" and "lazy trees" -> language with first-class CF matching
143 - perhaps linear boolean grammars instead? (linear time, quad space)
144 - Forest parsing => chained parsers
145 - unification parsing, attributes, etc
147 - Take another stab at maximal-match? Nonterminal not-followed-by is
149 - Error recovery based on substring parsing