1 _____________________________________________________________________________
4 - Fix the metagrammar (really?)
6 - Repeat, Sequence, Tree
7 - simplify Forest (considerably)
9 - decent/better error messages
10 - fix the location stuff, it's broken
15 ______________________________________________________________________________
18 - finalize metagrammar and rdp-op's
23 - RFC2822 (email message/headers)
24 - clean up the whole Walk situation (?)
27 ______________________________________________________________________________
30 - serialization of parse tables
32 - "ambiguity modulo dropped fragments"?
33 - can this be checked statically?
34 - eliminated statically?
36 - substring parsing for better error messages
38 - right now I can only lift the last child in a forest... begs
39 the question of what the right representation for Forests is
40 if we need to be able to do lift operations on it.
43 - "Regular Right Part" grammars (NP Chapman, etc)
44 - Attribute unification
46 - inference of rejections for literals
47 - "prefer whitespace higher up" (?)
49 - Labeled edges on trees (associate a label with each slot in the
50 child array in Forest.Body? might make equality tough) --
51 equivalent to Feature Structures. Colon-labeling.
53 ______________________________________________________________________________
56 - Partly-Linear-PATR? (O(n^6) unification grammar)
58 - Implement a k-token peek buffer (for each state, see if it "dead
59 ends" during the next k Phases based solely on state -- ignoring
62 - Arrange for the SPPF corresponding to dropped subtrees to never be
63 generated (or merged, etc)
65 - Is there any way we can avoid creating a GSS.Node instance for
66 nodes which are transient in the sense that they have only one
69 - Re-read Rekers, particularly the stuff on optimal sharing
71 - Isolate the Element objects from Parse.Table/GSS so we can move
74 - consider allowing a Forest.Body to represent some other Tree whose
75 Body's should be [recursively] considered part of this Forest.
77 - perhaps not: right now we have a nice situation where
78 Forest.Ref instances become immutable once iterator()ed. This
79 also gives us a strong place to to culling with the certainty
80 that we won't throw out a Body which would later be salvaged
81 by some yet-to-be-added dependency.
83 - Figure out if there is a way to:
85 - allow unwrapping of children other than the very last one.
87 - fold repetitions into an array form in Forest, before
88 conversion to Tree. The major problem here is that multiple
89 tree-arrays are possible, all of different lengths. Worse,
90 even if they're all the same length, not all elements belong
91 in the same "possibility vector" as all others. You
92 essentially need a GSS to represent the array, which perhaps
93 is what the unfolded form was in the first place.
95 - Wikipedia grammar (needs to be both lexerless and boolean)
98 => Ordered Choice (";" operator)
100 - bring back in parse-table phase resolution of precedence (just
101 like associativity). This can be inferred from the use of ">"
102 when the rules are in one of these special forms:
113 where "_" is anything and "E" is the defining nonterminal.
114 Essentially what we're looking for is the situation where the
115 leftmost portion of one rule produces another rule, and the
116 rightmost portion of the latter produces the former.
118 I'm not 100% certain that this is as "strong" as the prefer/avoid
119 form (try to prove this, you probably can), but it's "what people
120 intend" most of the time.
122 - implement Johnstone's algorithm for "reduced, resolved LR
123 tables" to eliminate superfluous reductions on
126 ______________________________________________________________________________
129 - Rekers & Koorn note that GLR Substring Parsing can be used to do
130 really elegant and generalized "autocompletion".
133 ______________________________________________________________________________
136 - Incremental parse table construction
137 - "lazy GLR" and "lazy trees" -> language with first-class CF matching
138 - perhaps linear boolean grammars instead? (linear time, quad space)
139 - Forest parsing => chained parsers
140 - unification parsing, attributes, etc
142 - Take another stab at maximal-match? Nonterminal not-followed-by is
144 - Error recovery based on substring parsing