checkpoint

[sbp.git] / TODO

diff --git a/TODO b/TODO
index 3e67a2e..1e53bb8 100644 (file)
--- a/TODO
+++ b/TODO
@@ -1,49 +1,57 @@
 _____________________________________________________________________________
 Immediately
 
-  - Performance
+- If a top-level rule has labels but no head-tag, like this
+    Foo = a:Bar b:Baz
+  then infer the name of the rule it belongs to
 
-     - Next target: TopologicalBag (make it wickedfast: preoptimize)
+create( $c:{...}, class ) =
+  return create($c:{...})
 
-     - Forest: keep() and valid() -- can we do this with states
-       rather than subtrees?
+create( h:{...}, class ) = 
 
-     - hash Long->long: it's all bogus
+create( , String)
 
-  * huge performance improvement (try for more)
-  * pick back up cleaning up end of Parser.java (Reduction)
-  * some weird edge cases; check last regression test, 'make doc'
+create( _:{...}, String) = treat as char[]
+create( _:{...}, c[] )   = { create(.,c), create(.,c), ... }
 
+create( $c:{...} ) =
 
-  - Sensible tree-printout
-  - make Tib.Block extend Tree<>
 
-  - more natural phrasing of metagrammar?
+  - better ambiguity debugging tools / visualization
 
-  - finalize metagrammar and rdp-op's
+  - ParseFailed, GSS, Walk, Parser, Sequence, Forest
 
-  - Deal with the problem of zero-rep productions and whitespace insertion
+  - Fix the metagrammar (really?)
+      - evil problems with      (x y? z /ws)
 
-  - should Union.add() be there?
-  - should Atom.top() be there?
+  - copyright notices
+  - documentation
 
-  - fix the location stuff, it's broken
-  - decent/better error messages
-      - substring parsing required
+______________________________________________________________________________
+v1.1
 
+  - finalize metagrammar and rdp-op's
   - write some grammars
       - Java grammar
       - TeX (math?)
       - URL (RFC)
       - RFC2822 (email message/headers)
+  - clean up the whole Walk situation (?)
+
+  - what if Tree<> could unwrap itself?
 
-  - PL-PATR?
 
 ______________________________________________________________________________
 Soon
 
-  - clean up the whole Walk situation
+  - serialization of parse tables
+
+  - "ambiguity modulo dropped fragments"?
+       - can this be checked statically?
+       - eliminated statically?
 
+  - substring parsing for better error messages
   - "lift" cases:
       - right now I can only lift the last child in a forest...  begs
         the question of what the right representation for Forests is
@@ -53,16 +61,18 @@ Soon
   - "Regular Right Part" grammars (NP Chapman, etc)
   - Attribute unification
 
-  - serialization of parse tables
   - inference of rejections for literals
   - "prefer whitespace higher up" (?)
-  - "ambiguity modulo dropped fragments"?
-       - can this be checked statically?
-       - eliminated statically?
+
+  - Labeled edges on trees (associate a label with each slot in the
+    child array in Forest.Body?  might make equality tough) --
+    equivalent to Feature Structures.  Colon-labeling.
 
 ______________________________________________________________________________
 Later
 
+  - Partly-Linear-PATR? (O(n^6) unification grammar)
+
   - Implement a k-token peek buffer (for each state, see if it "dead
     ends" during the next k Phases based solely on state -- ignoring
     result SPPF)
@@ -136,3 +146,17 @@ Neat Ideas
 
   - Rekers & Koorn note that GLR Substring Parsing can be used to do
     really elegant and generalized "autocompletion".
+
+
+______________________________________________________________________________
+Ideas for the Future
+
+- Incremental parse table construction
+- "lazy GLR" and "lazy trees" -> language with first-class CF matching
+    - perhaps linear boolean grammars instead? (linear time, quad space)
+- Forest parsing => chained parsers
+- unification parsing, attributes, etc
+- RRP grammars?
+- Take another stab at maximal-match?  Nonterminal not-followed-by is
+  too strong.
+- Error recovery based on substring parsing