X-Git-Url: http://git.megacz.com/?p=sbp.git;a=blobdiff_plain;f=TODO;h=3e67a2ea84ad521ccecfe58a5b03f8e7641fb70a;hp=a1510616f995bb046ecf7dadd4755c803f6cd73b;hb=c4431d19cc5ddaae29d22c8c56366b53b0bad352;hpb=9ca5aa82948f06917a81983c677e0f8b9cc86af3

diff --git a/TODO b/TODO
index a151061..3e67a2e 100644
--- a/TODO
+++ b/TODO
@@ -1,46 +1,49 @@
 _____________________________________________________________________________
 Immediately
 
-  - What is our use model?
-    - Parse, attribute, unify       (ag)
-    - Parse, transform tree, walk   (rdp)
-    - Parse, walk
-    - cascading tree rewrites
+  - Performance
 
-      ==> use the middle formalism "for now" and in meta.g; layer others
-          on top of it
+     - Next target: TopologicalBag (make it wickedfast: preoptimize)
 
-  - Lay down the law on the different kinds of Sequence productions
-    and how they work.
+     - Forest: keep() and valid() -- can we do this with states
+       rather than subtrees?
 
-     => mydrop
-     => mylift
+     - hash Long->long: it's all bogus
 
-  - whitespace-in-braces?
-  - Deal with the problem of zero-rep productions and whitespace insertion
+  * 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'
+
+
+  - Sensible tree-printout
+  - make Tib.Block extend Tree<>
 
-  - switch maximal to not-followed-by (~/~)
+  - more natural phrasing of metagrammar?
+
+  - finalize metagrammar and rdp-op's
+
+  - Deal with the problem of zero-rep productions and whitespace insertion
 
   - should Union.add() be there?
   - should Atom.top() be there?
 
   - fix the location stuff, it's broken
   - decent/better error messages
+      - substring parsing required
 
   - write some grammars
       - Java grammar
       - TeX (math?)
       - URL (RFC)
       - RFC2822 (email message/headers)
-      - Wiki grammar
+
+  - PL-PATR?
 
 ______________________________________________________________________________
 Soon
 
   - clean up the whole Walk situation
 
-  - cleaner solution to "maximal"?
-
   - "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
@@ -71,11 +74,6 @@ Later
     nodes which are transient in the sense that they have only one
     eligible reduction?
 
-  - Implement "GLR syntactic predicates" -- the ability to do
-    arbitrary lookahead (ie "followed-by" and "not-followed-by" for
-    arbitrary patterns).  This enables generalized longest-match and
-    lets us drop the Maximal hack.
-
   - Re-read Rekers, particularly the stuff on optimal sharing
 
   - Isolate the Element objects from Parse.Table/GSS so we can move
@@ -132,3 +130,9 @@ Later
   - implement Johnstone's algorithm for "reduced, resolved LR
     tables" to eliminate superfluous reductions on
     epsilon-transitions.
+
+______________________________________________________________________________
+Neat Ideas
+
+  - Rekers & Koorn note that GLR Substring Parsing can be used to do
+    really elegant and generalized "autocompletion".