checkpoint

[sbp.git] / TODO

diff --git a/TODO b/TODO
index 78baf01..1185209 100644 (file)
--- a/TODO
+++ b/TODO
@@ -1,3 +1,33 @@
+// priorities are all messy and dont get serialized
+// 1. Error messages
+// 2. Java MetaGrammar (proof of concept)
+// 3. Ivan's MetaGrammar
+// 4. Documentation format
+//       - TIB
+
+// TODO: better API for interfacing with Java
+// TODO: error messages
+// TODO: integrate with TIB
+
+// Element
+// Walk
+// ParseTable / GSS
+// MetaGrammar (necessary/relevant?)
+// Tree<String> (cleanup?)
+// Union.SubUnion
+// Repeat
+
+// FEATURE: serialization of ParseTable's, generation of Java code
+// FEATURE: infer reject elements for literals
+// FEATURE: prefer whitespace higher up
+// FEATURE: full conjunctive and boolean grammars
+// FEATURE: "ambiguity modulo dropped fragments"?  can this be checked for statically?  eliminated statically?
+//            - drop stuff during the parsing process (drop nodes)
+
+// LATER: Element<A> -- parameterize over the input token type?  Makes a huge mess...
+// LATER: Go back to where Sequence is not an Element?
+//            - The original motivation for making Sequence "first class" was the fact that 
+//              in order to do associativity right you need to have per-Sequence follow sets
 
 ______________________________________________________________________________
 Immediately