use more efficient matrixLBH implementation for repair

breandan · Jan 15, 2025 · a800e06 · a800e06
1 parent 0e6ddfb
commit a800e06
Show file tree

Hide file tree

Showing 4 changed files with 21 additions and 1 deletion.
diff --git a/latex/bib/acmart.bib b/latex/bib/acmart.bib
@@ -2340,4 +2340,19 @@ @article{floyd1967nondeterministic
     pages={636--644},
     year={1967},
     publisher={ACM New York, NY, USA}
+}
+
+@Inbook{Grune2008,
+    author="Grune, Dick
+and Jacobs, Ceriel J. H.",
+    title="Parsing as Intersection",
+    bookTitle="Parsing Techniques: A Practical Guide",
+    year="2008",
+    publisher="Springer New York",
+    address="New York, NY",
+    pages="425--442",
+    abstract="In 1961 Bar-Hillel, Perles and Shamir [219] proved that ``the intersection of a context-free language with a regular language is again a context-free language''. On the face of it, this means that when we take the set of strings that constitute a given CF language and remove from it all strings that do not occur in a given FS language, we get a set of strings for which a CF grammar exists. Actually, it means quite a bit more.",
+    isbn="978-0-387-68954-8",
+    doi="10.1007/978-0-387-68954-8_13",
+    url="https://doi.org/10.1007/978-0-387-68954-8_13"
 }
diff --git a/src/commonMain/kotlin/ai/hypergraph/kaliningraph/automata/FSA.kt b/src/commonMain/kotlin/ai/hypergraph/kaliningraph/automata/FSA.kt
@@ -3,6 +3,7 @@ package ai.hypergraph.kaliningraph.automata
 import ai.hypergraph.kaliningraph.*
 import ai.hypergraph.kaliningraph.graphs.*
 import ai.hypergraph.kaliningraph.parsing.*
+import ai.hypergraph.kaliningraph.repair.MAX_RADIUS
 import ai.hypergraph.kaliningraph.types.*
 import kotlin.lazy
 import kotlin.random.Random
@@ -161,6 +162,9 @@ open class FSA constructor(open val Q: TSA, open val init: Set<Σᐩ>, open val
       return false
     }
 
+    fun LED(cfg: CFG, brokeToks: Σᐩ): Int =
+      (1 until MAX_RADIUS).firstOrNull { FSA.nonemptyLevInt(brokeToks, cfg, it) } ?: MAX_RADIUS
+
     fun intersectPTree(str: Σᐩ, cfg: CFG, radius: Int): PTree? {
       // 1) Build the Levenshtein automaton (acyclic)
       val levFSA = makeLevFSA(str, radius)

diff --git a/src/commonMain/kotlin/ai/hypergraph/kaliningraph/parsing/BarHillel.kt b/src/commonMain/kotlin/ai/hypergraph/kaliningraph/parsing/BarHillel.kt
@@ -176,7 +176,7 @@ fun CFG.dropVestigialProductions(
 
 
 fun CFG.LED(brokeToks: Σᐩ): Int =
-  (1..MAX_RADIUS).firstOrNull {
+  (1 until MAX_RADIUS).firstOrNull {
     try {
       intersectLevFSA(fsa = makeLevFSA(brokeToks.tokenizeByWhitespace(), it)).isNotEmpty()
     } catch (_: Exception) { println("Failed $it, increasing..."); false }

diff --git a/src/commonMain/kotlin/ai/hypergraph/kaliningraph/types/Graph.kt b/src/commonMain/kotlin/ai/hypergraph/kaliningraph/types/Graph.kt
@@ -268,6 +268,7 @@ val <G: IGraph<G, E, V>, E: IEdge<G, E, V>, V: IVertex<G, E, V>> IGraph<G, E, V>
 
 val cache = LRUCache<String, Any>()
 
+// If you see a JS error get_first_irdx8n_k, it DEPTH is set incorrectly or something is funny with the stacktrace
 object PlatformVars { var PLATFORM_CALLER_STACKTRACE_DEPTH: Int = 3 }
 // This is somewhat of a hack and may break depending on the platform.
 // We do this because Kotlin Common has poor reflection capabilities.