From 2a18c0678ebd1632bb3d99dd247e3e55b5840c65 Mon Sep 17 00:00:00 2001
From: mikera <mike@mikera.net>
Date: Sat, 12 Oct 2024 13:45:29 +0100
Subject: [PATCH] More CAD3 encoding updates

---
 .../src/main/java/convex/core/data/ACell.java |  6 +--
 .../src/main/java/convex/core/data/ACode.java | 21 +++++++++-
 .../main/java/convex/core/data/ANonCVM.java   |  2 +-
 .../main/java/convex/core/data/Format.java    | 13 +++++--
 .../java/convex/core/data/impl/DummyCell.java |  2 +-
 .../java/convex/core/data/impl/KeySet.java    |  2 +-
 .../convex/core/data/prim/CVMBigInteger.java  |  2 +-
 .../main/java/convex/core/lang/ACVMCode.java  | 14 +++++++
 .../src/main/java/convex/core/lang/AFn.java   |  8 +---
 .../src/main/java/convex/core/lang/AOp.java   |  7 +---
 .../java/convex/core/lang/impl/ADataFn.java   | 39 ++++++-------------
 .../java/convex/core/lang/impl/KeywordFn.java |  5 +++
 .../java/convex/core/lang/impl/MapFn.java     |  5 +++
 .../java/convex/core/lang/impl/SeqFn.java     |  5 +++
 .../java/convex/core/lang/impl/SetFn.java     |  7 +++-
 .../java/convex/core/data/ObjectsTest.java    |  2 +
 16 files changed, 86 insertions(+), 54 deletions(-)
 create mode 100644 convex-core/src/main/java/convex/core/lang/ACVMCode.java

diff --git a/convex-core/src/main/java/convex/core/data/ACell.java b/convex-core/src/main/java/convex/core/data/ACell.java
index 9a5935a5a..eb730b2e2 100644
--- a/convex-core/src/main/java/convex/core/data/ACell.java
+++ b/convex-core/src/main/java/convex/core/data/ACell.java
@@ -183,7 +183,7 @@ protected static boolean genericEquals(ACell a, ACell b) {
 	 * @param pos The offset into the byte array
 	 * @return New position after writing
 	 */
-	protected abstract int encodeRaw(byte[] bs, int pos);
+	public abstract int encodeRaw(byte[] bs, int pos);
 	
 	/**
 	 * Creates the encoding for this cell. Cell must be canonical, or else an error may occur.
@@ -399,11 +399,11 @@ protected <R extends ACell> Ref<R> createRef() {
 	
 	/**
 	 * Gets the number of Refs contained within this Cell. This number is
-	 * final / immutable for any given instance and is defined by the Cell encoding rules.
+	 * final / immutable for any given instance and is defined by the Cell,s CAD3 encoding rules.
 	 * 
 	 * Contained Refs may be either external or embedded.
 	 * 
-	 * @return The number of Refs in this Cell
+	 * @return The number of Refs in this Cell (0-63)
 	 */
 	public int getRefCount() {
 		ACell canonical=getCanonical();
diff --git a/convex-core/src/main/java/convex/core/data/ACode.java b/convex-core/src/main/java/convex/core/data/ACode.java
index 791aeba28..4a067a8d2 100644
--- a/convex-core/src/main/java/convex/core/data/ACode.java
+++ b/convex-core/src/main/java/convex/core/data/ACode.java
@@ -1,8 +1,27 @@
 package convex.core.data;
 
+import convex.core.data.impl.ALongBlob;
+import convex.core.util.Utils;
+import convex.core.exceptions.InvalidDataException;
+
 /**
  * Abstract base class for CVM code data types
  */
-public abstract class ACode extends ACell {
+public abstract class ACode extends ALongBlob {
+
+	protected byte tag;
+	
+	protected ACode(long value) {
+		super(value);
+	}
 
+	@Override
+	public void validateCell() throws InvalidDataException {
+		if ((byte)(this.tag&0xF0)!=Tag.CODE_BASE) {
+			throw new InvalidDataException("Invalide Code tag: 0x"+Utils.toHexString(tag),this);
+		}
+		if (this.value<0) {
+			throw new InvalidDataException("Negaitive code value",this);
+		}
+	}
 }
diff --git a/convex-core/src/main/java/convex/core/data/ANonCVM.java b/convex-core/src/main/java/convex/core/data/ANonCVM.java
index f1fbac4d8..b6a8b736e 100644
--- a/convex-core/src/main/java/convex/core/data/ANonCVM.java
+++ b/convex-core/src/main/java/convex/core/data/ANonCVM.java
@@ -39,7 +39,7 @@ public int encode(byte[] bs, int pos) {
 	}
 
 	@Override
-	protected int encodeRaw(byte[] bs, int pos) {
+	public int encodeRaw(byte[] bs, int pos) {
 		encoding.slice(1).getBytes(bs, pos);
 		return pos+encoding.size()-1;
 	}
diff --git a/convex-core/src/main/java/convex/core/data/Format.java b/convex-core/src/main/java/convex/core/data/Format.java
index 6bdfc296c..4e3e30185 100644
--- a/convex-core/src/main/java/convex/core/data/Format.java
+++ b/convex-core/src/main/java/convex/core/data/Format.java
@@ -43,10 +43,9 @@
 import convex.core.util.Utils;
 
 /**
- * Static utility class for message format encoding
+ * Static utility class for CAD3 encoding format
  *
- * "Standards are always out of date. That's what makes them standards." - Alan
- * Bennett
+ * "Standards are always out of date. That's what makes them standards." - Alan Bennett
  */
 public class Format {
 
@@ -57,7 +56,8 @@ public class Format {
 	 * <ul>
 	 * <li>This is the maximum length that can be VLQ encoded in 2 bytes. This simplifies message encoding and decoding.</li>
 	 * <li>It is big enough to include a 4096-byte Blob</li>
-	 * <li>It is small enough to fit in a UDP message</li>
+	 * <li>It is big enough to include a Record with 63 directly referenced fields</li>
+	 * <li>It is small enough to guarantee fitting in a UDP message</li>
 	 * </ul>
 	 */
 	public static final int LIMIT_ENCODING_LENGTH = 0x3FFF; 
@@ -97,6 +97,11 @@ public class Format {
 	 */
 	public static final long FULL_EMBEDDED_MEMORY_SIZE = 0L;
 
+	/**
+	 * Maximum number of Refs for any single Cell
+	 */
+	public static final int MAX_REF_COUNT = 63;
+
 	/**
 	 * Gets the length in bytes of VLQ encoding for the given long value
 	 * @param x Long value to encode
diff --git a/convex-core/src/main/java/convex/core/data/impl/DummyCell.java b/convex-core/src/main/java/convex/core/data/impl/DummyCell.java
index 4c9f9ac4f..273c5f301 100644
--- a/convex-core/src/main/java/convex/core/data/impl/DummyCell.java
+++ b/convex-core/src/main/java/convex/core/data/impl/DummyCell.java
@@ -37,7 +37,7 @@ public int encode(byte[] bs, int pos) {
 	}
 	
 	@Override
-	protected int encodeRaw(byte[] bs, int pos) {
+	public int encodeRaw(byte[] bs, int pos) {
 		// Nothing to encode
 		return pos;
 	}
diff --git a/convex-core/src/main/java/convex/core/data/impl/KeySet.java b/convex-core/src/main/java/convex/core/data/impl/KeySet.java
index 8aec9b648..1b55ad89b 100644
--- a/convex-core/src/main/java/convex/core/data/impl/KeySet.java
+++ b/convex-core/src/main/java/convex/core/data/impl/KeySet.java
@@ -137,7 +137,7 @@ public boolean equals(ACell a) {
 	}
 
 	@Override
-	protected int encodeRaw(byte[] bs, int pos) {
+	public int encodeRaw(byte[] bs, int pos) {
 		throw new UnsupportedOperationException();
 	}
 
diff --git a/convex-core/src/main/java/convex/core/data/prim/CVMBigInteger.java b/convex-core/src/main/java/convex/core/data/prim/CVMBigInteger.java
index 3e5d735e8..9413e92d1 100644
--- a/convex-core/src/main/java/convex/core/data/prim/CVMBigInteger.java
+++ b/convex-core/src/main/java/convex/core/data/prim/CVMBigInteger.java
@@ -182,7 +182,7 @@ public int encode(byte[] bs, int pos) {
 	}
 
 	@Override
-	protected int encodeRaw(byte[] bs, int pos) {
+	public int encodeRaw(byte[] bs, int pos) {
 		ABlob b=blob();
 		return b.encodeRaw(bs, pos);
 	}
diff --git a/convex-core/src/main/java/convex/core/lang/ACVMCode.java b/convex-core/src/main/java/convex/core/lang/ACVMCode.java
new file mode 100644
index 000000000..35ab876e0
--- /dev/null
+++ b/convex-core/src/main/java/convex/core/lang/ACVMCode.java
@@ -0,0 +1,14 @@
+package convex.core.lang;
+
+import convex.core.data.ACell;
+
+/**
+ * Abstract base class for CVM code constructs
+ */
+public abstract class ACVMCode extends ACell {
+
+	@Override public final boolean isCVMValue() {
+		// CVM code objects are CVM values by definition
+		return true;
+	}
+}
diff --git a/convex-core/src/main/java/convex/core/lang/AFn.java b/convex-core/src/main/java/convex/core/lang/AFn.java
index 61668a841..d277feffc 100644
--- a/convex-core/src/main/java/convex/core/lang/AFn.java
+++ b/convex-core/src/main/java/convex/core/lang/AFn.java
@@ -1,7 +1,6 @@
 package convex.core.lang;
 
 import convex.core.data.ACell;
-import convex.core.data.ACode;
 import convex.core.data.IRefFunction;
 import convex.core.data.Tag;
 import convex.core.data.type.AType;
@@ -15,7 +14,7 @@
  *
  * @param <T> Return type of functions.
  */
-public abstract class AFn<T extends ACell> extends ACode implements IFn<T> {
+public abstract class AFn<T extends ACell> extends ACVMCode implements IFn<T> {
 	
 	@Override
 	public abstract Context invoke(Context context, ACell[] args);
@@ -48,11 +47,6 @@ public boolean supportsArgs(ACell[] args) {
 	 */
 	public abstract boolean hasArity(int arity);
 	
-	@Override 
-	public boolean isCVMValue() {
-		return true;
-	}
-	
 	@Override
 	public byte getTag() {
 		return Tag.FN;
diff --git a/convex-core/src/main/java/convex/core/lang/AOp.java b/convex-core/src/main/java/convex/core/lang/AOp.java
index 2960eccca..0ea8ec831 100644
--- a/convex-core/src/main/java/convex/core/lang/AOp.java
+++ b/convex-core/src/main/java/convex/core/lang/AOp.java
@@ -1,7 +1,6 @@
 package convex.core.lang;
 
 import convex.core.data.ACell;
-import convex.core.data.ACode;
 import convex.core.data.Format;
 import convex.core.data.IRefFunction;
 import convex.core.data.Tag;
@@ -20,7 +19,7 @@
  *
  * @param <T> the type of the operation return value
  */
-public abstract class AOp<T extends ACell> extends ACode {
+public abstract class AOp<T extends ACell> extends ACVMCode {
 
 	/**
 	 * Executes this op with the given context. Must preserve depth unless an
@@ -51,10 +50,6 @@ public boolean isCanonical() {
 	public ACell toCanonical() {
 		return this;
 	}
-	
-	@Override public final boolean isCVMValue() {
-		return true;
-	}
 
 	/**
 	 * Returns the opcode for this op
diff --git a/convex-core/src/main/java/convex/core/lang/impl/ADataFn.java b/convex-core/src/main/java/convex/core/lang/impl/ADataFn.java
index 66ad06032..0ca425525 100644
--- a/convex-core/src/main/java/convex/core/lang/impl/ADataFn.java
+++ b/convex-core/src/main/java/convex/core/lang/impl/ADataFn.java
@@ -4,12 +4,12 @@
 import convex.core.data.IRefFunction;
 import convex.core.exceptions.InvalidDataException;
 import convex.core.lang.AFn;
-import convex.core.lang.Context;
+import convex.core.lang.RT;
 
 /**
- * Abstract base class for data structure lookup functions.
+ * Abstract base wrapper class for data structure lookup functions.
  * 
- * Not a canonical object, can't exist as CVM value.
+ * Not a canonical object, essentially a wrapper for a data structure interpreted as a function
  * 
  * @param <T> Type of function return value
  */
@@ -17,37 +17,32 @@ public abstract class ADataFn<T extends ACell> extends AFn<T> {
 
 	@Override
 	public int estimatedEncodingSize() {
-		throw new UnsupportedOperationException();
-	}
-
-	@Override
-	public Context invoke(Context context, ACell[] args) {
-		throw new UnsupportedOperationException();
+		return getCanonical().estimatedEncodingSize();
 	}
 
 	@Override
 	public AFn<T> updateRefs(IRefFunction func) {
-		throw new UnsupportedOperationException();
+		return RT.castFunction(getCanonical().updateRefs(func));
 	}
 
 	@Override
 	public boolean hasArity(int n) {
-		throw new UnsupportedOperationException();
+		return (n==1)||(n==2);
 	}
 
 	@Override
 	public void validateCell() throws InvalidDataException {
-		throw new UnsupportedOperationException();
+		getCanonical().validateCell();;
 	}
 
 	@Override
 	public int encode(byte[] bs, int pos) {
-		throw new UnsupportedOperationException();
+		return getCanonical().encode(bs, pos);
 	}
 
 	@Override
 	public int encodeRaw(byte[] bs, int pos) {
-		throw new UnsupportedOperationException();
+		return getCanonical().encodeRaw(bs, pos);
 	}
 
 	@Override
@@ -56,22 +51,10 @@ public boolean isCanonical() {
 	}
 	
 	@Override
-	public ACell toCanonical() {
-		throw new UnsupportedOperationException("Can't make canonical!");
-	}
-	
-	@Override
-	public boolean isCVMValue() {
-		return false;
-	}
+	public abstract ACell toCanonical();
 	
 	@Override
 	public byte getTag() {
-		throw new UnsupportedOperationException();
-	}
-
-	@Override
-	public int getRefCount() {
-		throw new UnsupportedOperationException();
+		return getCanonical().getTag();
 	}
 }
diff --git a/convex-core/src/main/java/convex/core/lang/impl/KeywordFn.java b/convex-core/src/main/java/convex/core/lang/impl/KeywordFn.java
index 05b8630f9..819a7c1f4 100644
--- a/convex-core/src/main/java/convex/core/lang/impl/KeywordFn.java
+++ b/convex-core/src/main/java/convex/core/lang/impl/KeywordFn.java
@@ -49,5 +49,10 @@ public boolean print(BlobBuilder sb,long limit) {
 		return key.print(sb,limit);
 	}
 
+	@Override
+	public ACell toCanonical() {
+		return key.getCanonical();
+	}
+
 
 }
diff --git a/convex-core/src/main/java/convex/core/lang/impl/MapFn.java b/convex-core/src/main/java/convex/core/lang/impl/MapFn.java
index 4dd43034f..1da477f4c 100644
--- a/convex-core/src/main/java/convex/core/lang/impl/MapFn.java
+++ b/convex-core/src/main/java/convex/core/lang/impl/MapFn.java
@@ -39,4 +39,9 @@ public boolean print(BlobBuilder sb,long limit) {
 		return map.print(sb,limit);
 	}
 
+	@Override
+	public ACell toCanonical() {
+		return map.getCanonical();
+	}
+
 }
diff --git a/convex-core/src/main/java/convex/core/lang/impl/SeqFn.java b/convex-core/src/main/java/convex/core/lang/impl/SeqFn.java
index 8714f49ce..4ef2ac1d2 100644
--- a/convex-core/src/main/java/convex/core/lang/impl/SeqFn.java
+++ b/convex-core/src/main/java/convex/core/lang/impl/SeqFn.java
@@ -54,4 +54,9 @@ public boolean print(BlobBuilder sb,long limit) {
 		return seq.print(sb,limit);
 	}
 
+	@Override
+	public ACell toCanonical() {
+		return seq.getCanonical();
+	}
+
 }
diff --git a/convex-core/src/main/java/convex/core/lang/impl/SetFn.java b/convex-core/src/main/java/convex/core/lang/impl/SetFn.java
index d7182d87d..29bbbd56e 100644
--- a/convex-core/src/main/java/convex/core/lang/impl/SetFn.java
+++ b/convex-core/src/main/java/convex/core/lang/impl/SetFn.java
@@ -10,7 +10,7 @@ public class SetFn<T extends ACell> extends ADataFn<CVMBool> {
 
 	private ASet<T> set;
 
-	public SetFn(ASet<T> m) {
+	SetFn(ASet<T> m) {
 		this.set = m;
 	}
 
@@ -35,4 +35,9 @@ public boolean print(BlobBuilder sb,long limit) {
 		return set.print(sb,limit);
 	}
 
+	@Override
+	public ACell toCanonical() {
+		return set.getCanonical();
+	}
+
 }
diff --git a/convex-core/src/test/java/convex/core/data/ObjectsTest.java b/convex-core/src/test/java/convex/core/data/ObjectsTest.java
index e1220fdb0..92d941df3 100644
--- a/convex-core/src/test/java/convex/core/data/ObjectsTest.java
+++ b/convex-core/src/test/java/convex/core/data/ObjectsTest.java
@@ -406,7 +406,9 @@ private static void doCellStorageTest(ACell a) throws InvalidDataException, IOEx
 	private static void doRefContainerTests(ACell a) {
 		if (!a.isCanonical()) return;
 		int rc=a.getRefCount();
+	
 		assertTrue(rc>=0);
+		assertTrue(rc<=Format.MAX_REF_COUNT);
 		assertEquals(rc,Cells.refCount(a));
 		if (rc>0) {
 			long tcount = Refs.totalRefCount(a);