correct version

also fix string assignment bug for sub args.

.github/workflows/all-ci.yml

@@ -10,7 +10,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: build and install recent 64tass
         run: |
@@ -21,16 +21,31 @@ jobs:
           sudo make install
 
       - name: Set up JDK 11
-        uses: actions/setup-java@v3
+        uses: actions/setup-java@v4
         with:
           java-version: 11
           distribution: adopt
 
       - name: Build and test with Gradle
         run: ./gradlew build shadowJar --no-daemon
-
+        
       - name: Create compiler shadowJar artifact
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
           name: prog8-compiler-jar-zipped
           path: compiler/build/libs/*-all.jar
+
+      - name: Calculate hash
+        uses: MCJack123/ghaction-generate-release-hashes@v4
+        if: github.event_name == 'release'
+        with:
+          get-assets: true
+          hash-type: sha256
+          file-name: hashes.txt
+
+      - name: Upload hashes
+        uses: actions/upload-artifact@v4
+        if: github.event_name == 'release'
+        with:
+          name: Artifact Hashes
+          path: hashes.txt        

.gitignore

@@ -30,7 +30,6 @@ parsetab.py
 compiler/lib/
 
 .gradle
-**/BuildVersion.kt
 /prog8compiler.jar
 sd*.img
 *.d64

.idea/inspectionProfiles/Project_Default.xml

@@ -9,6 +9,15 @@
     </inspection_tool>
     <inspection_tool class="IncompleteDestructuring" enabled="true" level="WARNING" enabled_by_default="true" />
     <inspection_tool class="PyInterpreterInspection" enabled="false" level="WARNING" enabled_by_default="false" />
+    <inspection_tool class="PyPackageRequirementsInspection" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ignoredPackages">
+        <value>
+          <list size="1">
+            <item index="0" class="java.lang.String" itemvalue="sphinx_rtd_theme" />
+          </list>
+        </value>
+      </option>
+    </inspection_tool>
     <inspection_tool class="SpellCheckingInspection" enabled="true" level="TYPO" enabled_by_default="true">
       <option name="processCode" value="false" />
       <option name="processLiterals" value="true" />

.idea/libraries/github_hypfvieh_dbus_java.xml

@@ -1,25 +0,0 @@
-<component name="libraryTable">
-  <library name="github.hypfvieh.dbus.java" type="repository">
-    <properties maven-id="com.github.hypfvieh:dbus-java:3.3.2" />
-    <CLASSES>
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/hypfvieh/dbus-java/3.3.2/dbus-java-3.3.2.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jnr-unixsocket/0.38.17/jnr-unixsocket-0.38.17.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jnr-ffi/2.2.11/jnr-ffi-2.2.11.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jffi/1.3.9/jffi-1.3.9.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jffi/1.3.9/jffi-1.3.9-native.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/ow2/asm/asm/9.2/asm-9.2.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/ow2/asm/asm-commons/9.2/asm-commons-9.2.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/ow2/asm/asm-analysis/9.2/asm-analysis-9.2.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/ow2/asm/asm-tree/9.2/asm-tree-9.2.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/ow2/asm/asm-util/9.2/asm-util-9.2.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jnr-a64asm/1.0.0/jnr-a64asm-1.0.0.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jnr-x86asm/1.0.2/jnr-x86asm-1.0.2.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jnr-constants/0.10.3/jnr-constants-0.10.3.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jnr-enxio/0.32.13/jnr-enxio-0.32.13.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/github/jnr/jnr-posix/3.1.15/jnr-posix-3.1.15.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/slf4j/slf4j-api/1.7.36/slf4j-api-1.7.36.jar!/" />
-    </CLASSES>
-    <JAVADOC />
-    <SOURCES />
-  </library>
-</component>

.idea/libraries/glassfish_javax_json.xml

@@ -1,10 +0,0 @@
-<component name="libraryTable">
-  <library name="glassfish.javax.json" type="repository">
-    <properties include-transitive-deps="false" maven-id="org.glassfish:javax.json:1.1.4" />
-    <CLASSES>
-      <root url="jar://$MAVEN_REPOSITORY$/org/glassfish/javax.json/1.1.4/javax.json-1.1.4.jar!/" />
-    </CLASSES>
-    <JAVADOC />
-    <SOURCES />
-  </library>
-</component>

.idea/libraries/michael_bull_kotlin_result_jvm.xml

@@ -1,13 +1,10 @@
 <component name="libraryTable">
   <library name="michael.bull.kotlin.result.jvm" type="repository">
-    <properties maven-id="com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.18" />
+    <properties maven-id="com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.20" />
     <CLASSES>
-      <root url="jar://$MAVEN_REPOSITORY$/com/michael-bull/kotlin-result/kotlin-result-jvm/1.1.18/kotlin-result-jvm-1.1.18.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/jetbrains/kotlin/kotlin-stdlib-jdk8/1.8.10/kotlin-stdlib-jdk8-1.8.10.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/jetbrains/kotlin/kotlin-stdlib/1.8.10/kotlin-stdlib-1.8.10.jar!/" />
+      <root url="jar://$MAVEN_REPOSITORY$/com/michael-bull/kotlin-result/kotlin-result-jvm/1.1.20/kotlin-result-jvm-1.1.20.jar!/" />
+      <root url="jar://$MAVEN_REPOSITORY$/org/jetbrains/kotlin/kotlin-stdlib/1.9.22/kotlin-stdlib-1.9.22.jar!/" />
       <root url="jar://$MAVEN_REPOSITORY$/org/jetbrains/annotations/13.0/annotations-13.0.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/jetbrains/kotlin/kotlin-stdlib-jdk7/1.8.10/kotlin-stdlib-jdk7-1.8.10.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/jetbrains/kotlin/kotlin-stdlib-common/1.8.10/kotlin-stdlib-common-1.8.10.jar!/" />
     </CLASSES>
     <JAVADOC />
     <SOURCES />

.idea/libraries/slf4j_simple.xml

@@ -1,11 +0,0 @@
-<component name="libraryTable">
-  <library name="slf4j.simple" type="repository">
-    <properties maven-id="org.slf4j:slf4j-simple:2.0.11" />
-    <CLASSES>
-      <root url="jar://$MAVEN_REPOSITORY$/org/slf4j/slf4j-simple/2.0.11/slf4j-simple-2.0.11.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/slf4j/slf4j-api/2.0.11/slf4j-api-2.0.11.jar!/" />
-    </CLASSES>
-    <JAVADOC />
-    <SOURCES />
-  </library>
-</component>

.idea/libraries/takes.xml

@@ -1,18 +0,0 @@
-<component name="libraryTable">
-  <library name="takes" type="repository">
-    <properties maven-id="org.takes:takes:1.24.4" />
-    <CLASSES>
-      <root url="jar://$MAVEN_REPOSITORY$/org/takes/takes/1.24.4/takes-1.24.4.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/cactoos/cactoos/0.54.0/cactoos-0.54.0.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/javax/xml/bind/jaxb-api/2.4.0-b180830.0359/jaxb-api-2.4.0-b180830.0359.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/javax/activation/javax.activation-api/1.2.0/javax.activation-api-1.2.0.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/sun/xml/bind/jaxb-core/4.0.0/jaxb-core-4.0.0.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/jakarta/xml/bind/jakarta.xml.bind-api/4.0.0/jakarta.xml.bind-api-4.0.0.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/jakarta/activation/jakarta.activation-api/2.1.0/jakarta.activation-api-2.1.0.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/org/eclipse/angus/angus-activation/1.0.0/angus-activation-1.0.0.jar!/" />
-      <root url="jar://$MAVEN_REPOSITORY$/com/sun/xml/bind/jaxb-impl/4.0.0/jaxb-impl-4.0.0.jar!/" />
-    </CLASSES>
-    <JAVADOC />
-    <SOURCES />
-  </library>
-</component>

.idea/modules.xml

@@ -10,10 +10,8 @@
       <module fileurl="file://$PROJECT_DIR$/codeOptimizers/codeOptimizers.iml" filepath="$PROJECT_DIR$/codeOptimizers/codeOptimizers.iml" />
       <module fileurl="file://$PROJECT_DIR$/compiler/compiler.iml" filepath="$PROJECT_DIR$/compiler/compiler.iml" />
       <module fileurl="file://$PROJECT_DIR$/compilerAst/compilerAst.iml" filepath="$PROJECT_DIR$/compilerAst/compilerAst.iml" />
-      <module fileurl="file://$PROJECT_DIR$/dbusCompilerService/dbusCompilerService.iml" filepath="$PROJECT_DIR$/dbusCompilerService/dbusCompilerService.iml" />
       <module fileurl="file://$PROJECT_DIR$/docs/docs.iml" filepath="$PROJECT_DIR$/docs/docs.iml" />
       <module fileurl="file://$PROJECT_DIR$/examples/examples.iml" filepath="$PROJECT_DIR$/examples/examples.iml" />
-      <module fileurl="file://$PROJECT_DIR$/httpCompilerService/httpCompilerService.iml" filepath="$PROJECT_DIR$/httpCompilerService/httpCompilerService.iml" />
       <module fileurl="file://$PROJECT_DIR$/intermediate/intermediate.iml" filepath="$PROJECT_DIR$/intermediate/intermediate.iml" />
       <module fileurl="file://$PROJECT_DIR$/parser/parser.iml" filepath="$PROJECT_DIR$/parser/parser.iml" />
       <module fileurl="file://$PROJECT_DIR$/virtualmachine/virtualmachine.iml" filepath="$PROJECT_DIR$/virtualmachine/virtualmachine.iml" />

.readthedocs.yaml

@@ -9,17 +9,19 @@ version: 2
 build:
   os: ubuntu-22.04
   tools:
-    python: "3.11"
-
-# Build documentation in the docs/ directory with Sphinx
-sphinx:
-  configuration: docs/source/conf.py
-
-# If using Sphinx, optionally build your docs in additional formats such as PDF
-formats:
-  - pdf
+    python: "3.12"
 
 # Optionally declare the Python requirements required to build your docs
 python:
   install:
     - requirements: docs/requirements.txt
+
+# Build documentation in the docs/ directory with Sphinx
+sphinx:
+  configuration: docs/source/conf.py
+  fail_on_warning: true
+
+# If using Sphinx, optionally build your docs in additional formats such as PDF
+formats:
+  - pdf
+  - epub

Makefile

@@ -0,0 +1,12 @@
+# super simple Makefile to lauch the main gradle targets to build and/or test the prog8 compiler
+
+.PHONY: all test
+
+all:
+	gradle installdist installshadowdist
+	@echo "compiler launch script can be found here: compiler/build/install/compiler-shadow/bin/p8compile"
+
+test:
+	gradle build
+	@echo "compiler launch script can be found here: compiler/build/install/compiler-shadow/bin/p8compile"
+

README.md

@@ -29,6 +29,8 @@ How to get it/build it
 - Or, if you want/need a bleeding edge development version, you can:
   - download a build artifact zipfile from a recent [github action build](https://github.com/irmen/prog8/actions).
   - you can also compile it yourself from source. [Instructions here](https://prog8.readthedocs.io/en/latest/compiling.html).
+- Alternatively, you can also install the compiler as a package on some linux distros:
+  - Arch (via AUR): [`prog8`](https://aur.archlinux.org/packages/prog8)
 
 Community
 ---------
@@ -49,8 +51,8 @@ GNU GPL 3.0 (see file LICENSE), with exception for generated code:
 What does Prog8 provide?
 ------------------------
 
-- reduction of source code length over raw assembly
-- fast execution speed due to compilation to native assembly code. It's possible to write certain raster interrupt 'demoscene' effects purely in Prog8.
+- all advantages of a higher level language over having to write assembly code manually
+- programs run very fast because compilation to native machine code. It's possible to write games purely in Prog8, and even certain raster interrupt 'demoscene' effects.
 - modularity, symbol scoping, subroutines
 - various data types other than just bytes (16-bit words, floats, strings)
 - floating point math is supported if the target system provides floating point library routines (C64 and Cx16 both do)
@@ -58,6 +60,7 @@ What does Prog8 provide?
 - automatic static variable allocations, automatic string and array variables and string sharing
 - subroutines with input parameters and result values
 - high-level program optimizations
+- no need for forward declarations
 - small program boilerplate/compilersupport overhead
 - programs can be run multiple times without reloading because of automatic variable (re)initializations.
 - conditional branches

codeCore/build.gradle

@@ -26,7 +26,7 @@ compileTestKotlin {
 dependencies {
     // should have no dependencies to other modules
     implementation "org.jetbrains.kotlin:kotlin-stdlib-jdk8"
-    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.18"
+    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.20"
 }
 
 sourceSets {

codeCore/src/prog8/code/SymbolTable.kt

@@ -195,7 +195,7 @@ class StStaticVariable(name: String,
                 require(onetimeInitializationArrayValue.isEmpty() ||onetimeInitializationArrayValue.size==length)
         }
         if(onetimeInitializationNumericValue!=null) {
-            require(dt in NumericDatatypes)
+            require(dt in NumericDatatypes || dt==DataType.BOOL)
         }
         if(onetimeInitializationArrayValue!=null) {
             require(dt in ArrayDatatypes)
@@ -221,6 +221,7 @@ class StMemVar(name: String,
     StNode(name, StNodeType.MEMVAR, astNode) {
 
     init{
+        require(dt!=DataType.BOOL && dt!=DataType.ARRAY_BOOL)
         if(dt in ArrayDatatypes || dt == DataType.STR)
             require(length!=null) { "memory mapped array or string must have known length" }
     }
@@ -250,7 +251,7 @@ class StRomSub(name: String,
 
 class StSubroutineParameter(val name: String, val type: DataType)
 class StRomSubParameter(val register: RegisterOrStatusflag, val type: DataType)
-class StArrayElement(val number: Double?, val addressOfSymbol: String?)
+class StArrayElement(val number: Double?, val addressOfSymbol: String?, val boolean: Boolean?)
 
 typealias StString = Pair<String, Encoding>
 typealias StArray = List<StArrayElement>

codeCore/src/prog8/code/SymbolTableMaker.kt

@@ -128,10 +128,11 @@ class SymbolTableMaker(private val program: PtProgram, private val options: Comp
                 is PtAddressOf -> {
                     if(it.isFromArrayElement)
                         TODO("address-of array element $it in initial array value")
-                    StArrayElement(null, it.identifier.name)
+                    StArrayElement(null, it.identifier.name, null)
                 }
-                is PtIdentifier -> StArrayElement(null, it.name)
-                is PtNumber -> StArrayElement(it.number, null)
+                is PtIdentifier -> StArrayElement(null, it.name, null)
+                is PtNumber -> StArrayElement(it.number, null, null)
+                is PtBool -> StArrayElement(null, null, it.value)
                 else -> throw AssemblyError("invalid array element $it")
             }
         }

codeCore/src/prog8/code/ast/AstBase.kt

@@ -67,6 +67,7 @@ class PtProgram(
         children.asSequence().filterIsInstance<PtBlock>()
 
     fun entrypoint(): PtSub? =
+        // returns the main.start subroutine if it exists
         allBlocks().firstOrNull { it.name == "main" || it.name=="p8b_main" }
             ?.children
             ?.firstOrNull { it is PtSub && (it.name == "start" || it.name=="main.start" || it.name=="p8s_start" || it.name=="p8b_main.p8s_start") } as PtSub?

codeCore/src/prog8/code/ast/AstExpressions.kt

@@ -9,8 +9,6 @@ import kotlin.math.truncate
 sealed class PtExpression(val type: DataType, position: Position) : PtNode(position) {
 
     init {
-        if(type==DataType.BOOL)
-            throw IllegalArgumentException("bool should have become ubyte @$position")
         if(type==DataType.UNDEFINED) {
             @Suppress("LeakingThis")
             when(this) {
@@ -49,6 +47,7 @@ sealed class PtExpression(val type: DataType, position: Position) : PtNode(posit
             is PtMachineRegister -> other is PtMachineRegister && other.type==type && other.register==register
             is PtMemoryByte -> other is PtMemoryByte && other.address isSameAs address
             is PtNumber -> other is PtNumber && other.type==type && other.number==number
+            is PtBool -> other is PtBool && other.value==value
             is PtPrefix -> other is PtPrefix && other.type==type && other.operator==operator && other.value isSameAs value
             is PtRange -> other is PtRange && other.type==type && other.from==from && other.to==to && other.step==step
             is PtTypeCast -> other is PtTypeCast && other.type==type && other.value isSameAs value
@@ -71,7 +70,7 @@ sealed class PtExpression(val type: DataType, position: Position) : PtNode(posit
         }
     }
 
-    fun asConstInteger(): Int? = (this as? PtNumber)?.number?.toInt()
+    fun asConstInteger(): Int? = (this as? PtNumber)?.number?.toInt() ?: (this as? PtBool)?.asInt()
 
     fun isSimple(): Boolean {
         return when(this) {
@@ -90,6 +89,7 @@ sealed class PtExpression(val type: DataType, position: Position) : PtNode(posit
             is PtIdentifier -> true
             is PtMachineRegister -> true
             is PtMemoryByte -> address is PtNumber || address is PtIdentifier
+            is PtBool -> true
             is PtNumber -> true
             is PtPrefix -> value.isSimple()
             is PtRange -> true
@@ -116,6 +116,7 @@ sealed class PtExpression(val type: DataType, position: Position) : PtNode(posit
             is PtMachineRegister -> return withClonedChildrenFrom(this, PtMachineRegister(register, type, position))
             is PtMemoryByte -> return withClonedChildrenFrom(this, PtMemoryByte(position))
             is PtNumber -> return withClonedChildrenFrom(this, PtNumber(type, number, position))
+            is PtBool -> return withClonedChildrenFrom(this, PtBool(value, position))
             is PtPrefix -> return withClonedChildrenFrom(this, PtPrefix(operator, type, position))
             is PtRange -> return withClonedChildrenFrom(this, PtRange(type, position))
             is PtString -> return withClonedChildrenFrom(this, PtString(value, encoding, position))
@@ -149,7 +150,7 @@ class PtArrayIndexer(elementType: DataType, position: Position): PtExpression(el
         get() = variable.type in SplitWordArrayTypes
 
     init {
-        require(elementType in NumericDatatypes)
+        require(elementType in NumericDatatypesWithBoolean)
     }
 }
 
@@ -183,15 +184,21 @@ class PtBuiltinFunctionCall(val name: String,
 
 
 class PtBinaryExpression(val operator: String, type: DataType, position: Position): PtExpression(type, position) {
-    // note: "and", "or", "xor" do not occur anymore as operators. They've been replaced int the ast by their bitwise versions &, |, ^.
     val left: PtExpression
         get() = children[0] as PtExpression
     val right: PtExpression
         get() = children[1] as PtExpression
+
+    init {
+        if(operator in ComparisonOperators + LogicalOperators)
+            require(type==DataType.BOOL)
+        else
+            require(type!=DataType.BOOL) { "no bool allowed for this operator $operator"}
+    }
 }
 
 
-class PtContainmentCheck(position: Position): PtExpression(DataType.UBYTE, position) {
+class PtContainmentCheck(position: Position): PtExpression(DataType.BOOL, position) {
     val element: PtExpression
         get() = children[0] as PtExpression
     val iterable: PtIdentifier
@@ -228,6 +235,27 @@ class PtMemoryByte(position: Position) : PtExpression(DataType.UBYTE, position)
 }
 
 
+class PtBool(val value: Boolean, position: Position) : PtExpression(DataType.BOOL, position) {
+
+    companion object {
+        fun fromNumber(number: Number, position: Position): PtBool =
+            PtBool(if(number==0.0) false else true, position)
+    }
+
+    override fun hashCode(): Int = Objects.hash(type, value)
+
+    override fun equals(other: Any?): Boolean {
+        if(other==null || other !is PtBool)
+            return false
+        return value==other.value
+    }
+
+    override fun toString() = "PtBool:$value"
+
+    fun asInt(): Int = if(value) 1 else 0
+}
+
+
 class PtNumber(type: DataType, val number: Double, position: Position) : PtExpression(type, position) {
 
     companion object {
@@ -237,7 +265,7 @@ class PtNumber(type: DataType, val number: Double, position: Position) : PtExpre
 
     init {
         if(type==DataType.BOOL)
-            throw IllegalArgumentException("bool should have become ubyte @$position")
+            throw IllegalArgumentException("use PtBool instead")
         if(type!=DataType.FLOAT) {
             val trunc = truncate(number)
             if (trunc != number)
@@ -267,8 +295,7 @@ class PtPrefix(val operator: String, type: DataType, position: Position): PtExpr
         get() = children.single() as PtExpression
 
     init {
-        // note: the "not" operator may no longer occur in the ast; not x should have been replaced with x==0
-        require(operator in setOf("+", "-", "~")) { "invalid prefix operator: $operator" }
+        require(operator in setOf("+", "-", "~", "not")) { "invalid prefix operator: $operator" }
     }
 }
 
@@ -330,5 +357,5 @@ class PtTypeCast(type: DataType, position: Position) : PtExpression(type, positi
 class PtMachineRegister(val register: Int, type: DataType, position: Position) : PtExpression(type, position)
 
 
-fun constValue(expr: PtExpression): Double? = if(expr is PtNumber) expr.number else null
-fun constIntValue(expr: PtExpression): Int? = if(expr is PtNumber) expr.number.toInt() else null
+fun constValue(expr: PtExpression): Double? = if(expr is PtNumber) expr.number else if(expr is PtBool) expr.asInt().toDouble() else null
+fun constIntValue(expr: PtExpression): Int? = if(expr is PtNumber) expr.number.toInt() else if(expr is PtBool) expr.asInt() else null

codeCore/src/prog8/code/ast/AstPrinter.kt

@@ -15,7 +15,12 @@ fun printAst(root: PtNode, skipLibraries: Boolean, output: (text: String) -> Uni
             is PtAugmentedAssign -> "<inplace-assign> ${node.operator}"
             is PtBreakpoint -> "%breakpoint"
             is PtConditionalBranch -> "if_${node.condition.name.lowercase()}"
-            is PtAddressOf -> "&"
+            is PtAddressOf -> {
+                if(node.isFromArrayElement)
+                    "& array-element"
+                else
+                    "&"
+            }
             is PtArray -> "array len=${node.children.size} ${type(node.type)}"
             is PtArrayIndexer -> "<arrayindexer> ${type(node.type)} ${if(node.splitWords) "[splitwords]" else ""}"
             is PtBinaryExpression -> "<expr> ${node.operator} ${type(node.type)}"
@@ -35,6 +40,7 @@ fun printAst(root: PtNode, skipLibraries: Boolean, output: (text: String) -> Uni
                 val numstr = if(node.type == DataType.FLOAT) node.number.toString() else node.number.toHex()
                 "$numstr ${type(node.type)}"
             }
+            is PtBool -> node.value.toString()
             is PtPrefix -> node.operator
             is PtRange -> "<range>"
             is PtString -> "\"${node.value.escape()}\""
@@ -84,7 +90,11 @@ fun printAst(root: PtNode, skipLibraries: Boolean, output: (text: String) -> Uni
                 "\nblock '${node.name}' $addr $align"
             }
             is PtConstant -> {
-                val value = if(node.type in IntegerDatatypes) node.value.toInt().toString() else node.value.toString()
+                val value = when(node.type) {
+                    DataType.BOOL -> if(node.value==0.0) "false" else "true"
+                    in IntegerDatatypes -> node.value.toInt().toString()
+                    else -> node.value.toString()
+                }
                 "const ${node.type.name.lowercase()} ${node.name} = $value"
             }
             is PtLabel -> "${node.name}:"
@@ -121,9 +131,8 @@ fun printAst(root: PtNode, skipLibraries: Boolean, output: (text: String) -> Uni
                 else
                     str
             }
-            is PtNodeGroup -> "<group>"
+            is PtNodeGroup -> if(node.children.isNotEmpty()) "<group>" else ""
             is PtNop -> "nop"
-            is PtPostIncrDecr -> "<post> ${node.operator}"
             is PtProgram -> "PROGRAM ${node.name}"
             is PtRepeatLoop -> "repeat"
             is PtReturn -> "return"

codeCore/src/prog8/code/ast/AstStatements.kt

@@ -26,10 +26,10 @@ class PtSub(
 ) : PtNamedNode(name, position), IPtSubroutine, IPtStatementContainer {
     init {
         // params and return value should not be str
-        if(parameters.any{ it.type !in NumericDatatypes })
-            throw AssemblyError("non-numeric parameter")
-        if(returntype!=null && returntype !in NumericDatatypes)
-            throw AssemblyError("non-numeric returntype $returntype")
+        if(parameters.any{ it.type !in NumericDatatypes && it.type!=DataType.BOOL })
+            throw AssemblyError("non-numeric/non-bool parameter")
+        if(returntype!=null && returntype !in NumericDatatypes && returntype!=DataType.BOOL)
+            throw AssemblyError("non-numeric/non-bool returntype $returntype")
         parameters.forEach { it.parent=this }
     }
 }
@@ -98,6 +98,8 @@ class PtIfElse(position: Position) : PtNode(position) {
         get() = children[1] as PtNodeGroup
     val elseScope: PtNodeGroup
         get() = children[2] as PtNodeGroup
+
+    fun hasElse(): Boolean = children.size==3 && elseScope.children.isNotEmpty()
 }
 
 
@@ -110,12 +112,6 @@ class PtJump(val identifier: PtIdentifier?,     // note: even ad-hoc labels are
 }
 
 
-class PtPostIncrDecr(val operator: String, position: Position) : PtNode(position) {
-    val target: PtAssignTarget
-        get() = children.single() as PtAssignTarget
-}
-
-
 class PtRepeatLoop(position: Position) : PtNode(position) {
     val count: PtExpression
         get() = children[0] as PtExpression
@@ -152,7 +148,11 @@ class PtVariable(name: String, override val type: DataType, val zeropage: Zeropa
 class PtConstant(name: String, override val type: DataType, val value: Double, position: Position) : PtNamedNode(name, position), IPtVariable
 
 
-class PtMemMapped(name: String, override val type: DataType, val address: UInt, val arraySize: UInt?, position: Position) : PtNamedNode(name, position), IPtVariable
+class PtMemMapped(name: String, override val type: DataType, val address: UInt, val arraySize: UInt?, position: Position) : PtNamedNode(name, position), IPtVariable {
+    init {
+        require(type!=DataType.BOOL && type!=DataType.ARRAY_BOOL)
+    }
+}
 
 
 class PtWhen(position: Position) : PtNode(position) {

codeCore/src/prog8/code/core/BuiltinFunctions.kt

@@ -79,27 +79,28 @@ val BuiltinFunctions: Map<String, FSignature> = mapOf(
     "sort"      to FSignature(false, listOf(FParam("array", ArrayDatatypes)), null),
     "reverse"   to FSignature(false, listOf(FParam("array", ArrayDatatypes)), null),
     // cmp returns a status in the carry flag, but not a proper return value
-    "cmp"       to FSignature(false, listOf(FParam("value1", IntegerDatatypesNoBool), FParam("value2", NumericDatatypesNoBool)), null),
+    "cmp"       to FSignature(false, listOf(FParam("value1", IntegerDatatypes), FParam("value2", NumericDatatypes)), null),
     "prog8_lib_stringcompare"     to FSignature(true, listOf(FParam("str1", arrayOf(DataType.STR)), FParam("str2", arrayOf(DataType.STR))), DataType.BYTE),
+    "prog8_lib_arraycopy"         to FSignature(false, listOf(FParam("source", ArrayDatatypes), FParam("target", ArrayDatatypes)), null),
     "prog8_lib_square_byte"       to FSignature(true, listOf(FParam("value", arrayOf(DataType.BYTE, DataType.UBYTE))), DataType.UBYTE),
     "prog8_lib_square_word"       to FSignature(true, listOf(FParam("value", arrayOf(DataType.WORD, DataType.UWORD))), DataType.UWORD),
-    "abs"       to FSignature(true, listOf(FParam("value", NumericDatatypesNoBool)), null),
+    "abs"       to FSignature(true, listOf(FParam("value", NumericDatatypes)), null),
     "abs__byte"  to FSignature(true, listOf(FParam("value", arrayOf(DataType.BYTE))), DataType.BYTE),
     "abs__word"  to FSignature(true, listOf(FParam("value", arrayOf(DataType.WORD))), DataType.WORD),
     "abs__float" to FSignature(true, listOf(FParam("value", arrayOf(DataType.FLOAT))), DataType.FLOAT),
     "len"       to FSignature(true, listOf(FParam("values", IterableDatatypes)), DataType.UWORD),
     // normal functions follow:
     "sizeof"    to FSignature(true, listOf(FParam("object", DataType.entries.toTypedArray())), DataType.UBYTE),
-    "sgn"       to FSignature(true, listOf(FParam("value", NumericDatatypesNoBool)), DataType.BYTE),
-    "sqrt"        to FSignature(true, listOf(FParam("value", NumericDatatypesNoBool)), null),
+    "sgn"       to FSignature(true, listOf(FParam("value", NumericDatatypes)), DataType.BYTE),
+    "sqrt"        to FSignature(true, listOf(FParam("value", NumericDatatypes)), null),
     "sqrt__ubyte" to FSignature(true, listOf(FParam("value", arrayOf(DataType.UBYTE))), DataType.UBYTE),
     "sqrt__uword" to FSignature(true, listOf(FParam("value", arrayOf(DataType.UWORD))), DataType.UBYTE),
     "sqrt__float" to FSignature(true, listOf(FParam("value", arrayOf(DataType.FLOAT))), DataType.FLOAT),
     "divmod"           to FSignature(false, listOf(FParam("number", arrayOf(DataType.UBYTE, DataType.UWORD)), FParam("divident", arrayOf(DataType.UBYTE, DataType.UWORD)), FParam("division", arrayOf(DataType.UBYTE, DataType.UWORD)), FParam("remainder", arrayOf(DataType.UBYTE, DataType.UWORD))), null),
     "divmod__ubyte"    to FSignature(false, listOf(FParam("number", arrayOf(DataType.UBYTE)), FParam("divident", arrayOf(DataType.UBYTE)), FParam("division", arrayOf(DataType.UBYTE)), FParam("remainder", arrayOf(DataType.UBYTE))), null),
     "divmod__uword"    to FSignature(false, listOf(FParam("number", arrayOf(DataType.UWORD)), FParam("divident", arrayOf(DataType.UWORD)), FParam("division", arrayOf(DataType.UWORD)), FParam("remainder", arrayOf(DataType.UWORD))), null),
-    "any"       to FSignature(true, listOf(FParam("values", ArrayDatatypes)), DataType.UBYTE),
-    "all"       to FSignature(true, listOf(FParam("values", ArrayDatatypes)), DataType.UBYTE),
+    "any"       to FSignature(true, listOf(FParam("values", ArrayDatatypes)), DataType.BOOL),
+    "all"       to FSignature(true, listOf(FParam("values", ArrayDatatypes)), DataType.BOOL),
     "lsb"       to FSignature(true, listOf(FParam("value", arrayOf(DataType.UWORD, DataType.WORD))), DataType.UBYTE),
     "msb"       to FSignature(true, listOf(FParam("value", arrayOf(DataType.UWORD, DataType.WORD))), DataType.UBYTE),
     "mkword"    to FSignature(true, listOf(FParam("msb", arrayOf(DataType.UBYTE)), FParam("lsb", arrayOf(DataType.UBYTE))), DataType.UWORD),

codeCore/src/prog8/code/core/CompilationOptions.kt

@@ -19,12 +19,14 @@ class CompilationOptions(val output: OutputType,
                          var asmQuiet: Boolean = false,
                          var asmListfile: Boolean = false,
                          var includeSourcelines: Boolean = false,
+                         var dumpVariables: Boolean = false,
                          var experimentalCodegen: Boolean = false,
                          var varsHighBank: Int? = null,
                          var varsGolden: Boolean = false,
                          var slabsHighBank: Int? = null,
                          var slabsGolden: Boolean = false,
                          var splitWordArrays: Boolean = false,
+                         var strictBool: Boolean = true,
                          var breakpointCpuInstruction: String? = null,
                          var outputDir: Path = Path(""),
                          var symbolDefs: Map<String, String> = emptyMap()

codeCore/src/prog8/code/core/Conversions.kt

@@ -58,6 +58,7 @@ fun String.unescape(): String {
                 '\\' -> '\\'
                 'n' -> '\n'
                 'r' -> '\r'
+                't' -> '\t'
                 '"' -> '"'
                 '\'' -> '\''
                 'u' -> {

codeCore/src/prog8/code/core/Enumerations.kt

@@ -24,8 +24,8 @@ enum class DataType {
      */
     infix fun isAssignableTo(targetType: DataType) =
         when(this) {
-            BOOL -> targetType.oneOf(BOOL, BYTE, UBYTE, WORD, UWORD, LONG, FLOAT)
-            UBYTE -> targetType.oneOf(UBYTE, WORD, UWORD, LONG, FLOAT, BOOL)
+            BOOL -> targetType == BOOL
+            UBYTE -> targetType.oneOf(UBYTE, WORD, UWORD, LONG, FLOAT)
             BYTE -> targetType.oneOf(BYTE, WORD, LONG, FLOAT)
             UWORD -> targetType.oneOf(UWORD, LONG, FLOAT)
             WORD -> targetType.oneOf(WORD, LONG, FLOAT)
@@ -41,9 +41,9 @@ enum class DataType {
     infix fun largerThan(other: DataType) =
         when {
             this == other -> false
-            this in ByteDatatypes -> false
-            this in WordDatatypes -> other in ByteDatatypes
-            this == LONG -> other in ByteDatatypes+WordDatatypes
+            this in ByteDatatypesWithBoolean -> false
+            this in WordDatatypes -> other in ByteDatatypesWithBoolean
+            this == LONG -> other in ByteDatatypesWithBoolean+WordDatatypes
             this == STR && other == UWORD || this == UWORD && other == STR -> false
             else -> true
         }
@@ -51,7 +51,7 @@ enum class DataType {
     infix fun equalsSize(other: DataType) =
         when {
             this == other -> true
-            this in ByteDatatypes -> other in ByteDatatypes
+            this in ByteDatatypesWithBoolean -> other in ByteDatatypesWithBoolean
             this in WordDatatypes -> other in WordDatatypes
             this== STR && other== UWORD || this== UWORD && other== STR -> true
             else -> false
@@ -124,12 +124,13 @@ enum class BranchCondition {
 }
 
 
-val ByteDatatypes = arrayOf(DataType.UBYTE, DataType.BYTE, DataType.BOOL)
+val ByteDatatypes = arrayOf(DataType.UBYTE, DataType.BYTE)
+val ByteDatatypesWithBoolean = ByteDatatypes + DataType.BOOL
 val WordDatatypes = arrayOf(DataType.UWORD, DataType.WORD)
-val IntegerDatatypesNoBool = arrayOf(DataType.UBYTE, DataType.BYTE, DataType.UWORD, DataType.WORD, DataType.LONG)
-val IntegerDatatypes = IntegerDatatypesNoBool + DataType.BOOL
-val NumericDatatypesNoBool = arrayOf(DataType.UBYTE, DataType.BYTE, DataType.UWORD, DataType.WORD, DataType.LONG, DataType.FLOAT)
-val NumericDatatypes = NumericDatatypesNoBool + DataType.BOOL
+val IntegerDatatypes = arrayOf(DataType.UBYTE, DataType.BYTE, DataType.UWORD, DataType.WORD, DataType.LONG)
+val IntegerDatatypesWithBoolean = IntegerDatatypes + DataType.BOOL
+val NumericDatatypes = arrayOf(DataType.UBYTE, DataType.BYTE, DataType.UWORD, DataType.WORD, DataType.LONG, DataType.FLOAT)
+val NumericDatatypesWithBoolean = NumericDatatypes + DataType.BOOL
 val SignedDatatypes =  arrayOf(DataType.BYTE, DataType.WORD, DataType.LONG, DataType.FLOAT)
 val ArrayDatatypes = arrayOf(DataType.ARRAY_UB, DataType.ARRAY_B, DataType.ARRAY_UW, DataType.ARRAY_UW_SPLIT, DataType.ARRAY_W, DataType.ARRAY_W_SPLIT, DataType.ARRAY_F, DataType.ARRAY_BOOL)
 val StringlyDatatypes = arrayOf(DataType.STR, DataType.ARRAY_UB, DataType.ARRAY_B, DataType.UWORD)
@@ -141,7 +142,7 @@ val IterableDatatypes = arrayOf(
     DataType.ARRAY_UW_SPLIT, DataType.ARRAY_W_SPLIT,
     DataType.ARRAY_F, DataType.ARRAY_BOOL
 )
-val PassByValueDatatypes = NumericDatatypes
+val PassByValueDatatypes = NumericDatatypesWithBoolean
 val PassByReferenceDatatypes = IterableDatatypes
 val ArrayToElementTypes = mapOf(
     DataType.STR to DataType.UBYTE,

codeCore/src/prog8/code/core/IErrorReporter.kt

@@ -11,4 +11,6 @@ interface IErrorReporter {
         if(numErrors>0)
             throw ErrorsReportedException("There are $numErrors errors, $numWarnings warnings, and $numInfos infos.")
     }
+
+    fun noErrorForLine(position: Position): Boolean
 }

codeCore/src/prog8/code/core/MemoryRegions.kt

@@ -71,7 +71,7 @@ abstract class Zeropage(options: CompilationOptions): MemoryAllocator(options) {
 
         val size: Int =
                 when (datatype) {
-                    in IntegerDatatypes -> options.compTarget.memorySize(datatype)
+                    in IntegerDatatypesWithBoolean -> options.compTarget.memorySize(datatype)
                     DataType.STR, in ArrayDatatypes  -> {
                         val memsize = options.compTarget.memorySize(datatype, numElements!!)
                         if(position!=null)
@@ -119,7 +119,7 @@ abstract class Zeropage(options: CompilationOptions): MemoryAllocator(options) {
         free.removeAll(address until address+size.toUInt())
         if(name.isNotEmpty()) {
             allocatedVariables[name] = when(datatype) {
-                in NumericDatatypes -> VarAllocation(address, datatype, size)        // numerical variables in zeropage never have an initial value here because they are set in separate initializer assignments
+                in NumericDatatypes, DataType.BOOL -> VarAllocation(address, datatype, size)        // numerical variables in zeropage never have an initial value here because they are set in separate initializer assignments
                 DataType.STR -> VarAllocation(address, datatype, size)
                 in ArrayDatatypes -> VarAllocation(address, datatype, size)
                 else -> throw AssemblyError("invalid dt")
@@ -151,7 +151,7 @@ class GoldenRam(options: CompilationOptions, val region: UIntRange): MemoryAlloc
 
         val size: Int =
             when (datatype) {
-                in IntegerDatatypes -> options.compTarget.memorySize(datatype)
+                in IntegerDatatypesWithBoolean -> options.compTarget.memorySize(datatype)
                 DataType.STR, in ArrayDatatypes  -> {
                     options.compTarget.memorySize(datatype, numElements!!)
                 }

codeCore/src/prog8/code/core/Operators.kt

@@ -1,8 +1,8 @@
 package prog8.code.core
 
-val AssociativeOperators = setOf("+", "*", "&", "|", "^", "==", "!=", "and", "or", "xor")
+val AssociativeOperators = setOf("+", "*", "&", "|", "^", "==", "!=", "xor")       // note: and,or are no longer associative because of Shortcircuit/McCarthy evaluation
 val ComparisonOperators = setOf("==", "!=", "<", ">", "<=", ">=")
-val LogicalOperators = setOf("and", "or", "xor", "not")
+val LogicalOperators = setOf("and", "or", "xor", "not", "in")
 val BitwiseOperators = setOf("&", "|", "^", "~")
 val PrefixOperators = setOf("+", "-", "~", "not")
 

codeCore/src/prog8/code/target/AtariTarget.kt

@@ -15,7 +15,7 @@ class AtariTarget: ICompilationTarget, IStringEncoding by Encoder, IMemSizer {
 
     override fun memorySize(dt: DataType): Int {
         return when(dt) {
-            in ByteDatatypes -> 1
+            in ByteDatatypesWithBoolean -> 1
             in WordDatatypes, in PassByReferenceDatatypes -> 2
             DataType.FLOAT -> machine.FLOAT_MEM_SIZE
             else -> throw IllegalArgumentException("invalid datatype")
@@ -23,5 +23,9 @@ class AtariTarget: ICompilationTarget, IStringEncoding by Encoder, IMemSizer {
     }
 
     override fun memorySize(arrayDt: DataType, numElements: Int) =
-        memorySize(ArrayToElementTypes.getValue(arrayDt)) * numElements
+        if(arrayDt==DataType.UWORD)
+            numElements    // pointer to bytes.
+        else
+            memorySize(ArrayToElementTypes.getValue(arrayDt)) * numElements
+
 }

codeCore/src/prog8/code/target/VMTarget.kt

@@ -14,7 +14,7 @@ class VMTarget: ICompilationTarget, IStringEncoding by Encoder, IMemSizer {
 
     override fun memorySize(dt: DataType): Int {
         return when(dt) {
-            in ByteDatatypes -> 1
+            in ByteDatatypesWithBoolean -> 1
             in WordDatatypes, in PassByReferenceDatatypes -> 2
             DataType.FLOAT -> machine.FLOAT_MEM_SIZE
             else -> throw IllegalArgumentException("invalid datatype")
@@ -22,5 +22,8 @@ class VMTarget: ICompilationTarget, IStringEncoding by Encoder, IMemSizer {
     }
 
     override fun memorySize(arrayDt: DataType, numElements: Int) =
-        memorySize(ArrayToElementTypes.getValue(arrayDt)) * numElements
+        if(arrayDt==DataType.UWORD)
+            numElements    // pointer to bytes.
+        else
+            memorySize(ArrayToElementTypes.getValue(arrayDt)) * numElements
 }

codeCore/src/prog8/code/target/cbm/CbmMemorySizer.kt

@@ -6,7 +6,7 @@ import prog8.code.core.*
 internal object CbmMemorySizer: IMemSizer {
     override fun memorySize(dt: DataType): Int {
         return when(dt) {
-            in ByteDatatypes -> 1
+            in ByteDatatypesWithBoolean -> 1
             in WordDatatypes, in PassByReferenceDatatypes -> 2
             DataType.FLOAT -> Mflpt5.FLOAT_MEM_SIZE
             else -> throw IllegalArgumentException("invalid datatype")
@@ -14,5 +14,8 @@ internal object CbmMemorySizer: IMemSizer {
     }
 
     override fun memorySize(arrayDt: DataType, numElements: Int) =
-        memorySize(ArrayToElementTypes.getValue(arrayDt)) * numElements
+        if(arrayDt==DataType.UWORD)
+            numElements    // pointer to bytes.
+        else
+            memorySize(ArrayToElementTypes.getValue(arrayDt)) * numElements
 }

codeGenCpu6502/build.gradle

@@ -1,9 +1,7 @@
-
 plugins {
     id 'java'
     id 'application'
     id "org.jetbrains.kotlin.jvm"
-    id "io.kotest" version "0.3.9"
 }
 
 java {
@@ -28,9 +26,11 @@ dependencies {
     implementation project(':codeCore')
     implementation "org.jetbrains.kotlin:kotlin-stdlib-jdk8"
     // implementation "org.jetbrains.kotlin:kotlin-reflect"
-    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.18"
+    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.20"
 
     testImplementation 'io.kotest:kotest-runner-junit5-jvm:5.8.0'
+    testImplementation 'org.junit.jupiter:junit-jupiter:5.9.2'
+    testRuntimeOnly 'org.junit.platform:junit-platform-launcher'
 }
 
 sourceSets {

codeGenCpu6502/src/prog8/codegen/cpu6502/AsmOptimizer.kt

@@ -58,7 +58,7 @@ internal fun optimizeAssembly(lines: MutableList<String>, machine: IMachineDefin
         numberOfOptimizations++
     }
 
-    mods = optimizeSamePointerIndexing(linesByFourteen)
+    mods = optimizeSamePointerIndexingAndUselessBeq(linesByFourteen)
     if(mods.isNotEmpty()) {
         apply(mods, lines)
         linesByFourteen = getLinesBy(lines, 14)
@@ -75,23 +75,48 @@ private fun String.isStoreReg() = this.startsWith("sta") || this.startsWith("sty
 private fun String.isStoreRegOrZero() = this.isStoreReg() || this.startsWith("stz")
 private fun String.isLoadReg() = this.startsWith("lda") || this.startsWith("ldy") || this.startsWith("ldx")
 
-private class Modification(val lineIndex: Int, val remove: Boolean, val replacement: String?)
+private class Modification(val lineIndex: Int, val remove: Boolean, val replacement: String?, val removeLabel: Boolean=false)
 
 private fun apply(modifications: List<Modification>, lines: MutableList<String>) {
     for (modification in modifications.sortedBy { it.lineIndex }.reversed()) {
-        if(modification.remove)
-            lines.removeAt(modification.lineIndex)
+        if(modification.remove) {
+            if(modification.removeLabel)
+                lines.removeAt(modification.lineIndex)
+            else {
+                val line = lines[modification.lineIndex]
+                if (line.length < 2 || line[0] == ';' || line.trimStart()[0] == ';')
+                    lines.removeAt(modification.lineIndex)
+                else if (haslabel(line)) {
+                    val label = keeplabel(line)
+                    if (label.isNotEmpty())
+                        lines[modification.lineIndex] = label
+                    else
+                        lines.removeAt(modification.lineIndex)
+                } else lines.removeAt(modification.lineIndex)
+            }
+        }
         else
             lines[modification.lineIndex] = modification.replacement!!
     }
 }
 
+private fun haslabel(line: String): Boolean {
+    return line.length>1 && line[0]!=';' && (!line[0].isWhitespace() || ':' in line)
+}
+
+private fun keeplabel(line: String): String {
+    if(':' in line)
+        return line.substringBefore(':') + ':'
+    val splits = line.split('\t', ' ', limit=2)
+    return if(splits.size>1) splits[0] + ':' else ""
+}
+
 private fun getLinesBy(lines: MutableList<String>, windowSize: Int) =
 // all lines (that aren't empty or comments) in sliding windows of certain size
-        lines.withIndex().filter { it.value.isNotBlank() && !it.value.trimStart().startsWith(';') }.windowed(windowSize, partialWindows = false)
+        lines.asSequence().withIndex().filter { it.value.isNotBlank() && !it.value.trimStart().startsWith(';') }.windowed(windowSize, partialWindows = false)
 
 private fun optimizeSameAssignments(
-    linesByFourteen: List<List<IndexedValue<String>>>,
+    linesByFourteen: Sequence<List<IndexedValue<String>>>,
     machine: IMachineDefinition,
     symbolTable: SymbolTable
 ): List<Modification> {
@@ -281,7 +306,7 @@ private fun optimizeSameAssignments(
     return mods
 }
 
-private fun optimizeSamePointerIndexing(linesByFourteen: List<List<IndexedValue<String>>>): List<Modification> {
+private fun optimizeSamePointerIndexingAndUselessBeq(linesByFourteen: Sequence<List<IndexedValue<String>>>): List<Modification> {
 
     // Optimize same pointer indexing where for instance we load and store to the same ptr index in Y
     // if Y isn't modified in between we can omit the second LDY:
@@ -318,31 +343,57 @@ private fun optimizeSamePointerIndexing(linesByFourteen: List<List<IndexedValue<
                 mods.add(Modification(lines[4].index, true, null))
             }
         }
+
+
+        /*
+    beq  +
+    lda  #1
++
+[ optional:  label_xxxx_shortcut   line here]
+    beq  label_xxxx_shortcut   /  bne label_xxxx_shortcut
+or *_afterif labels.
+
+This gets generated after certain if conditions, and only the branch instruction is needed in these cases.
+         */
+
+        if(first=="beq  +" && second=="lda  #1" && third=="+") {
+            if((fourth.startsWith("beq  label_") || fourth.startsWith("bne  label_")) &&
+                (fourth.endsWith("_shortcut") || fourth.endsWith("_afterif") || fourth.endsWith("_shortcut:") || fourth.endsWith("_afterif:"))) {
+                mods.add(Modification(lines[0].index, true, null))
+                mods.add(Modification(lines[1].index, true, null))
+                mods.add(Modification(lines[2].index, true, null))
+            }
+            else if(fourth.startsWith("label_") && (fourth.endsWith("_shortcut") || fourth.endsWith("_shortcut:"))) {
+                if((fifth.startsWith("beq  label_") || fifth.startsWith("bne  label_")) &&
+                    (fifth.endsWith("_shortcut") || fifth.endsWith("_afterif") || fifth.endsWith("_shortcut:") || fifth.endsWith("_afterif:"))) {
+                    mods.add(Modification(lines[0].index, true, null))
+                    mods.add(Modification(lines[1].index, true, null))
+                    mods.add(Modification(lines[2].index, true, null))
+                }
+            }
+        }
     }
 
     return mods
 }
 
 private fun optimizeStoreLoadSame(
-    linesByFour: List<List<IndexedValue<String>>>,
+    linesByFour: Sequence<List<IndexedValue<String>>>,
     machine: IMachineDefinition,
     symbolTable: SymbolTable
 ): List<Modification> {
-    // sta X + lda X,  sty X + ldy X,   stx X + ldx X  -> the second instruction can OFTEN be eliminated
     val mods = mutableListOf<Modification>()
     for (lines in linesByFour) {
         val first = lines[1].value.trimStart()
         val second = lines[2].value.trimStart()
 
+        // sta X + lda X,  sty X + ldy X,   stx X + ldx X  -> the second instruction can OFTEN be eliminated
         if ((first.startsWith("sta ") && second.startsWith("lda ")) ||
                 (first.startsWith("stx ") && second.startsWith("ldx ")) ||
                 (first.startsWith("sty ") && second.startsWith("ldy ")) ||
                 (first.startsWith("lda ") && second.startsWith("lda ")) ||
                 (first.startsWith("ldy ") && second.startsWith("ldy ")) ||
-                (first.startsWith("ldx ") && second.startsWith("ldx ")) ||
-                (first.startsWith("sta ") && second.startsWith("lda ")) ||
-                (first.startsWith("sty ") && second.startsWith("ldy ")) ||
-                (first.startsWith("stx ") && second.startsWith("ldx "))
+                (first.startsWith("ldx ") && second.startsWith("ldx "))
         ) {
             val third = lines[3].value.trimStart()
             val attemptRemove =
@@ -385,6 +436,18 @@ private fun optimizeStoreLoadSame(
             mods.add(Modification(lines[1].index, true, null))
             mods.add(Modification(lines[2].index, false, "  tya"))
         }
+
+
+        // lda X + sta X,  ldy X + sty X,   ldx X + stx X  -> the second instruction can be eliminated
+        if ((first.startsWith("lda ") && second.startsWith("sta ")) ||
+            (first.startsWith("ldx ") && second.startsWith("stx ")) ||
+            (first.startsWith("ldy ") && second.startsWith("sty "))
+        ) {
+            val firstLoc = first.substring(4).trimStart()
+            val secondLoc = second.substring(4).trimStart()
+            if (firstLoc == secondLoc)
+                mods.add(Modification(lines[2].index, true, null))
+        }
     }
     return mods
 }
@@ -414,7 +477,7 @@ private fun getAddressArg(line: String, symbolTable: SymbolTable): UInt? {
     }
 }
 
-private fun optimizeIncDec(linesByFour: List<List<IndexedValue<String>>>): List<Modification> {
+private fun optimizeIncDec(linesByFour: Sequence<List<IndexedValue<String>>>): List<Modification> {
     // sometimes, iny+dey / inx+dex / dey+iny / dex+inx sequences are generated, these can be eliminated.
     val mods = mutableListOf<Modification>()
     for (lines in linesByFour) {
@@ -431,14 +494,17 @@ private fun optimizeIncDec(linesByFour: List<List<IndexedValue<String>>>): List<
             mods.add(Modification(lines[0].index, true, null))
             mods.add(Modification(lines[1].index, true, null))
         }
+
     }
     return mods
 }
 
-private fun optimizeJsrRtsAndOtherCombinations(linesByFour: List<List<IndexedValue<String>>>): List<Modification> {
+private fun optimizeJsrRtsAndOtherCombinations(linesByFour: Sequence<List<IndexedValue<String>>>): List<Modification> {
     // jsr Sub + rts -> jmp Sub
     // rts + jmp -> remove jmp
     // rts + bxx -> remove bxx
+    // lda  + cmp #0 -> remove cmp,  same for cpy and cpx.
+    // and some other optimizations.
 
     val mods = mutableListOf<Modification>()
     for (lines in linesByFour) {
@@ -470,11 +536,54 @@ private fun optimizeJsrRtsAndOtherCombinations(linesByFour: List<List<IndexedVal
             else if (" bvc" in second || "\tbvc" in second)
                 mods += Modification(lines[1].index, true, null)
         }
+
+        if (!haslabel(second)) {
+            if ((" lda" in first || "\tlda" in first) && (" cmp  #0" in second || "\tcmp  #0" in second) ||
+                (" ldx" in first || "\tldx" in first) && (" cpx  #0" in second || "\tcpx  #0" in second) ||
+                (" ldy" in first || "\tldy" in first) && (" cpy  #0" in second || "\tcpy  #0" in second)
+            ) {
+                mods.add(Modification(lines[1].index, true, null))
+            }
+            else if(" cmp  #0" in second || "\tcmp  #0" in second) {
+                // there are many instructions that modify A and set the bits...
+                for(instr in arrayOf("lda", "ora", "and", "eor", "adc", "sbc", "asl", "cmp", "inc  a", "lsr", "pla", "rol", "ror", "txa", "tya")) {
+                    if(" $instr" in first || "\t$instr" in first) {
+                        mods.add(Modification(lines[1].index, true, null))
+                    }
+                }
+            }
+        }
+
+        /*
+    LDA NUM1
+    CMP NUM2
+    BCC LABEL
+    BEQ LABEL
+
+(or something similar) which branches to LABEL when NUM1 <= NUM2. (In this case NUM1 and NUM2 are unsigned numbers.) However, consider the following sequence:
+
+    LDA NUM2
+    CMP NUM1
+    BCS LABEL
+         */
+        val tfirst = first.trimStart()
+        val tsecond = second.trimStart()
+        val tthird = lines[2].value.trimStart()
+        val tfourth = lines[3].value.trimStart()
+        if(tfirst.startsWith("lda") && tsecond.startsWith("cmp") && tthird.startsWith("bcc") && tfourth.startsWith("beq")) {
+            val label = tthird.substring(4)
+            if(label==tfourth.substring(4)) {
+                mods += Modification(lines[0].index, false, "  lda  ${tsecond.substring(4)}")
+                mods += Modification(lines[1].index, false, "  cmp  ${tfirst.substring(4)}")
+                mods += Modification(lines[2].index, false, "  bcs  $label")
+                mods += Modification(lines[3].index, true, null)
+            }
+        }
     }
     return mods
 }
 
-private fun optimizeUselessPushPopStack(linesByFour: List<List<IndexedValue<String>>>): List<Modification> {
+private fun optimizeUselessPushPopStack(linesByFour: Sequence<List<IndexedValue<String>>>): List<Modification> {
     val mods = mutableListOf<Modification>()
 
     fun optimize(register: Char, lines: List<IndexedValue<String>>) {
@@ -511,7 +620,7 @@ private fun optimizeUselessPushPopStack(linesByFour: List<List<IndexedValue<Stri
     return mods
 }
 
-private fun optimizeUnneededTempvarInAdd(linesByFour: List<List<IndexedValue<String>>>): List<Modification> {
+private fun optimizeUnneededTempvarInAdd(linesByFour: Sequence<List<IndexedValue<String>>>): List<Modification> {
     // sequence:  sta  P8ZP_SCRATCH_XX  / lda  something / clc / adc  P8ZP_SCRATCH_XX
     // this can be performed without the scratch variable:  clc  /  adc  something
     val mods = mutableListOf<Modification>()
@@ -523,8 +632,8 @@ private fun optimizeUnneededTempvarInAdd(linesByFour: List<List<IndexedValue<Str
         val fourth = lines[3].value.trimStart()
         if(first.startsWith("sta  P8ZP_SCRATCH_") && second.startsWith("lda") && third.startsWith("clc") && fourth.startsWith("adc  P8ZP_SCRATCH_") ) {
             if(fourth.substring(4)==first.substring(4)) {
-                mods.add(Modification(lines[0].index, false, "    clc"))
-                mods.add(Modification(lines[1].index, false, "    adc  ${second.substring(3).trimStart()}"))
+                mods.add(Modification(lines[0].index, false, "  clc"))
+                mods.add(Modification(lines[1].index, false, "  adc  ${second.substring(3).trimStart()}"))
                 mods.add(Modification(lines[2].index, true, null))
                 mods.add(Modification(lines[3].index, true, null))
             }

codeGenCpu6502/src/prog8/codegen/cpu6502/BuiltinFunctionsAsmGen.kt

@@ -71,12 +71,102 @@ internal class BuiltinFunctionsAsmGen(private val program: PtProgram,
             "prog8_lib_stringcompare" -> funcStringCompare(fcall, resultRegister)
             "prog8_lib_square_byte" -> funcSquare(fcall, DataType.UBYTE, resultRegister)
             "prog8_lib_square_word" -> funcSquare(fcall, DataType.UWORD, resultRegister)
+            "prog8_lib_arraycopy" -> funcArrayCopy(fcall)
             else -> throw AssemblyError("missing asmgen for builtin func ${fcall.name}")
         }
 
         return BuiltinFunctions.getValue(fcall.name).returnType
     }
 
+    private fun funcArrayCopy(fcall: PtBuiltinFunctionCall) {
+        val source = fcall.args[0] as PtIdentifier
+        val target = fcall.args[1] as PtIdentifier
+
+        val sourceSymbol = asmgen.symbolTable.lookup(source.name)
+        val numElements = when(sourceSymbol) {
+            is StStaticVariable -> sourceSymbol.length!!
+            is StMemVar -> sourceSymbol.length!!
+            else -> 0
+        }
+        val sourceAsm = asmgen.asmVariableName(source)
+        val targetAsm = asmgen.asmVariableName(target)
+
+        if(source.type in SplitWordArrayTypes && target.type in SplitWordArrayTypes) {
+            // split -> split words (copy lsb and msb arrays separately)
+            asmgen.out("""
+                lda  #<${sourceAsm}_lsb
+                ldy  #>${sourceAsm}_lsb
+                sta  P8ZP_SCRATCH_W1
+                sty  P8ZP_SCRATCH_W1+1
+                lda  #<${targetAsm}_lsb
+                ldy  #>${targetAsm}_lsb
+                sta  P8ZP_SCRATCH_W2
+                sty  P8ZP_SCRATCH_W2+1
+                ldy  #${numElements and 255}
+                jsr  prog8_lib.memcopy_small
+                lda  #<${sourceAsm}_msb
+                ldy  #>${sourceAsm}_msb
+                sta  P8ZP_SCRATCH_W1
+                sty  P8ZP_SCRATCH_W1+1
+                lda  #<${targetAsm}_msb
+                ldy  #>${targetAsm}_msb
+                sta  P8ZP_SCRATCH_W2
+                sty  P8ZP_SCRATCH_W2+1
+                ldy  #${numElements and 255}
+                jsr  prog8_lib.memcopy_small""")
+        }
+        else if(source.type in SplitWordArrayTypes) {
+            // split word array to normal word array (copy lsb and msb arrays separately)
+            require(target.type==DataType.ARRAY_UW || target.type==DataType.ARRAY_W)
+            asmgen.out("""
+                lda  #<${sourceAsm}_lsb
+                ldy  #>${sourceAsm}_lsb
+                sta  P8ZP_SCRATCH_W1
+                sty  P8ZP_SCRATCH_W1+1
+                lda  #<${sourceAsm}_msb
+                ldy  #>${sourceAsm}_msb
+                sta  P8ZP_SCRATCH_W2
+                sty  P8ZP_SCRATCH_W2+1
+                lda  #<${targetAsm}
+                ldy  #>${targetAsm}
+                ldx  #${numElements and 255}
+                jsr  prog8_lib.arraycopy_split_to_normal_words""")
+        }
+        else if(target.type in SplitWordArrayTypes) {
+            // normal word array to split array
+            require(source.type==DataType.ARRAY_UW || source.type==DataType.ARRAY_W)
+            asmgen.out("""
+                lda  #<${targetAsm}_lsb
+                ldy  #>${targetAsm}_lsb
+                sta  P8ZP_SCRATCH_W1
+                sty  P8ZP_SCRATCH_W1+1
+                lda  #<${targetAsm}_msb
+                ldy  #>${targetAsm}_msb
+                sta  P8ZP_SCRATCH_W2
+                sty  P8ZP_SCRATCH_W2+1
+                lda  #<${sourceAsm}
+                ldy  #>${sourceAsm}
+                ldx  #${numElements and 255}
+                jsr  prog8_lib.arraycopy_normal_to_split_words""")
+        }
+        else {
+            // normal array to array copy, various element types
+            val eltsize = asmgen.options.compTarget.memorySize(ArrayToElementTypes.getValue(source.type))
+            val numBytes = numElements * eltsize
+            asmgen.out("""
+                lda  #<${sourceAsm}
+                ldy  #>${sourceAsm}
+                sta  P8ZP_SCRATCH_W1
+                sty  P8ZP_SCRATCH_W1+1
+                lda  #<${targetAsm}
+                ldy  #>${targetAsm}
+                sta  P8ZP_SCRATCH_W2
+                sty  P8ZP_SCRATCH_W2+1
+                ldy  #${numBytes and 255}
+                jsr  prog8_lib.memcopy_small""")
+        }
+    }
+
     private fun funcSquare(fcall: PtBuiltinFunctionCall, resultType: DataType, resultRegister: RegisterOrPair?) {
         // square of word value is faster with dedicated routine, square of byte just use the regular multiplication routine.
         when (resultType) {
@@ -225,6 +315,10 @@ internal class BuiltinFunctionsAsmGen(private val program: PtProgram,
                         asmgen.assignExpressionToRegister(arg1, RegisterOrPair.A)
                         asmgen.out("  cmp  #${arg2.number.toInt()}")
                     }
+                    is PtBool -> {
+                        asmgen.assignExpressionToRegister(arg1, RegisterOrPair.A)
+                        asmgen.out("  cmp  #${arg2.asInt()}")
+                    }
                     is PtMemoryByte -> {
                         if(arg2.address is PtNumber) {
                             asmgen.assignExpressionToRegister(arg1, RegisterOrPair.A)
@@ -253,6 +347,7 @@ internal class BuiltinFunctionsAsmGen(private val program: PtProgram,
                             cmp  ${asmgen.asmVariableName(arg2)}
 +""")
                     }
+                    is PtBool -> TODO("word compare against bool")
                     is PtNumber -> {
                         asmgen.assignExpressionToRegister(arg1, RegisterOrPair.AY)
                         asmgen.out("""
@@ -665,6 +760,8 @@ internal class BuiltinFunctionsAsmGen(private val program: PtProgram,
             }
             is PtAddressOf -> {
                 val mem = PtMemoryByte(fcall.position)
+                if((fcall.args[0] as PtAddressOf).isFromArrayElement)
+                    TODO("address-of arrayelement")
                 if(msb) {
                     val address = PtBinaryExpression("+", DataType.UWORD, fcall.args[0].position)
                     address.add(fcall.args[0])
@@ -1313,8 +1410,7 @@ internal class BuiltinFunctionsAsmGen(private val program: PtProgram,
                     ldy  #>$identifierName
                     sta  P8ZP_SCRATCH_W1
                     sty  P8ZP_SCRATCH_W1+1
-                    lda  #$numElements
-                    """)
+                    lda  #${numElements and 255}""")
     }
 
     private fun translateArguments(call: PtBuiltinFunctionCall, scope: IPtSubroutine?) {

codeGenCpu6502/src/prog8/codegen/cpu6502/Extensions.kt

@@ -17,7 +17,7 @@ internal fun IPtSubroutine.returnsWhatWhere(): List<Pair<RegisterOrStatusflag, D
                 emptyList()
             else {
                 val register = when (returntype!!) {
-                    in ByteDatatypes -> RegisterOrStatusflag(RegisterOrPair.A, null)
+                    in ByteDatatypesWithBoolean -> RegisterOrStatusflag(RegisterOrPair.A, null)
                     in WordDatatypes -> RegisterOrStatusflag(RegisterOrPair.AY, null)
                     DataType.FLOAT -> RegisterOrStatusflag(RegisterOrPair.FAC1, null)
                     else -> RegisterOrStatusflag(RegisterOrPair.AY, null)
@@ -31,7 +31,7 @@ internal fun IPtSubroutine.returnsWhatWhere(): List<Pair<RegisterOrStatusflag, D
 
 internal fun PtSub.returnRegister(): RegisterOrStatusflag? {
     return when(returntype) {
-        in ByteDatatypes -> RegisterOrStatusflag(RegisterOrPair.A, null)
+        in ByteDatatypesWithBoolean -> RegisterOrStatusflag(RegisterOrPair.A, null)
         in WordDatatypes -> RegisterOrStatusflag(RegisterOrPair.AY, null)
         DataType.FLOAT -> RegisterOrStatusflag(RegisterOrPair.FAC1, null)
         null -> null

codeGenCpu6502/src/prog8/codegen/cpu6502/ForLoopsAsmGen.kt

@@ -359,7 +359,7 @@ $loopLabel          lda  ${65535.toHex()}       ; modified
                     bne  $loopLabel
 $endLabel""")
             }
-            DataType.ARRAY_UB, DataType.ARRAY_B -> {
+            DataType.ARRAY_UB, DataType.ARRAY_B, DataType.ARRAY_BOOL -> {
                 val indexVar = asmgen.makeLabel("for_index")
                 asmgen.out("""
                     ldy  #0

codeGenCpu6502/src/prog8/codegen/cpu6502/FunctionCallAsmGen.kt

@@ -16,8 +16,8 @@ internal class FunctionCallAsmGen(private val program: PtProgram, private val as
     }
 
     internal fun optimizeIntArgsViaRegisters(sub: PtSub) =
-        (sub.parameters.size==1 && sub.parameters[0].type in IntegerDatatypes)
-                || (sub.parameters.size==2 && sub.parameters[0].type in ByteDatatypes && sub.parameters[1].type in ByteDatatypes)
+        (sub.parameters.size==1 && sub.parameters[0].type in IntegerDatatypesWithBoolean)
+                || (sub.parameters.size==2 && sub.parameters[0].type in ByteDatatypesWithBoolean && sub.parameters[1].type in ByteDatatypesWithBoolean)
 
     internal fun translateFunctionCall(call: PtFunctionCall) {
         // Output only the code to set up the parameters and perform the actual call
@@ -46,7 +46,7 @@ internal class FunctionCallAsmGen(private val program: PtProgram, private val as
         else if(sub is PtSub) {
             if(optimizeIntArgsViaRegisters(sub)) {
                 if(sub.parameters.size==1) {
-                    val register = if (sub.parameters[0].type in ByteDatatypes) RegisterOrPair.A else RegisterOrPair.AY
+                    val register = if (sub.parameters[0].type in ByteDatatypesWithBoolean) RegisterOrPair.A else RegisterOrPair.AY
                     argumentViaRegister(sub, IndexedValue(0, sub.parameters[0]), call.args[0], register)
                 } else {
                     // 2 byte params, second in Y, first in A
@@ -82,8 +82,9 @@ internal class FunctionCallAsmGen(private val program: PtProgram, private val as
             is PtAddressOf -> false
             is PtIdentifier -> false
             is PtMachineRegister -> false
-            is PtMemoryByte -> false
+            is PtMemoryByte -> return usesOtherRegistersWhileEvaluating(arg.address)
             is PtNumber -> false
+            is PtBool -> false
             else -> true
         }
     }
@@ -152,7 +153,7 @@ internal class FunctionCallAsmGen(private val program: PtProgram, private val as
         }
         if (statusflag!=null) {
             if(requiredDt!=value.type)
-                throw AssemblyError("for statusflag, byte value is required")
+                throw AssemblyError("for statusflag, byte or bool value is required")
             if (statusflag == Statusflag.Pc) {
                 // this boolean param needs to be set last, right before the jsr
                 // for now, this is already enforced on the subroutine definition by the Ast Checker
@@ -161,6 +162,9 @@ internal class FunctionCallAsmGen(private val program: PtProgram, private val as
                         val carrySet = value.number.toInt() != 0
                         asmgen.out(if(carrySet) "  sec" else "  clc")
                     }
+                    is PtBool -> {
+                        asmgen.out(if(value.value) "  sec" else "  clc")
+                    }
                     is PtIdentifier -> {
                         val sourceName = asmgen.asmVariableName(value)
                         // note: cannot use X register here to store A because it might be used for other arguments
@@ -218,6 +222,8 @@ internal class FunctionCallAsmGen(private val program: PtProgram, private val as
     private fun isArgumentTypeCompatible(argType: DataType, paramType: DataType): Boolean {
         if(argType isAssignableTo paramType)
             return true
+        if(argType==DataType.BOOL && paramType==DataType.BOOL)
+            return true
         if(argType in ByteDatatypes && paramType in ByteDatatypes)
             return true
         if(argType in WordDatatypes && paramType in WordDatatypes)

codeGenCpu6502/src/prog8/codegen/cpu6502/PostIncrDecrAsmGen.kt

@@ -1,191 +0,0 @@
-package prog8.codegen.cpu6502
-
-import prog8.code.ast.*
-import prog8.code.core.*
-
-
-internal class PostIncrDecrAsmGen(private val program: PtProgram, private val asmgen: AsmGen6502Internal) {
-    internal fun translate(stmt: PtPostIncrDecr) {
-        val incr = stmt.operator=="++"
-        val targetIdent = stmt.target.identifier
-        val targetMemory = stmt.target.memory
-        val targetArrayIdx = stmt.target.array
-        when {
-            targetIdent!=null -> {
-                val what = asmgen.asmVariableName(targetIdent)
-                when (stmt.target.type) {
-                    in ByteDatatypes -> asmgen.out(if (incr) "  inc  $what" else "  dec  $what")
-                    in WordDatatypes -> {
-                        if(incr)
-                            asmgen.out(" inc  $what |  bne  + |  inc  $what+1 |+")
-                        else
-                            asmgen.out("""
-        lda  $what
-        bne  +
-        dec  $what+1
-+       dec  $what 
-""")
-                    }
-                    DataType.FLOAT -> {
-                        asmgen.out("  lda  #<$what |  ldy  #>$what")
-                        asmgen.out(if(incr) "  jsr  floats.inc_var_f" else "  jsr  floats.dec_var_f")
-                    }
-                    else -> throw AssemblyError("need numeric type")
-                }
-            }
-            targetMemory!=null -> {
-                fun incDecViaExprEval() {
-                    asmgen.assignExpressionToRegister(targetMemory.address, RegisterOrPair.AY)
-                    asmgen.out("  sta  (+) + 1 |  sty  (+) + 2")
-                    if(incr)
-                        asmgen.out("+\tinc  ${'$'}ffff\t; modified")
-                    else
-                        asmgen.out("+\tdec  ${'$'}ffff\t; modified")
-                }
-
-                fun tryOptimizedPointerIncDec(address: PtBinaryExpression): Boolean {
-                    if(address.operator=="+") {
-                        val offset = address.right.asConstInteger()
-                        if(offset!=null && offset<256) {
-                            // we have @(ptr + 255) ++ , or  @(ptr+255)--
-                            asmgen.assignExpressionToRegister(address.left, RegisterOrPair.AY, false)
-                            asmgen.out("""
-                                sta  (+) + 1
-                                sty  (+) + 2
-                                ldx  #$offset""")
-                            if(incr)
-                                asmgen.out("+\tinc  ${'$'}ffff,x\t; modified")
-                            else
-                                asmgen.out("+\tdec  ${'$'}ffff,x\t; modified")
-                            return true
-                        } else if(address.right.type in ByteDatatypes) {
-                            // we have @(ptr + bytevar) ++ , or  @(ptr+bytevar)--
-                            asmgen.assignExpressionToRegister(address.left, RegisterOrPair.AY, false)
-                            asmgen.out(" sta  (+) + 1 |  sty  (+) + 2")
-                            asmgen.assignExpressionToRegister(address.right, RegisterOrPair.X, false)
-                            if(incr)
-                                asmgen.out("+\tinc  ${'$'}ffff,x\t; modified")
-                            else
-                                asmgen.out("+\tdec  ${'$'}ffff,x\t; modified")
-                            return true
-                        } else if((address.right as? PtTypeCast)?.value?.type in ByteDatatypes) {
-                            // we have @(ptr + bytevar) ++ , or  @(ptr+bytevar)--
-                            asmgen.assignExpressionToRegister(address.left, RegisterOrPair.AY, false)
-                            asmgen.out(" sta  (+) + 1 |  sty  (+) + 2")
-                            asmgen.assignExpressionToRegister((address.right as PtTypeCast).value, RegisterOrPair.X, false)
-                            if(incr)
-                                asmgen.out("+\tinc  ${'$'}ffff,x\t; modified")
-                            else
-                                asmgen.out("+\tdec  ${'$'}ffff,x\t; modified")
-                            return true
-                        }
-                    }
-                    return false
-                }
-
-                when (val addressExpr = targetMemory.address) {
-                    is PtNumber -> {
-                        val what = addressExpr.number.toHex()
-                        asmgen.out(if(incr) "  inc  $what" else "  dec  $what")
-                    }
-                    is PtIdentifier -> {
-                        val what = asmgen.asmVariableName(addressExpr)
-                        asmgen.out("  lda  $what |  sta  (+) +1 |  lda  $what+1 |  sta  (+) +2")
-                        if(incr)
-                            asmgen.out("+\tinc  ${'$'}ffff\t; modified")
-                        else
-                            asmgen.out("+\tdec  ${'$'}ffff\t; modified")
-                    }
-                    is PtBinaryExpression -> {
-                        if(!tryOptimizedPointerIncDec(addressExpr)) {
-                            incDecViaExprEval()
-                        }
-                    }
-                    else -> {
-                        incDecViaExprEval()
-                    }
-                }
-            }
-            targetArrayIdx!=null -> {
-                val asmArrayvarname = asmgen.asmVariableName(targetArrayIdx.variable)
-                val elementDt = targetArrayIdx.type
-                val constIndex = targetArrayIdx.index.asConstInteger()
-                if(targetArrayIdx.splitWords) {
-                    if(constIndex!=null) {
-                        if(incr)
-                            asmgen.out(" inc  ${asmArrayvarname}_lsb+$constIndex |  bne  + |  inc  ${asmArrayvarname}_msb+$constIndex |+")
-                        else
-                            asmgen.out("""
-        lda  ${asmArrayvarname}_lsb+$constIndex
-        bne  +
-        dec  ${asmArrayvarname}_msb+$constIndex
-+       dec  ${asmArrayvarname}_lsb+$constIndex""")
-                    } else {
-                        asmgen.loadScaledArrayIndexIntoRegister(targetArrayIdx, CpuRegister.X)
-                        if(incr)
-                            asmgen.out(" inc  ${asmArrayvarname}_lsb,x |  bne  + |  inc  ${asmArrayvarname}_msb,x |+")
-                        else
-                            asmgen.out("""
-        lda  ${asmArrayvarname}_lsb,x
-        bne  +
-        dec  ${asmArrayvarname}_msb,x
-+       dec  ${asmArrayvarname}_lsb,x""")
-                    }
-                    return
-                }
-                if(constIndex!=null) {
-                    val indexValue = constIndex * program.memsizer.memorySize(elementDt)
-                    when(elementDt) {
-                        in ByteDatatypes -> {
-                            asmgen.out(if (incr) "  inc  $asmArrayvarname+$indexValue" else "  dec  $asmArrayvarname+$indexValue")
-                        }
-                        in WordDatatypes -> {
-                            if(incr)
-                                asmgen.out(" inc  $asmArrayvarname+$indexValue |  bne  + |  inc  $asmArrayvarname+$indexValue+1 |+")
-                            else
-                                asmgen.out("""
-        lda  $asmArrayvarname+$indexValue
-        bne  +
-        dec  $asmArrayvarname+$indexValue+1
-+       dec  $asmArrayvarname+$indexValue""")
-                        }
-                        DataType.FLOAT -> {
-                            asmgen.out("  lda  #<($asmArrayvarname+$indexValue) |  ldy  #>($asmArrayvarname+$indexValue)")
-                            asmgen.out(if(incr) "  jsr  floats.inc_var_f" else "  jsr  floats.dec_var_f")
-                        }
-                        else -> throw AssemblyError("need numeric type")
-                    }
-                }
-                else
-                {
-                    asmgen.loadScaledArrayIndexIntoRegister(targetArrayIdx, CpuRegister.X)
-                    when(elementDt) {
-                        in ByteDatatypes -> {
-                            asmgen.out(if (incr) "  inc  $asmArrayvarname,x" else "  dec  $asmArrayvarname,x")
-                        }
-                        in WordDatatypes -> {
-                            if(incr)
-                                asmgen.out(" inc  $asmArrayvarname,x |  bne  + |  inc  $asmArrayvarname+1,x |+")
-                            else
-                                asmgen.out("""
-        lda  $asmArrayvarname,x
-        bne  +
-        dec  $asmArrayvarname+1,x
-+       dec  $asmArrayvarname,x""")
-                        }
-                        DataType.FLOAT -> {
-                            asmgen.out("""
-        ldy  #>$asmArrayvarname
-        clc
-        adc  #<$asmArrayvarname
-        bcc  +
-        iny
-+       jsr  floats.inc_var_f""")
-                        }
-                        else -> throw AssemblyError("weird array elt dt")
-                    }
-                }
-            }
-        }
-    }
-}

codeGenCpu6502/src/prog8/codegen/cpu6502/ProgramAndVarsGen.kt

@@ -173,6 +173,7 @@ internal class ProgramAndVarsGen(
                 for(num in 1..count) {
                     val name = asmgen.buildTempVarName(dt, num)
                     when (dt) {
+                        DataType.BOOL  -> asmgen.out("$name    .byte  ?")
                         DataType.BYTE  -> asmgen.out("$name    .char  ?")
                         DataType.UBYTE -> asmgen.out("$name    .byte  ?")
                         DataType.WORD  -> asmgen.out("$name    .sint  ?")
@@ -404,7 +405,7 @@ internal class ProgramAndVarsGen(
             if(sub.parameters.size==1) {
                 val dt = sub.parameters[0].type
                 val target = AsmAssignTarget(TargetStorageKind.VARIABLE, asmgen, dt, sub, sub.parameters[0].position, variableAsmName = sub.parameters[0].name)
-                if(dt in ByteDatatypes)
+                if(dt in ByteDatatypesWithBoolean)
                     asmgen.assignRegister(RegisterOrPair.A, target)
                 else
                     asmgen.assignRegister(RegisterOrPair.AY, target)
@@ -604,7 +605,7 @@ internal class ProgramAndVarsGen(
 
     private fun uninitializedVariable2asm(variable: StStaticVariable) {
         when (variable.dt) {
-            DataType.UBYTE -> asmgen.out("${variable.name}\t.byte  ?")
+            DataType.BOOL, DataType.UBYTE -> asmgen.out("${variable.name}\t.byte  ?")
             DataType.BYTE -> asmgen.out("${variable.name}\t.char  ?")
             DataType.UWORD -> asmgen.out("${variable.name}\t.word  ?")
             DataType.WORD -> asmgen.out("${variable.name}\t.sint  ?")
@@ -634,6 +635,7 @@ internal class ProgramAndVarsGen(
             } else 0
 
         when (variable.dt) {
+            DataType.BOOL -> TODO("bool var to asm")
             DataType.UBYTE -> asmgen.out("${variable.name}\t.byte  ${initialValue.toHex()}")
             DataType.BYTE -> asmgen.out("${variable.name}\t.char  $initialValue")
             DataType.UWORD -> asmgen.out("${variable.name}\t.word  ${initialValue.toHex()}")
@@ -658,7 +660,7 @@ internal class ProgramAndVarsGen(
 
     private fun arrayVariable2asm(varname: String, dt: DataType, value: StArray?, orNumberOfZeros: Int?) {
         when(dt) {
-            DataType.ARRAY_UB -> {
+            DataType.ARRAY_UB, DataType.ARRAY_BOOL -> {
                 val data = makeArrayFillDataUnsigned(dt, value, orNumberOfZeros)
                 if (data.size <= 16)
                     asmgen.out("$varname\t.byte  ${data.joinToString()}")
@@ -726,7 +728,7 @@ internal class ProgramAndVarsGen(
     private fun zeroFilledArray(numElts: Int): StArray {
         val values = mutableListOf<StArrayElement>()
         repeat(numElts) {
-            values.add(StArrayElement(0.0, null))
+            values.add(StArrayElement(0.0, null, null))
         }
         return values
     }
@@ -763,6 +765,16 @@ internal class ProgramAndVarsGen(
     private fun makeArrayFillDataUnsigned(dt: DataType, value: StArray?, orNumberOfZeros: Int?): List<String> {
         val array = value ?: zeroFilledArray(orNumberOfZeros!!)
         return when (dt) {
+            DataType.ARRAY_BOOL ->
+                // byte array can never contain pointer-to types, so treat values as all integers
+                array.map {
+                    if(it.boolean!=null)
+                        if(it.boolean==true) "1" else "0"
+                    else {
+                        val number = it.number!!
+                        if(number==0.0) "0" else "1"
+                    }
+                }
             DataType.ARRAY_UB ->
                 // byte array can never contain pointer-to types, so treat values as all integers
                 array.map {

codeGenCpu6502/src/prog8/codegen/cpu6502/VariableAllocator.kt

@@ -91,7 +91,7 @@ internal class VariableAllocator(private val symboltable: SymbolTable,
             if(errors.noErrors()) {
                 val sortedList = varsDontCare.sortedByDescending { it.scopedName }
                 for (variable in sortedList) {
-                    if(variable.dt in IntegerDatatypes) {
+                    if(variable.dt in IntegerDatatypesWithBoolean) {
                         if(zeropage.free.isEmpty()) {
                             break
                         } else {

codeGenCpu6502/src/prog8/codegen/cpu6502/assignment/AnyExprAsmGen.kt

@@ -18,12 +18,12 @@ internal class AnyExprAsmGen(
 ) {
     fun assignAnyExpressionUsingStack(expr: PtBinaryExpression, assign: AsmAssignment): Boolean {
         when(expr.type) {
-            in ByteDatatypes -> {
-                if(expr.left.type in ByteDatatypes && expr.right.type in ByteDatatypes)
+            in ByteDatatypesWithBoolean -> {
+                if(expr.left.type in ByteDatatypesWithBoolean && expr.right.type in ByteDatatypesWithBoolean)
                     return assignByteBinExpr(expr, assign)
                 if (expr.left.type in WordDatatypes && expr.right.type in WordDatatypes) {
                     require(expr.operator in ComparisonOperators)
-                    throw AssemblyError("words operands comparison -> byte, should have been handled by assignOptimizedComparisonWords()")
+                    throw AssemblyError("words operands comparison -> byte, should have been handled elsewhere")
                 }
                 if (expr.left.type==DataType.FLOAT && expr.right.type==DataType.FLOAT) {
                     require(expr.operator in ComparisonOperators)

codeGenCpu6502/src/prog8/codegen/cpu6502/assignment/AsmAssignment.kt

@@ -14,6 +14,7 @@ internal enum class TargetStorageKind {
 }
 
 internal enum class SourceStorageKind {
+    LITERALBOOLEAN,
     LITERALNUMBER,
     VARIABLE,
     ARRAY,
@@ -43,8 +44,8 @@ internal class AsmAssignTarget(val kind: TargetStorageKind,
     }
 
     init {
-        if(register!=null && datatype !in NumericDatatypes)
-            throw AssemblyError("register must be integer or float type")
+        if(register!=null && datatype !in NumericDatatypesWithBoolean)
+            throw AssemblyError("must be numeric type")
     }
 
     companion object {
@@ -133,6 +134,7 @@ internal class AsmAssignSource(val kind: SourceStorageKind,
                                val memory: PtMemoryByte? = null,
                                val register: RegisterOrPair? = null,
                                val number: PtNumber? = null,
+                               val boolean: PtBool? = null,
                                val expression: PtExpression? = null
 )
 {
@@ -147,6 +149,9 @@ internal class AsmAssignSource(val kind: SourceStorageKind,
             val cv = value as? PtNumber
             if(cv!=null)
                 return AsmAssignSource(SourceStorageKind.LITERALNUMBER, program, asmgen, cv.type, number = cv)
+            val bv = value as? PtBool
+            if(bv!=null)
+                return AsmAssignSource(SourceStorageKind.LITERALBOOLEAN, program, asmgen, DataType.BOOL, boolean = bv)
 
             return when(value) {
                 // checked above:   is PtNumber -> throw AssemblyError("should have been constant value")
@@ -194,7 +199,7 @@ internal class AsmAssignSource(val kind: SourceStorageKind,
         // allow some signed/unsigned relaxations
 
         fun withAdjustedDt(newType: DataType) =
-                AsmAssignSource(kind, program, asmgen, newType, variableAsmName, array, memory, register, number, expression)
+                AsmAssignSource(kind, program, asmgen, newType, variableAsmName, array, memory, register, number, boolean, expression)
 
         if(target.datatype!=datatype) {
             if(target.datatype in ByteDatatypes && datatype in ByteDatatypes) {

codeGenCpu6502/test/Dummies.kt

@@ -5,7 +5,7 @@ import prog8.code.core.*
 
 internal object DummyMemsizer : IMemSizer {
     override fun memorySize(dt: DataType) = when(dt) {
-        in ByteDatatypes -> 1
+        in ByteDatatypesWithBoolean -> 1
         DataType.FLOAT -> 5
         else -> 2
     }
@@ -58,6 +58,7 @@ internal class ErrorReporterForTests(private val throwExceptionAtReportIfErrors:
     }
 
     override fun noErrors(): Boolean  = errors.isEmpty()
+    override fun noErrorForLine(position: Position) = !errors.any { ":${position.line}:" in it }
 
     override fun report() {
         infos.forEach { println("UNITTEST COMPILATION REPORT: INFO: $it") }

codeGenExperimental/build.gradle

@@ -29,7 +29,7 @@ dependencies {
     implementation project(':codeGenIntermediate')
     implementation "org.jetbrains.kotlin:kotlin-stdlib-jdk8"
     // implementation "org.jetbrains.kotlin:kotlin-reflect"
-    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.18"
+    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.20"
 
 }
 

codeGenIntermediate/build.gradle

@@ -28,9 +28,12 @@ dependencies {
     implementation project(':intermediate')
     implementation "org.jetbrains.kotlin:kotlin-stdlib-jdk8"
     // implementation "org.jetbrains.kotlin:kotlin-reflect"
-    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.18"
+    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.20"
 
     testImplementation 'io.kotest:kotest-runner-junit5-jvm:5.8.0'
+    testImplementation 'org.junit.jupiter:junit-jupiter:5.9.2'
+    testRuntimeOnly 'org.junit.platform:junit-platform-launcher'
+
 }
 
 sourceSets {

codeGenIntermediate/src/prog8/codegen/intermediate/BuiltinFuncGen.kt

@@ -1,10 +1,7 @@
 package prog8.codegen.intermediate
 
 import prog8.code.ast.*
-import prog8.code.core.AssemblyError
-import prog8.code.core.DataType
-import prog8.code.core.SignedDatatypes
-import prog8.code.core.SplitWordArrayTypes
+import prog8.code.core.*
 import prog8.intermediate.*
 
 
@@ -48,10 +45,72 @@ internal class BuiltinFuncGen(private val codeGen: IRCodeGen, private val exprGe
             "prog8_lib_stringcompare" -> funcStringCompare(call)
             "prog8_lib_square_byte" -> funcSquare(call, IRDataType.BYTE)
             "prog8_lib_square_word" -> funcSquare(call, IRDataType.WORD)
+            "prog8_lib_arraycopy" -> funcArrayCopy(call)
             else -> throw AssemblyError("missing builtinfunc for ${call.name}")
         }
     }
 
+    private fun funcArrayCopy(call: PtBuiltinFunctionCall): ExpressionCodeResult {
+        val source = call.args[0] as PtIdentifier
+        val target = call.args[1] as PtIdentifier
+        val sourceLength = codeGen.symbolTable.getLength(source.name)!!
+        val targetLength = codeGen.symbolTable.getLength(target.name)!!
+        require(sourceLength==targetLength)
+        val result = mutableListOf<IRCodeChunkBase>()
+        val fromReg = codeGen.registers.nextFree()
+        val toReg = codeGen.registers.nextFree()
+        val countReg = codeGen.registers.nextFree()
+        if(source.type in SplitWordArrayTypes && target.type in SplitWordArrayTypes) {
+            // split words -> split words, copy lsb and msb arrays separately
+            result += IRCodeChunk(null, null).also {
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=fromReg, labelSymbol = source.name+"_lsb")
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=toReg, labelSymbol = target.name+"_lsb")
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=countReg, immediate = sourceLength)
+                it += codeGen.makeSyscall(IMSyscall.MEMCOPY_SMALL, listOf(IRDataType.WORD to fromReg, IRDataType.WORD to toReg, IRDataType.BYTE to (countReg and 255)), returns = null)
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=fromReg, labelSymbol = source.name+"_msb")
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=toReg, labelSymbol = target.name+"_msb")
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=countReg, immediate = sourceLength)
+                it += codeGen.makeSyscall(IMSyscall.MEMCOPY_SMALL, listOf(IRDataType.WORD to fromReg, IRDataType.WORD to toReg, IRDataType.BYTE to (countReg and 255)), returns = null)
+            }
+        }
+        else if(source.type in SplitWordArrayTypes) {
+            // split -> normal words
+            require(target.type==DataType.ARRAY_UW || target.type==DataType.ARRAY_W)
+            val fromRegMsb = codeGen.registers.nextFree()
+            result += IRCodeChunk(null, null).also {
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=fromReg, labelSymbol = source.name+"_lsb")
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=fromRegMsb, labelSymbol = source.name+"_msb")
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=toReg, labelSymbol = target.name)
+                it += IRInstruction(Opcode.LOAD, IRDataType.BYTE, reg1=countReg, immediate = sourceLength)
+            }
+            result += codeGen.makeSyscall(IMSyscall.ARRAYCOPY_SPLITW_TO_NORMAL, listOf(IRDataType.WORD to fromReg, IRDataType.WORD to fromRegMsb, IRDataType.WORD to toReg, IRDataType.BYTE to countReg), returns = null)
+        }
+        else if(target.type in SplitWordArrayTypes) {
+            // normal -> split words
+            require(source.type==DataType.ARRAY_UW || source.type==DataType.ARRAY_W)
+            val toRegMsb = codeGen.registers.nextFree()
+            result += IRCodeChunk(null, null).also {
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=fromReg, labelSymbol = source.name)
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=toReg, labelSymbol = target.name+"_lsb")
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=toRegMsb, labelSymbol = target.name+"_msb")
+                it += IRInstruction(Opcode.LOAD, IRDataType.BYTE, reg1=countReg, immediate = sourceLength)
+            }
+            result += codeGen.makeSyscall(IMSyscall.ARRAYCOPY_NORMAL_TO_SPLITW, listOf(IRDataType.WORD to fromReg, IRDataType.WORD to toReg, IRDataType.WORD to toRegMsb, IRDataType.BYTE to countReg), returns = null)
+        }
+        else {
+            // normal array to array copy (various element types)
+            val eltsize = codeGen.options.compTarget.memorySize(ArrayToElementTypes.getValue(source.type))
+            result += IRCodeChunk(null, null).also {
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=fromReg, labelSymbol = source.name)
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=toReg, labelSymbol = target.name)
+                it += IRInstruction(Opcode.LOAD, IRDataType.WORD, reg1=countReg, immediate = sourceLength * eltsize)
+            }
+            result += codeGen.makeSyscall(IMSyscall.MEMCOPY_SMALL, listOf(IRDataType.WORD to fromReg, IRDataType.WORD to toReg, IRDataType.BYTE to (countReg and 255)), returns = null)
+        }
+
+        return ExpressionCodeResult(result, IRDataType.BYTE, -1, -1)
+    }
+
     private fun funcSquare(call: PtBuiltinFunctionCall, resultType: IRDataType): ExpressionCodeResult {
         val result = mutableListOf<IRCodeChunkBase>()
         val valueTr = exprGen.translateExpression(call.args[0])
@@ -185,7 +244,7 @@ internal class BuiltinFuncGen(private val codeGen: IRCodeGen, private val exprGe
         addInstr(result, IRInstruction(Opcode.PREPARECALL, immediate = 2), null)
         val tr = exprGen.translateExpression(arrayName)
         addToResult(result, tr, tr.resultReg, -1)
-        addInstr(result, IRInstruction(Opcode.LOAD, IRDataType.BYTE, reg1 = lengthReg, immediate = arrayLength), null)
+        addInstr(result, IRInstruction(Opcode.LOAD, IRDataType.BYTE, reg1 = lengthReg, immediate = arrayLength!! and 255), null)
         result += codeGen.makeSyscall(syscall, listOf(IRDataType.WORD to tr.resultReg, IRDataType.BYTE to lengthReg), IRDataType.BYTE to tr.resultReg)
         return ExpressionCodeResult(result, IRDataType.BYTE, tr.resultReg, -1)
     }
@@ -213,7 +272,7 @@ internal class BuiltinFuncGen(private val codeGen: IRCodeGen, private val exprGe
         val tr = exprGen.translateExpression(arrayName)
         addToResult(result, tr, tr.resultReg, -1)
         val lengthReg = codeGen.registers.nextFree()
-        addInstr(result, IRInstruction(Opcode.LOAD, IRDataType.BYTE, reg1 = lengthReg, immediate = arrayLength), null)
+        addInstr(result, IRInstruction(Opcode.LOAD, IRDataType.BYTE, reg1 = lengthReg, immediate = arrayLength!! and 255), null)
         result += codeGen.makeSyscall(syscall, listOf(IRDataType.WORD to tr.resultReg, IRDataType.BYTE to lengthReg), IRDataType.BYTE to tr.resultReg)
         return ExpressionCodeResult(result, IRDataType.BYTE, tr.resultReg, -1)
     }

codeGenIntermediate/src/prog8/codegen/intermediate/IRPeepholeOptimizer.kt

@@ -337,10 +337,33 @@ class IRPeepholeOptimizer(private val irprog: IRProgram) {
                     }
                 }
             }
+
+            // a SNZ etc. whose target register is not used can be removed altogether
+            if(ins.opcode in OpcodesThatSetRegFromStatusbits) {
+                val usages = regUsages(ins.reg1!!)
+                if(usages.toList().sumOf { it.second } <= 1) {
+                    chunk.instructions.removeAt(idx)
+                    changed = true
+                }
+            }
         }
         return changed
     }
 
+    private fun regUsages(register: Int): Map<IRCodeChunkBase, Int> {
+        val chunks = mutableMapOf<IRCodeChunkBase, Int>()
+        irprog.foreachSub { sub ->
+            sub.chunks.forEach { chunk ->
+                val used = chunk.usedRegisters()
+                val numUsages = used.readRegs.getOrDefault(register, 0) + used.writeRegs.getOrDefault(register, 0)
+                if(numUsages>0) {
+                    chunks[chunk] = numUsages
+                }
+            }
+        }
+        return chunks
+    }
+
     private fun removeUselessArithmetic(chunk: IRCodeChunk, indexedInstructions: List<IndexedValue<IRInstruction>>): Boolean {
         // note: this is hard to solve for the non-immediate instructions atm because the values are loaded into registers first
         var changed = false

codeGenIntermediate/src/prog8/codegen/intermediate/IRUnusedCodeRemover.kt

@@ -148,6 +148,7 @@ class IRUnusedCodeRemover(
         val entrypointSub = irprog.blocks.single { it.label=="main" }
             .children.single { it is IRSubroutine && it.label=="main.start" }
         val reachable = mutableSetOf((entrypointSub as IRSubroutine).chunks.first())
+        reachable.add(irprog.globalInits)
 
         // all chunks referenced in array initializer values are also 'reachable':
         irprog.st.allVariables()
@@ -168,7 +169,7 @@ class IRUnusedCodeRemover(
                 it.instructions.forEach { instr ->
                     if (instr.branchTarget == null)
                         instr.labelSymbol?.let { label ->
-                            val chunk = allLabeledChunks[label.substringBeforeLast('.')]
+                            val chunk = allLabeledChunks[label] ?: allLabeledChunks[label.substringBeforeLast('.')]
                             if(chunk!=null)
                                 new+=chunk
                             else
@@ -230,6 +231,7 @@ class IRUnusedCodeRemover(
             }
         }
 
+        linkedChunks.add(irprog.globalInits)
         return removeUnlinkedChunks(linkedChunks)
     }
 

codeGenIntermediate/src/prog8/codegen/intermediate/PreProcess.kt

@@ -0,0 +1,47 @@
+package prog8.codegen.intermediate
+
+import prog8.code.ast.*
+
+
+internal fun makeAllNodenamesScoped(program: PtProgram) {
+    val renames = mutableListOf<Pair<PtNamedNode, String>>()
+    fun recurse(node: PtNode) {
+        node.children.forEach {
+            if(it is PtNamedNode)
+                renames.add(it to it.scopedName)
+            recurse(it)
+        }
+    }
+    recurse(program)
+    renames.forEach { it.first.name = it.second }
+}
+
+internal fun moveAllNestedSubroutinesToBlockScope(program: PtProgram) {
+    val movedSubs = mutableListOf<Pair<PtBlock, PtSub>>()
+    val removedSubs = mutableListOf<Pair<PtSub, PtSub>>()
+
+    fun moveToBlock(block: PtBlock, parent: PtSub, asmsub: PtAsmSub) {
+        block.add(asmsub)
+        parent.children.remove(asmsub)
+    }
+
+    fun moveToBlock(block: PtBlock, parent: PtSub, sub: PtSub) {
+        sub.children.filterIsInstance<PtSub>().forEach { subsub -> moveToBlock(block, sub, subsub) }
+        sub.children.filterIsInstance<PtAsmSub>().forEach { asmsubsub -> moveToBlock(block, sub, asmsubsub) }
+        movedSubs += Pair(block, sub)
+        removedSubs += Pair(parent, sub)
+    }
+
+    program.allBlocks().forEach { block ->
+        block.children.toList().forEach {
+            if (it is PtSub) {
+                // Only regular subroutines can have nested subroutines.
+                it.children.filterIsInstance<PtSub>().forEach { subsub -> moveToBlock(block, it, subsub) }
+                it.children.filterIsInstance<PtAsmSub>().forEach { asmsubsub -> moveToBlock(block, it, asmsubsub) }
+            }
+        }
+    }
+
+    removedSubs.forEach { (parent, sub) -> parent.children.remove(sub) }
+    movedSubs.forEach { (block, sub) -> block.add(sub) }
+}

codeGenIntermediate/test/Dummies.kt

@@ -3,7 +3,7 @@ import prog8.code.core.*
 
 internal object DummyMemsizer : IMemSizer {
     override fun memorySize(dt: DataType) = when(dt) {
-        in ByteDatatypes -> 1
+        in ByteDatatypesWithBoolean -> 1
         DataType.FLOAT -> 5
         else -> 2
     }
@@ -56,6 +56,7 @@ internal class ErrorReporterForTests(private val throwExceptionAtReportIfErrors:
     }
 
     override fun noErrors(): Boolean  = errors.isEmpty()
+    override fun noErrorForLine(position: Position) = !errors.any { ":${position.line}:" in it }
 
     override fun report() {
         infos.forEach { println("UNITTEST COMPILATION REPORT: INFO: $it") }

codeGenIntermediate/test/TestVmCodeGen.kt

@@ -125,7 +125,7 @@ class TestVmCodeGen: FunSpec({
         val sub = PtSub("start", emptyList(), null, Position.DUMMY)
         sub.add(PtVariable("f1", DataType.FLOAT, ZeropageWish.DONTCARE, null, null, Position.DUMMY))
         val if1 = PtIfElse(Position.DUMMY)
-        val cmp1 = PtBinaryExpression("==", DataType.UBYTE, Position.DUMMY)
+        val cmp1 = PtBinaryExpression("==", DataType.BOOL, Position.DUMMY)
         cmp1.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp1.add(PtNumber(DataType.FLOAT, 0.0, Position.DUMMY))
         if1.add(cmp1)
@@ -133,7 +133,7 @@ class TestVmCodeGen: FunSpec({
         if1.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if1)
         val if2 = PtIfElse(Position.DUMMY)
-        val cmp2 = PtBinaryExpression("!=", DataType.UBYTE, Position.DUMMY)
+        val cmp2 = PtBinaryExpression("!=", DataType.BOOL, Position.DUMMY)
         cmp2.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp2.add(PtNumber(DataType.FLOAT, 0.0, Position.DUMMY))
         if2.add(cmp2)
@@ -141,7 +141,7 @@ class TestVmCodeGen: FunSpec({
         if2.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if2)
         val if3 = PtIfElse(Position.DUMMY)
-        val cmp3 = PtBinaryExpression("<", DataType.UBYTE, Position.DUMMY)
+        val cmp3 = PtBinaryExpression("<", DataType.BOOL, Position.DUMMY)
         cmp3.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp3.add(PtNumber(DataType.FLOAT, 0.0, Position.DUMMY))
         if3.add(cmp3)
@@ -149,7 +149,7 @@ class TestVmCodeGen: FunSpec({
         if3.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if3)
         val if4 = PtIfElse(Position.DUMMY)
-        val cmp4 = PtBinaryExpression(">", DataType.UBYTE, Position.DUMMY)
+        val cmp4 = PtBinaryExpression(">", DataType.BOOL, Position.DUMMY)
         cmp4.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp4.add(PtNumber(DataType.FLOAT, 0.0, Position.DUMMY))
         if4.add(cmp4)
@@ -188,7 +188,7 @@ class TestVmCodeGen: FunSpec({
         val sub = PtSub("start", emptyList(), null, Position.DUMMY)
         sub.add(PtVariable("f1", DataType.FLOAT, ZeropageWish.DONTCARE, null, null, Position.DUMMY))
         val if1 = PtIfElse(Position.DUMMY)
-        val cmp1 = PtBinaryExpression("==", DataType.UBYTE, Position.DUMMY)
+        val cmp1 = PtBinaryExpression("==", DataType.BOOL, Position.DUMMY)
         cmp1.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp1.add(PtNumber(DataType.FLOAT, 42.0, Position.DUMMY))
         if1.add(cmp1)
@@ -196,7 +196,7 @@ class TestVmCodeGen: FunSpec({
         if1.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if1)
         val if2 = PtIfElse(Position.DUMMY)
-        val cmp2 = PtBinaryExpression("!=", DataType.UBYTE, Position.DUMMY)
+        val cmp2 = PtBinaryExpression("!=", DataType.BOOL, Position.DUMMY)
         cmp2.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp2.add(PtNumber(DataType.FLOAT, 42.0, Position.DUMMY))
         if2.add(cmp2)
@@ -204,7 +204,7 @@ class TestVmCodeGen: FunSpec({
         if2.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if2)
         val if3 = PtIfElse(Position.DUMMY)
-        val cmp3 = PtBinaryExpression("<", DataType.UBYTE, Position.DUMMY)
+        val cmp3 = PtBinaryExpression("<", DataType.BOOL, Position.DUMMY)
         cmp3.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp3.add(PtNumber(DataType.FLOAT, 42.0, Position.DUMMY))
         if3.add(cmp3)
@@ -212,7 +212,7 @@ class TestVmCodeGen: FunSpec({
         if3.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if3)
         val if4 = PtIfElse(Position.DUMMY)
-        val cmp4 = PtBinaryExpression(">", DataType.UBYTE, Position.DUMMY)
+        val cmp4 = PtBinaryExpression(">", DataType.BOOL, Position.DUMMY)
         cmp4.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp4.add(PtNumber(DataType.FLOAT, 42.0, Position.DUMMY))
         if4.add(cmp4)
@@ -247,7 +247,7 @@ class TestVmCodeGen: FunSpec({
         val sub = PtSub("start", emptyList(), null, Position.DUMMY)
         sub.add(PtVariable("f1", DataType.FLOAT, ZeropageWish.DONTCARE, null, null, Position.DUMMY))
         val if1 = PtIfElse(Position.DUMMY)
-        val cmp1 = PtBinaryExpression("==", DataType.UBYTE, Position.DUMMY)
+        val cmp1 = PtBinaryExpression("==", DataType.BOOL, Position.DUMMY)
         cmp1.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp1.add(PtNumber(DataType.FLOAT, 42.0, Position.DUMMY))
         if1.add(cmp1)
@@ -255,7 +255,7 @@ class TestVmCodeGen: FunSpec({
         if1.add(PtNodeGroup())
         sub.add(if1)
         val if2 = PtIfElse(Position.DUMMY)
-        val cmp2 = PtBinaryExpression(">", DataType.UBYTE, Position.DUMMY)
+        val cmp2 = PtBinaryExpression(">", DataType.BOOL, Position.DUMMY)
         cmp2.add(PtIdentifier("main.start.f1", DataType.FLOAT, Position.DUMMY))
         cmp2.add(PtNumber(DataType.FLOAT, 42.0, Position.DUMMY))
         if2.add(cmp2)
@@ -294,7 +294,7 @@ class TestVmCodeGen: FunSpec({
         val sub = PtSub("start", emptyList(), null, Position.DUMMY)
         sub.add(PtVariable("sb1", DataType.BYTE, ZeropageWish.DONTCARE, null, null, Position.DUMMY))
         val if1 = PtIfElse(Position.DUMMY)
-        val cmp1 = PtBinaryExpression("==", DataType.BYTE, Position.DUMMY)
+        val cmp1 = PtBinaryExpression("==", DataType.BOOL, Position.DUMMY)
         cmp1.add(PtIdentifier("main.start.sb1", DataType.BYTE, Position.DUMMY))
         cmp1.add(PtNumber(DataType.BYTE, 0.0, Position.DUMMY))
         if1.add(cmp1)
@@ -302,7 +302,7 @@ class TestVmCodeGen: FunSpec({
         if1.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if1)
         val if2 = PtIfElse(Position.DUMMY)
-        val cmp2 = PtBinaryExpression("!=", DataType.BYTE, Position.DUMMY)
+        val cmp2 = PtBinaryExpression("!=", DataType.BOOL, Position.DUMMY)
         cmp2.add(PtIdentifier("main.start.sb1", DataType.BYTE, Position.DUMMY))
         cmp2.add(PtNumber(DataType.BYTE, 0.0, Position.DUMMY))
         if2.add(cmp2)
@@ -310,7 +310,7 @@ class TestVmCodeGen: FunSpec({
         if2.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if2)
         val if3 = PtIfElse(Position.DUMMY)
-        val cmp3 = PtBinaryExpression("<", DataType.BYTE, Position.DUMMY)
+        val cmp3 = PtBinaryExpression("<", DataType.BOOL, Position.DUMMY)
         cmp3.add(PtIdentifier("main.start.sb1", DataType.BYTE, Position.DUMMY))
         cmp3.add(PtNumber(DataType.BYTE, 0.0, Position.DUMMY))
         if3.add(cmp3)
@@ -318,7 +318,7 @@ class TestVmCodeGen: FunSpec({
         if3.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if3)
         val if4 = PtIfElse(Position.DUMMY)
-        val cmp4 = PtBinaryExpression(">", DataType.BYTE, Position.DUMMY)
+        val cmp4 = PtBinaryExpression(">", DataType.BOOL, Position.DUMMY)
         cmp4.add(PtIdentifier("main.start.sb1", DataType.BYTE, Position.DUMMY))
         cmp4.add(PtNumber(DataType.BYTE, 0.0, Position.DUMMY))
         if4.add(cmp4)
@@ -357,7 +357,7 @@ class TestVmCodeGen: FunSpec({
         val sub = PtSub("start", emptyList(), null, Position.DUMMY)
         sub.add(PtVariable("sb1", DataType.BYTE, ZeropageWish.DONTCARE, null, null, Position.DUMMY))
         val if1 = PtIfElse(Position.DUMMY)
-        val cmp1 = PtBinaryExpression("==", DataType.BYTE, Position.DUMMY)
+        val cmp1 = PtBinaryExpression("==", DataType.BOOL, Position.DUMMY)
         cmp1.add(PtIdentifier("main.start.sb1", DataType.BYTE, Position.DUMMY))
         cmp1.add(PtNumber(DataType.BYTE, 42.0, Position.DUMMY))
         if1.add(cmp1)
@@ -365,7 +365,7 @@ class TestVmCodeGen: FunSpec({
         if1.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if1)
         val if2 = PtIfElse(Position.DUMMY)
-        val cmp2 = PtBinaryExpression("!=", DataType.BYTE, Position.DUMMY)
+        val cmp2 = PtBinaryExpression("!=", DataType.BOOL, Position.DUMMY)
         cmp2.add(PtIdentifier("main.start.sb1", DataType.BYTE, Position.DUMMY))
         cmp2.add(PtNumber(DataType.BYTE, 42.0, Position.DUMMY))
         if2.add(cmp2)
@@ -373,7 +373,7 @@ class TestVmCodeGen: FunSpec({
         if2.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if2)
         val if3 = PtIfElse(Position.DUMMY)
-        val cmp3 = PtBinaryExpression("<", DataType.BYTE, Position.DUMMY)
+        val cmp3 = PtBinaryExpression("<", DataType.BOOL, Position.DUMMY)
         cmp3.add(PtIdentifier("main.start.sb1", DataType.BYTE, Position.DUMMY))
         cmp3.add(PtNumber(DataType.BYTE, 42.0, Position.DUMMY))
         if3.add(cmp3)
@@ -381,7 +381,7 @@ class TestVmCodeGen: FunSpec({
         if3.add(PtNodeGroup().also { it.add(PtNop(Position.DUMMY)) })
         sub.add(if3)
         val if4 = PtIfElse(Position.DUMMY)
-        val cmp4 = PtBinaryExpression(">", DataType.BYTE, Position.DUMMY)
+        val cmp4 = PtBinaryExpression(">", DataType.BOOL, Position.DUMMY)
         cmp4.add(PtIdentifier("main.start.sb1", DataType.BYTE, Position.DUMMY))
         cmp4.add(PtNumber(DataType.BYTE, 42.0, Position.DUMMY))
         if4.add(cmp4)
@@ -416,7 +416,7 @@ class TestVmCodeGen: FunSpec({
         val sub = PtSub("start", emptyList(), null, Position.DUMMY)
         sub.add(PtVariable("ub1", DataType.UBYTE, ZeropageWish.DONTCARE, null, null, Position.DUMMY))
         val if1 = PtIfElse(Position.DUMMY)
-        val cmp1 = PtBinaryExpression("==", DataType.UBYTE, Position.DUMMY)
+        val cmp1 = PtBinaryExpression("==", DataType.BOOL, Position.DUMMY)
         cmp1.add(PtIdentifier("main.start.ub1", DataType.UBYTE, Position.DUMMY))
         cmp1.add(PtNumber(DataType.UBYTE, 42.0, Position.DUMMY))
         if1.add(cmp1)
@@ -424,7 +424,7 @@ class TestVmCodeGen: FunSpec({
         if1.add(PtNodeGroup())
         sub.add(if1)
         val if2 = PtIfElse(Position.DUMMY)
-        val cmp2 = PtBinaryExpression(">", DataType.UBYTE, Position.DUMMY)
+        val cmp2 = PtBinaryExpression(">", DataType.BOOL, Position.DUMMY)
         cmp2.add(PtIdentifier("main.start.ub1", DataType.UBYTE, Position.DUMMY))
         cmp2.add(PtNumber(DataType.UBYTE, 42.0, Position.DUMMY))
         if2.add(cmp2)

codeOptimizers/build.gradle

@@ -27,7 +27,7 @@ dependencies {
     implementation project(':codeCore')
     implementation project(':compilerAst')
     implementation "org.jetbrains.kotlin:kotlin-stdlib-jdk8"
-    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.18"
+    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.20"
     // implementation "org.jetbrains.kotlin:kotlin-reflect"
 }
 

codeOptimizers/src/prog8/optimizer/ConstExprEvaluator.kt

@@ -7,6 +7,7 @@ import prog8.ast.expressions.FunctionCallExpression
 import prog8.ast.expressions.NumericLiteral
 import prog8.code.core.DataType
 import prog8.code.core.IntegerDatatypes
+import prog8.code.core.IntegerDatatypesWithBoolean
 import prog8.code.core.Position
 import kotlin.math.*
 
@@ -69,15 +70,15 @@ class ConstExprEvaluator {
     }
 
     private fun bitwiseXor(left: NumericLiteral, right: NumericLiteral): NumericLiteral {
-        if(left.type== DataType.UBYTE) {
-            if(right.type in IntegerDatatypes) {
+        if(left.type==DataType.UBYTE || left.type==DataType.BOOL) {
+            if(right.type in IntegerDatatypesWithBoolean) {
                 return NumericLiteral(DataType.UBYTE, (left.number.toInt() xor (right.number.toInt() and 255)).toDouble(), left.position)
             }
-        } else if(left.type== DataType.UWORD) {
+        } else if(left.type==DataType.UWORD) {
             if(right.type in IntegerDatatypes) {
                 return NumericLiteral(DataType.UWORD, (left.number.toInt() xor right.number.toInt() and 65535).toDouble(), left.position)
             }
-        } else if(left.type== DataType.LONG) {
+        } else if(left.type==DataType.LONG) {
             if(right.type in IntegerDatatypes) {
                 return NumericLiteral.optimalNumeric((left.number.toInt() xor right.number.toInt()).toDouble(), left.position)
             }
@@ -86,15 +87,15 @@ class ConstExprEvaluator {
     }
 
     private fun bitwiseOr(left: NumericLiteral, right: NumericLiteral): NumericLiteral {
-        if(left.type== DataType.UBYTE) {
-            if(right.type in IntegerDatatypes) {
+        if(left.type==DataType.UBYTE || left.type==DataType.BOOL) {
+            if(right.type in IntegerDatatypesWithBoolean) {
                 return NumericLiteral(DataType.UBYTE, (left.number.toInt() or (right.number.toInt() and 255)).toDouble(), left.position)
             }
-        } else if(left.type== DataType.UWORD) {
+        } else if(left.type==DataType.UWORD) {
             if(right.type in IntegerDatatypes) {
                 return NumericLiteral(DataType.UWORD, (left.number.toInt() or right.number.toInt() and 65535).toDouble(), left.position)
             }
-        } else if(left.type== DataType.LONG) {
+        } else if(left.type==DataType.LONG) {
             if(right.type in IntegerDatatypes) {
                 return NumericLiteral.optimalNumeric((left.number.toInt() or right.number.toInt()).toDouble(), left.position)
             }
@@ -103,11 +104,11 @@ class ConstExprEvaluator {
     }
 
     private fun bitwiseAnd(left: NumericLiteral, right: NumericLiteral): NumericLiteral {
-        if(left.type== DataType.UBYTE) {
-            if(right.type in IntegerDatatypes) {
+        if(left.type==DataType.UBYTE || left.type==DataType.BOOL) {
+            if(right.type in IntegerDatatypesWithBoolean) {
                 return NumericLiteral(DataType.UBYTE, (left.number.toInt() and (right.number.toInt() and 255)).toDouble(), left.position)
             }
-        } else if(left.type== DataType.UWORD) {
+        } else if(left.type==DataType.UWORD) {
             if(right.type in IntegerDatatypes) {
                 return NumericLiteral(DataType.UWORD, (left.number.toInt() and right.number.toInt() and 65535).toDouble(), left.position)
             }

codeOptimizers/src/prog8/optimizer/ConstantFoldingOptimizer.kt

@@ -36,9 +36,9 @@ class ConstantFoldingOptimizer(private val program: Program, private val errors:
         if(parent is Assignment) {
             val iDt = parent.target.inferType(program)
             if(iDt.isKnown && !iDt.isBool && !iDt.istype(numLiteral.type)) {
-                val casted = numLiteral.cast(iDt.getOr(DataType.UNDEFINED))
+                val casted = numLiteral.cast(iDt.getOr(DataType.UNDEFINED), true)
                 if(casted.isValid) {
-                    return listOf(IAstModification.ReplaceNode(numLiteral, casted.value!!, parent))
+                    return listOf(IAstModification.ReplaceNode(numLiteral, casted.valueOrZero(), parent))
                 }
             }
         }
@@ -313,23 +313,23 @@ class ConstantFoldingOptimizer(private val program: Program, private val errors:
 
     override fun after(forLoop: ForLoop, parent: Node): Iterable<IAstModification> {
         fun adjustRangeDt(rangeFrom: NumericLiteral, targetDt: DataType, rangeTo: NumericLiteral, stepLiteral: NumericLiteral?, range: RangeExpression): RangeExpression? {
-            val fromCast = rangeFrom.cast(targetDt)
-            val toCast = rangeTo.cast(targetDt)
+            val fromCast = rangeFrom.cast(targetDt, true)
+            val toCast = rangeTo.cast(targetDt, true)
             if(!fromCast.isValid || !toCast.isValid)
                 return null
 
             val newStep =
                 if(stepLiteral!=null) {
-                    val stepCast = stepLiteral.cast(targetDt)
+                    val stepCast = stepLiteral.cast(targetDt, true)
                     if(stepCast.isValid)
-                        stepCast.value!!
+                        stepCast.valueOrZero()
                     else
                         range.step
                 } else {
                     range.step
                 }
 
-            return RangeExpression(fromCast.value!!, toCast.value!!, newStep, range.position)
+            return RangeExpression(fromCast.valueOrZero(), toCast.valueOrZero(), newStep, range.position)
         }
 
         // adjust the datatype of a range expression in for loops to the loop variable.
@@ -386,9 +386,9 @@ class ConstantFoldingOptimizer(private val program: Program, private val errors:
             val valueDt = numval.inferType(program)
             if(valueDt isnot decl.datatype) {
                 if(decl.datatype!=DataType.BOOL || valueDt.isnot(DataType.UBYTE)) {
-                    val cast = numval.cast(decl.datatype)
+                    val cast = numval.cast(decl.datatype, true)
                     if (cast.isValid)
-                        return listOf(IAstModification.ReplaceNode(numval, cast.value!!, decl))
+                        return listOf(IAstModification.ReplaceNode(numval, cast.valueOrZero(), decl))
                 }
             }
         }

codeOptimizers/src/prog8/optimizer/ConstantIdentifierReplacer.kt

@@ -35,12 +35,12 @@ class VarConstantValueTypeAdjuster(
                 // avoid silent float roundings
                 if(decl.datatype in IntegerDatatypes && declConstValue.type == DataType.FLOAT) {
                     errors.err("refused truncating of float to avoid loss of precision", decl.value!!.position)
-                } else {
-                    // cast the numeric literal to the appropriate datatype of the variable
+                } else if(decl.datatype!=DataType.BOOL) {
+                    // cast the numeric literal to the appropriate datatype of the variable if it's not boolean
                     declConstValue.linkParents(decl)
-                    val cast = declConstValue.cast(decl.datatype)
+                    val cast = declConstValue.cast(decl.datatype, true)
                     if (cast.isValid)
-                        return listOf(IAstModification.ReplaceNode(decl.value!!, cast.value!!, decl))
+                        return listOf(IAstModification.ReplaceNode(decl.value!!, cast.valueOrZero(), decl))
                 }
             }
         } catch (x: UndefinedSymbolError) {
@@ -270,7 +270,11 @@ class VarConstantValueTypeAdjuster(
 // Replace all constant identifiers with their actual value,
 // and the array var initializer values and sizes.
 // This is needed because further constant optimizations depend on those.
-internal class ConstantIdentifierReplacer(private val program: Program, private val errors: IErrorReporter, private val compTarget: ICompilationTarget) : AstWalker() {
+internal class ConstantIdentifierReplacer(
+    private val program: Program,
+    private val options: CompilationOptions,
+    private val errors: IErrorReporter
+) : AstWalker() {
 
     override fun before(addressOf: AddressOf, parent: Node): Iterable<IAstModification> {
         val constValue = addressOf.constValue(program)
@@ -292,7 +296,7 @@ internal class ConstantIdentifierReplacer(private val program: Program, private
             return noModifications
 
         val dt = identifier.inferType(program)
-        if(!dt.isKnown || !dt.isNumeric)
+        if(!dt.isKnown || !dt.isNumeric && !dt.isBool)
             return noModifications
 
         try {
@@ -313,7 +317,7 @@ internal class ConstantIdentifierReplacer(private val program: Program, private
                 }
             }
             when (cval.type) {
-                in NumericDatatypes -> {
+                in NumericDatatypesWithBoolean -> {
                     if(parent is AddressOf)
                         return noModifications      // cannot replace the identifier INSIDE the addr-of here, let's do it later.
                     return listOf(
@@ -369,7 +373,7 @@ internal class ConstantIdentifierReplacer(private val program: Program, private
                 DataType.FLOAT -> {
                     // vardecl: for scalar float vars, promote constant integer initialization values to floats
                     val litval = decl.value as? NumericLiteral
-                    if (litval!=null && litval.type in IntegerDatatypes) {
+                    if (litval!=null && litval.type in IntegerDatatypesWithBoolean) {
                         val newValue = NumericLiteral(DataType.FLOAT, litval.number, litval.position)
                         return listOf(IAstModification.ReplaceNode(decl.value!!, newValue, decl))
                     }
@@ -470,12 +474,11 @@ internal class ConstantIdentifierReplacer(private val program: Program, private
                 val numericLv = decl.value as? NumericLiteral
                 val size = decl.arraysize?.constIndex() ?: return null
                 if(rangeExpr==null && numericLv!=null) {
-                    // arraysize initializer is a single int, and we know the size.
+                    // arraysize initializer is a single int, and we know the array size.
                     val fillvalue = numericLv.number
-                    if (fillvalue < compTarget.machine.FLOAT_MAX_NEGATIVE || fillvalue > compTarget.machine.FLOAT_MAX_POSITIVE)
+                    if (fillvalue < options.compTarget.machine.FLOAT_MAX_NEGATIVE || fillvalue > options.compTarget.machine.FLOAT_MAX_POSITIVE)
                         errors.err("float value overflow", numericLv.position)
                     else {
-                        // create the array itself, filled with the fillvalue.
                         val array = Array(size) {fillvalue}.map { NumericLiteral(DataType.FLOAT, it, numericLv.position) }.toTypedArray<Expression>()
                         return ArrayLiteral(InferredTypes.InferredType.known(DataType.ARRAY_F), array, position = numericLv.position)
                     }
@@ -485,9 +488,13 @@ internal class ConstantIdentifierReplacer(private val program: Program, private
                 val numericLv = decl.value as? NumericLiteral
                 val size = decl.arraysize?.constIndex() ?: return null
                 if(numericLv!=null) {
-                    // arraysize initializer is a single int, and we know the size.
-                    val fillvalue = if(numericLv.number==0.0) 0.0 else 1.0
-                    val array = Array(size) {fillvalue}.map { NumericLiteral(DataType.UBYTE, fillvalue, numericLv.position) }.toTypedArray<Expression>()
+                    // arraysize initializer is a single value, and we know the array size.
+                    if(numericLv.type!=DataType.BOOL) {
+                        if(options.strictBool || numericLv.type !in ByteDatatypes)
+                            errors.err("initializer value is not a boolean", numericLv.position)
+                        return null
+                    }
+                    val array = Array(size) {numericLv.number}.map { NumericLiteral(DataType.BOOL, it, numericLv.position) }.toTypedArray<Expression>()
                     return ArrayLiteral(InferredTypes.InferredType.known(DataType.ARRAY_BOOL), array, position = numericLv.position)
                 }
             }

codeOptimizers/src/prog8/optimizer/ExpressionSimplifier.kt

@@ -1,13 +1,9 @@
 package prog8.optimizer
 
-import prog8.ast.IStatementContainer
-import prog8.ast.Node
-import prog8.ast.Program
+import prog8.ast.*
 import prog8.ast.base.FatalAstException
 import prog8.ast.expressions.*
-import prog8.ast.maySwapOperandOrder
 import prog8.ast.statements.AnonymousScope
-import prog8.ast.statements.Assignment
 import prog8.ast.statements.IfElse
 import prog8.ast.statements.Jump
 import prog8.ast.walk.AstWalker
@@ -19,9 +15,7 @@ import kotlin.math.pow
 
 // TODO add more peephole expression optimizations? Investigate what optimizations binaryen has?
 
-class ExpressionSimplifier(private val program: Program,
-                           private val errors: IErrorReporter,
-                           private val compTarget: ICompilationTarget) : AstWalker() {
+class ExpressionSimplifier(private val program: Program, private val options: CompilationOptions, private val errors: IErrorReporter) : AstWalker() {
     private val powersOfTwo = (1..16).map { (2.0).pow(it) }.toSet()
     private val negativePowersOfTwo = powersOfTwo.map { -it }.toSet()
 
@@ -31,9 +25,9 @@ class ExpressionSimplifier(private val program: Program,
         // try to statically convert a literal value into one of the desired type
         val literal = typecast.expression as? NumericLiteral
         if (literal != null) {
-            val newLiteral = literal.cast(typecast.type)
-            if (newLiteral.isValid && newLiteral.value!! !== literal) {
-                mods += IAstModification.ReplaceNode(typecast, newLiteral.value!!, parent)
+            val newLiteral = literal.cast(typecast.type, typecast.implicit)
+            if (newLiteral.isValid && newLiteral.valueOrZero() !== literal) {
+                mods += IAstModification.ReplaceNode(typecast, newLiteral.valueOrZero(), parent)
             }
         }
 
@@ -114,16 +108,6 @@ class ExpressionSimplifier(private val program: Program,
         if (!leftIDt.isKnown || !rightIDt.isKnown)
             throw FatalAstException("can't determine datatype of both expression operands $expr")
 
-        // ConstValue <associativeoperator> X -->  X <associativeoperator> ConstValue
-        if (leftVal != null && expr.operator in AssociativeOperators && rightVal == null && maySwapOperandOrder(expr))
-            return listOf(IAstModification.SwapOperands(expr))
-
-        // NonBinaryExpression  <associativeoperator>  BinaryExpression  -->  BinaryExpression  <associativeoperator>  NonBinaryExpression
-        if (expr.operator in AssociativeOperators && expr.left !is BinaryExpression && expr.right is BinaryExpression) {
-            if(parent !is Assignment || !(expr.left isSameAs parent.target) && maySwapOperandOrder(expr))
-                return listOf(IAstModification.SwapOperands(expr))
-        }
-
         // X + (-A)  -->  X - A
         if (expr.operator == "+" && (expr.right as? PrefixExpression)?.operator == "-") {
             return listOf(IAstModification.ReplaceNode(
@@ -255,26 +239,98 @@ class ExpressionSimplifier(private val program: Program,
             }
         }
 
-        // boolvar & 1  --> boolvar
-        // boolvar & 2  --> false
-        if(expr.operator=="&" && rightDt in IntegerDatatypes && (leftDt == DataType.BOOL || (expr.left as? TypecastExpression)?.expression?.inferType(program)?.istype(DataType.BOOL)==true)) {
-            if(rightVal?.number==1.0) {
-                return listOf(IAstModification.ReplaceNode(expr, expr.left, parent))
-            } else if(rightVal?.number!=null && (rightVal.number.toInt() and 1)==0) {
-                return listOf(IAstModification.ReplaceNode(expr, NumericLiteral.fromBoolean(false, expr.position), parent))
+        // optimize boolean constant comparisons
+        if(expr.operator=="==") {
+            if(rightDt==DataType.BOOL && leftDt==DataType.BOOL) {
+                if(rightVal?.asBooleanValue==true)
+                    return listOf(IAstModification.ReplaceNode(expr, expr.left, parent))
+                else
+                    return listOf(IAstModification.ReplaceNode(expr, PrefixExpression("not", expr.left, expr.position), parent))
+            }
+            if (rightVal?.number == 1.0) {
+                if (options.strictBool) {
+                    if (rightDt != leftDt) {
+                        val right = NumericLiteral(leftDt, rightVal.number, rightVal.position)
+                        return listOf(IAstModification.ReplaceNode(expr.right, right, expr))
+                    }
+                } else
+                    return listOf(IAstModification.ReplaceNode(expr, expr.left, parent))
+            }
+            else if (rightVal?.number == 0.0) {
+                if (options.strictBool) {
+                    if (rightDt != leftDt) {
+                        val right = NumericLiteral(leftDt, rightVal.number, rightVal.position)
+                        return listOf(IAstModification.ReplaceNode(expr.right, right, expr))
+                    }
+                }
+            }
+        }
+        if (expr.operator=="!=") {
+            if(rightDt==DataType.BOOL && leftDt==DataType.BOOL) {
+                val rightConstBool = rightVal?.asBooleanValue
+                if(rightConstBool!=null) {
+                    return if (rightConstBool)
+                        listOf(IAstModification.ReplaceNode(expr, PrefixExpression("not", expr.left, expr.position), parent))
+                    else
+                        listOf(IAstModification.ReplaceNode(expr, expr.left, parent))
+                }
+            }
+            if (rightVal?.number == 1.0) {
+                if(options.strictBool) {
+                    if(rightDt!=leftDt) {
+                        val right = NumericLiteral(leftDt, rightVal.number, rightVal.position)
+                        return listOf(IAstModification.ReplaceNode(expr.right, right, expr))
+                    }
+                }
+            }
+            else if (rightVal?.number == 0.0) {
+                if(options.strictBool) {
+                    if(rightDt!=leftDt) {
+                        val right = NumericLiteral(leftDt, rightVal.number, rightVal.position)
+                        return listOf(IAstModification.ReplaceNode(expr.right, right, expr))
+                    }
+                } else
+                    return listOf(IAstModification.ReplaceNode(expr, expr.left, parent))
             }
         }
 
-        if(leftDt==DataType.BOOL) {
-            // optimize boolean constant comparisons
-//            if(expr.operator=="==" && rightVal?.number==0.0)
-//                return listOf(IAstModification.ReplaceNode(expr, PrefixExpression("not", expr.left, expr.position), parent))
-//            if(expr.operator=="!=" && rightVal?.number==1.0)
-//                return listOf(IAstModification.ReplaceNode(expr, PrefixExpression("not", expr.left, expr.position), parent))
-            if(expr.operator=="==" && rightVal?.number==1.0)
-                return listOf(IAstModification.ReplaceNode(expr, expr.left, parent))
-            if(expr.operator=="!=" && rightVal?.number==0.0)
-                return listOf(IAstModification.ReplaceNode(expr, expr.left, parent))
+        if(expr.operator in arrayOf("and", "or", "xor")) {
+            if(leftVal!=null) {
+                val result = if(leftVal.asBooleanValue) {
+                    when(expr.operator) {
+                        "and" -> expr.right
+                        "or" -> NumericLiteral.fromBoolean(true, expr.position)
+                        "xor" -> PrefixExpression("not", expr.right, expr.position)
+                        else -> throw FatalAstException("weird op")
+                    }
+                } else {
+                    when(expr.operator) {
+                        "and" -> NumericLiteral.fromBoolean(false, expr.position)
+                        "or" -> expr.right
+                        "xor" -> expr.right
+                        else -> throw FatalAstException("weird op")
+                    }
+                }
+                return listOf(IAstModification.ReplaceNode(expr, result, parent))
+            }
+            else if(rightVal!=null) {
+                val result = if(rightVal.asBooleanValue) {
+                    when(expr.operator) {
+                        "and" -> expr.left
+                        "or" -> NumericLiteral.fromBoolean(true, expr.position)
+                        "xor" -> PrefixExpression("not", expr.left, expr.position)
+                        else -> throw FatalAstException("weird op")
+                    }
+                } else {
+                    when(expr.operator) {
+                        "and" -> NumericLiteral.fromBoolean(false, expr.position)
+                        "or" -> expr.left
+                        "xor" -> expr.left
+                        else -> throw FatalAstException("weird op")
+                    }
+                }
+                return listOf(IAstModification.ReplaceNode(expr, result, parent))
+            }
         }
 
         // simplify when a term is constant and directly determines the outcome
@@ -387,10 +443,14 @@ class ExpressionSimplifier(private val program: Program,
         return noModifications
     }
 
-
     private fun applyAbsorptionLaws(expr: BinaryExpression): Expression? {
+        // NOTE: only when the terms are not function calls!!!
+        if(expr.left is IFunctionCall || expr.right is IFunctionCall)
+            return null
         val rightB = expr.right as? BinaryExpression
         if(rightB!=null) {
+            if(rightB.left is IFunctionCall || rightB.right is IFunctionCall)
+                return null
             // absorption laws:  a or (a and b) --> a,  a and (a or b) --> a
             if(expr.operator=="or" && rightB.operator=="and") {
                 if(expr.left isSameAs rightB.left || expr.left isSameAs rightB.right) {
@@ -402,12 +462,26 @@ class ExpressionSimplifier(private val program: Program,
                     return expr.left
                 }
             }
+            else if(expr.operator=="or" && rightB.operator=="or") {
+                if(expr.left isSameAs rightB.left || expr.left isSameAs rightB.right) {
+                    // a or (a or b) -> a or b
+                    return expr.right
+                }
+            }
+            else if(expr.operator=="and" && rightB.operator=="and") {
+                if(expr.left isSameAs rightB.left || expr.left isSameAs rightB.right) {
+                    // a and (a and b) -> a and b
+                    return expr.right
+                }
+            }
         }
         val leftB = expr.left as? BinaryExpression
         if(leftB!=null) {
+            if(leftB.left is IFunctionCall || leftB.right is IFunctionCall)
+                return null
             // absorption laws:  (a and b) or a --> a,  (a or b) and a --> a
             if(expr.operator=="or" && leftB.operator=="and") {
-                if(expr.right isSameAs leftB.left || expr.left isSameAs leftB.right) {
+                if(expr.right isSameAs leftB.left || expr.right isSameAs leftB.right) {
                     return expr.right
                 }
             }
@@ -416,6 +490,18 @@ class ExpressionSimplifier(private val program: Program,
                     return expr.right
                 }
             }
+            else if(expr.operator=="or" && leftB.operator=="or") {
+                if(expr.right isSameAs leftB.left || expr.right isSameAs leftB.right) {
+                    // (a or b) or a -> a or b
+                    return expr.left
+                }
+            }
+            else if(expr.operator=="and" && leftB.operator=="and") {
+                if(expr.right isSameAs leftB.left || expr.right isSameAs leftB.right) {
+                    // (a and b) or a -> a and b
+                    return expr.left
+                }
+            }
         }
         return null
     }

codeOptimizers/src/prog8/optimizer/Extensions.kt

@@ -3,7 +3,6 @@ package prog8.optimizer
 import prog8.ast.IBuiltinFunctions
 import prog8.ast.Program
 import prog8.code.core.CompilationOptions
-import prog8.code.core.ICompilationTarget
 import prog8.code.core.IErrorReporter
 
 
@@ -13,7 +12,7 @@ fun Program.constantFold(errors: IErrorReporter, options: CompilationOptions) {
     if(errors.noErrors()) {
         valuetypefixer.applyModifications()
 
-        val replacer = ConstantIdentifierReplacer(this, errors, options.compTarget)
+        val replacer = ConstantIdentifierReplacer(this, options, errors)
         replacer.visit(this)
         if (errors.noErrors()) {
             replacer.applyModifications()
@@ -60,8 +59,8 @@ fun Program.inlineSubroutines(options: CompilationOptions): Int {
     return inliner.applyModifications()
 }
 
-fun Program.simplifyExpressions(errors: IErrorReporter, target: ICompilationTarget) : Int {
-    val opti = ExpressionSimplifier(this, errors, target)
+fun Program.simplifyExpressions(errors: IErrorReporter, options: CompilationOptions) : Int {
+    val opti = ExpressionSimplifier(this, options, errors)
     opti.visit(this)
     return opti.applyModifications()
 }

codeOptimizers/src/prog8/optimizer/Inliner.kt

@@ -87,6 +87,14 @@ class Inliner(private val program: Program, private val options: CompilationOpti
                                                 } else
                                                     false
                                             targetInline || valueInline
+                                        } else if(stmt.target.identifier!=null && stmt.isAugmentable) {
+                                            val binExpr = stmt.value as BinaryExpression
+                                            if(binExpr.operator in "+-" && binExpr.right.constValue(program)?.number==1.0) {
+                                                makeFullyScoped(stmt.target.identifier!!)
+                                                makeFullyScoped(binExpr.left as IdentifierReference)
+                                                true
+                                            } else
+                                                false
                                         } else
                                             false
                                     }
@@ -107,18 +115,6 @@ class Inliner(private val program: Program, private val options: CompilationOpti
                                         inline
                                     }
 
-                                    is PostIncrDecr -> {
-                                        if (stmt.target.identifier != null) {
-                                            makeFullyScoped(stmt.target.identifier!!)
-                                            true
-                                        } else if (stmt.target.memoryAddress?.addressExpression is NumericLiteral || stmt.target.memoryAddress?.addressExpression is IdentifierReference) {
-                                            if (stmt.target.memoryAddress?.addressExpression is IdentifierReference)
-                                                makeFullyScoped(stmt.target.memoryAddress?.addressExpression as IdentifierReference)
-                                            true
-                                        } else
-                                            false
-                                    }
-
                                     is Jump -> true
                                     else -> false
                                 }

codeOptimizers/src/prog8/optimizer/StatementOptimizer.kt

@@ -5,9 +5,10 @@ import prog8.ast.expressions.*
 import prog8.ast.statements.*
 import prog8.ast.walk.AstWalker
 import prog8.ast.walk.IAstModification
-import prog8.code.core.*
-import prog8.code.target.VMTarget
-import kotlin.math.floor
+import prog8.code.core.AssociativeOperators
+import prog8.code.core.CompilationOptions
+import prog8.code.core.DataType
+import prog8.code.core.IErrorReporter
 
 
 class StatementOptimizer(private val program: Program,
@@ -82,7 +83,7 @@ class StatementOptimizer(private val program: Program,
 
         // empty true part? switch with the else part
         if(ifElse.truepart.isEmpty() && ifElse.elsepart.isNotEmpty()) {
-            val invertedCondition = BinaryExpression(ifElse.condition, "==", NumericLiteral(DataType.UBYTE, 0.0, ifElse.condition.position), ifElse.condition.position)
+            val invertedCondition = invertCondition(ifElse.condition, program)
             val emptyscope = AnonymousScope(mutableListOf(), ifElse.elsepart.position)
             val truepart = AnonymousScope(ifElse.elsepart.statements, ifElse.truepart.position)
             return listOf(
@@ -102,25 +103,40 @@ class StatementOptimizer(private val program: Program,
             }
         }
 
-        // remove obvious dangling elses (else after a return)
-        if(ifElse.elsepart.isNotEmpty() && ifElse.truepart.statements.singleOrNull() is Return) {
-            val elsePart = AnonymousScope(ifElse.elsepart.statements, ifElse.elsepart.position)
-            return listOf(
-                IAstModification.ReplaceNode(ifElse.elsepart, AnonymousScope(mutableListOf(), ifElse.elsepart.position), ifElse),
-                IAstModification.InsertAfter(ifElse, elsePart, parent as IStatementContainer)
-            )
-        }
+        if(ifElse.elsepart.isNotEmpty()) {
+            // remove obvious dangling elses (else after a return)
+            if(ifElse.truepart.statements.singleOrNull() is Return) {
+                val elsePart = AnonymousScope(ifElse.elsepart.statements, ifElse.elsepart.position)
+                return listOf(
+                    IAstModification.ReplaceNode(ifElse.elsepart, AnonymousScope(mutableListOf(), ifElse.elsepart.position), ifElse),
+                    IAstModification.InsertAfter(ifElse, elsePart, parent as IStatementContainer)
+                )
+            }
 
-        // switch if/else around if the else is just a jump or branch
-        if(ifElse.elsepart.isNotEmpty() && ifElse.elsepart.statements.size==1) {
-            val jump = ifElse.elsepart.statements[0]
-            if(jump is Jump) {
-                val newTruePart = AnonymousScope(mutableListOf(jump), ifElse.elsepart.position)
+            // switch if/else around if the else is just a jump or branch
+            if(ifElse.elsepart.statements.size==1) {
+                val jump = ifElse.elsepart.statements[0]
+                if(jump is Jump) {
+                    val newTruePart = AnonymousScope(mutableListOf(jump), ifElse.elsepart.position)
+                    val newElsePart = AnonymousScope(ifElse.truepart.statements, ifElse.truepart.position)
+                    return listOf(
+                        IAstModification.ReplaceNode(ifElse.elsepart, newElsePart, ifElse),
+                        IAstModification.ReplaceNode(ifElse.truepart, newTruePart, ifElse),
+                        IAstModification.ReplaceNode(ifElse.condition, invertCondition(ifElse.condition, program), ifElse)
+                    )
+                }
+            }
+
+            // switch if/else around if the condition is a not
+            val prefixCond = ifElse.condition as? PrefixExpression
+            if(prefixCond?.operator=="not") {
+                errors.info("invert conditon and swap if/else blocks", ifElse.condition.position)
+                val newTruePart = AnonymousScope(ifElse.elsepart.statements, ifElse.elsepart.position)
                 val newElsePart = AnonymousScope(ifElse.truepart.statements, ifElse.truepart.position)
                 return listOf(
                     IAstModification.ReplaceNode(ifElse.elsepart, newElsePart, ifElse),
                     IAstModification.ReplaceNode(ifElse.truepart, newTruePart, ifElse),
-                    IAstModification.ReplaceNode(ifElse.condition, invertCondition(ifElse.condition, program), ifElse)
+                    IAstModification.ReplaceNode(ifElse.condition, prefixCond.expression, ifElse)
                 )
             }
         }
@@ -185,47 +201,12 @@ class StatementOptimizer(private val program: Program,
 
         val loopvarDt = forLoop.loopVarDt(program)
         if(loopvarDt.istype(DataType.UWORD) || loopvarDt.istype(DataType.UBYTE)) {
-            if (range != null && range.from.constValue(program)?.number == 0.0 && range.step.constValue(program)?.number==1.0) {
-                val toBinExpr = range.to as? BinaryExpression
-                if(toBinExpr!=null && toBinExpr.operator=="-" && toBinExpr.right.constValue(program)?.number==1.0) {
-                    // FOR var IN 0 TO X-1 .... ---> var=0, DO {... , var++} UNTIL var==X
-                    val pos = forLoop.position
-                    val condition = BinaryExpression(forLoop.loopVar.copy(), "==", toBinExpr.left, pos)
-                    val incOne = PostIncrDecr(AssignTarget(forLoop.loopVar.copy(), null, null, pos), "++", pos)
-                    forLoop.body.statements.add(incOne)
-                    val replacement = AnonymousScope(mutableListOf(
-                        Assignment(AssignTarget(forLoop.loopVar.copy(), null, null, pos),
-                            NumericLiteral.optimalNumeric(0.0, pos),
-                            AssignmentOrigin.OPTIMIZER, pos),
-                        UntilLoop(forLoop.body, condition, pos)
-                    ), pos)
-                    return listOf(IAstModification.ReplaceNode(forLoop, replacement, parent))
-                }
 
-                if(options.compTarget.name!=VMTarget.NAME) {
-                    // this optimization is not effective for the VM target.
-                    val toConst = range.to.constValue(program)
-                    if (toConst == null) {
-                        // FOR var in 0 TO X ... --->  var=0, REPEAT { ... , IF var==X break , var++ }
-                        val pos = forLoop.position
-                        val incOne = PostIncrDecr(AssignTarget(forLoop.loopVar.copy(), null, null, pos), "++", pos)
-                        val breakCondition = IfElse(
-                            BinaryExpression(forLoop.loopVar, "==", range.to, pos),
-                            AnonymousScope(mutableListOf(Break(pos)), pos),
-                            AnonymousScope(mutableListOf(), pos),
-                            pos
-                        )
-                        forLoop.body.statements.add(breakCondition)
-                        forLoop.body.statements.add(incOne)
-                        val replacement = AnonymousScope(mutableListOf(
-                            Assignment(AssignTarget(forLoop.loopVar.copy(), null, null, pos),
-                                NumericLiteral.optimalNumeric(0.0, pos),
-                                AssignmentOrigin.OPTIMIZER, pos),
-                            RepeatLoop(null, forLoop.body, pos)
-                        ), pos)
-                        return listOf(IAstModification.ReplaceNode(forLoop, replacement, parent))
-                    }
-                }
+            fun incOrDec(inc: Boolean): Assignment {
+                val pos = forLoop.position
+                val loopVar = forLoop.loopVar
+                val addSubOne = BinaryExpression(loopVar.copy(), if(inc) "+" else "-", NumericLiteral.optimalInteger(1, pos), pos, false)
+                return Assignment(AssignTarget(loopVar.copy(), null, null, pos), addSubOne, AssignmentOrigin.USERCODE, pos)
             }
 
             if (range != null && range.to.constValue(program)?.number == 0.0 && range.step.constValue(program)?.number==-1.0) {
@@ -235,7 +216,7 @@ class StatementOptimizer(private val program: Program,
                     val pos = forLoop.position
                     val checkValue = NumericLiteral(loopvarDt.getOr(DataType.UNDEFINED), if(loopvarDt.istype(DataType.UBYTE)) 255.0 else 65535.0, pos)
                     val condition = BinaryExpression(forLoop.loopVar.copy(), "==", checkValue, pos)
-                    val decOne = PostIncrDecr(AssignTarget(forLoop.loopVar.copy(), null, null, pos), "--", pos)
+                    val decOne = incOrDec(false)
                     forLoop.body.statements.add(decOne)
                     val replacement = AnonymousScope(mutableListOf(
                         Assignment(AssignTarget(forLoop.loopVar.copy(), null, null, pos),
@@ -390,35 +371,15 @@ class StatementOptimizer(private val program: Program,
                 // assignments of the form:  X = X <operator> <expr>
                 // remove assignments that have no effect (such as X=X+0)
                 // optimize/rewrite some other expressions
-                val targetDt = targetIDt.getOr(DataType.UNDEFINED)
-                val vardeclDt = (assignment.target.identifier?.targetVarDecl(program))?.type
                 when (bexpr.operator) {
                     "+" -> {
                         if (rightCv == 0.0) {
                             return listOf(IAstModification.Remove(assignment, parent as IStatementContainer))
-                        } else if (targetDt in IntegerDatatypes && floor(rightCv) == rightCv) {
-                            if (vardeclDt != VarDeclType.MEMORY && rightCv in 1.0..3.0 && options.compTarget.name!=VMTarget.NAME) {
-                                // replace by several INCs if it's not a memory address (inc on a memory mapped register doesn't work very well)
-                                val incs = AnonymousScope(mutableListOf(), assignment.position)
-                                repeat(rightCv.toInt()) {
-                                    incs.statements.add(PostIncrDecr(assignment.target.copy(), "++", assignment.position))
-                                }
-                                return listOf(IAstModification.ReplaceNode(assignment, if(incs.statements.size==1) incs.statements[0] else incs, parent))
-                            }
                         }
                     }
                     "-" -> {
                         if (rightCv == 0.0) {
                             return listOf(IAstModification.Remove(assignment, parent as IStatementContainer))
-                        } else if (targetDt in IntegerDatatypes && floor(rightCv) == rightCv) {
-                            if (vardeclDt != VarDeclType.MEMORY && rightCv in 1.0..3.0 && options.compTarget.name!=VMTarget.NAME) {
-                                // replace by several DECs if it's not a memory address (dec on a memory mapped register doesn't work very well)
-                                val decs = AnonymousScope(mutableListOf(), assignment.position)
-                                repeat(rightCv.toInt()) {
-                                    decs.statements.add(PostIncrDecr(assignment.target.copy(), "--", assignment.position))
-                                }
-                                return listOf(IAstModification.ReplaceNode(assignment, decs, parent))
-                            }
                         }
                     }
                     "*" -> if (rightCv == 1.0) return listOf(IAstModification.Remove(assignment, parent as IStatementContainer))
@@ -453,6 +414,33 @@ class StatementOptimizer(private val program: Program,
             }
         }
 
+        // xx+=2 -> xx++ xx++
+        // note: ideally this optimization should be done by the code generator, but doing it there
+        // requires doing it multiple times (because lots of different things can be incremented/decremented)
+        if(assignment.target.identifier!=null
+            || assignment.target.arrayindexed?.isSimple==true
+            || assignment.target.memoryAddress?.addressExpression?.isSimple==true) {
+            if(assignment.value.inferType(program).isBytes && assignment.isAugmentable) {
+                val binExpr = assignment.value as? BinaryExpression
+                if(binExpr!=null) {
+                    if(binExpr.operator in "+-") {
+                        val value = binExpr.right.constValue(program)?.number?.toInt()
+                        if(value!=null && value==2) {
+                            val stmts = mutableListOf<Statement>()
+                            repeat(value) {
+                                val incrdecr = Assignment(assignment.target.copy(),
+                                    BinaryExpression(assignment.target.toExpression(), binExpr.operator, NumericLiteral.optimalInteger(1, assignment.position), assignment.position),
+                                    AssignmentOrigin.OPTIMIZER, assignment.position)
+                                stmts.add(incrdecr)
+                            }
+                            val incrdecrs = AnonymousScope(stmts, assignment.position)
+                            return listOf(IAstModification.ReplaceNode(assignment, incrdecrs, parent))
+                        }
+                    }
+                }
+            }
+        }
+
         return noModifications
     }
 
@@ -472,38 +460,6 @@ class StatementOptimizer(private val program: Program,
             return listOf(IAstModification.ReplaceNode(whenStmt, ifStmt, parent))
         }
 
-        if(whenStmt.condition.inferType(program).isBool) {
-            if(whenStmt.choices.all { it.values?.size==1 }) {
-                if (whenStmt.choices.all { it.values!!.single().constValue(program)!!.number in arrayOf(0.0, 1.0) }) {
-                    // it's a when statement on booleans that can just be replaced by an if or if-else.
-                    if (whenStmt.choices.size == 1) {
-                        return if(whenStmt.choices[0].values!![0].constValue(program)!!.number==1.0) {
-                            replaceWithIf(whenStmt.condition, whenStmt.choices[0].statements, null)
-                        } else {
-                            val notCondition = BinaryExpression(whenStmt.condition, "==", NumericLiteral(DataType.UBYTE, 0.0, whenStmt.condition.position), whenStmt.condition.position)
-                            replaceWithIf(notCondition, whenStmt.choices[0].statements, null)
-                        }
-                    } else if (whenStmt.choices.size == 2) {
-                        var trueBlock: AnonymousScope? = null
-                        var elseBlock: AnonymousScope? = null
-                        if(whenStmt.choices[0].values!![0].constValue(program)!!.number==1.0) {
-                            trueBlock = whenStmt.choices[0].statements
-                        } else {
-                            elseBlock = whenStmt.choices[0].statements
-                        }
-                        if(whenStmt.choices[1].values!![0].constValue(program)!!.number==1.0) {
-                            trueBlock = whenStmt.choices[1].statements
-                        } else {
-                            elseBlock = whenStmt.choices[1].statements
-                        }
-                        if(trueBlock!=null && elseBlock!=null) {
-                            return replaceWithIf(whenStmt.condition, trueBlock, elseBlock)
-                        }
-                    }
-                }
-            }
-        }
-
         val constantValue = whenStmt.condition.constValue(program)?.number
         if(constantValue!=null) {
             // when condition is a constant

codeOptimizers/src/prog8/optimizer/UnusedCodeRemover.kt

@@ -20,7 +20,12 @@ class UnusedCodeRemover(private val program: Program,
                         private val compTarget: ICompilationTarget
 ): AstWalker() {
 
-    private val callgraph = CallGraph(program)
+    private lateinit var callgraph: CallGraph
+
+    override fun before(program: Program): Iterable<IAstModification> {
+        callgraph = CallGraph(program)
+        return noModifications
+    }
 
     override fun before(module: Module, parent: Node): Iterable<IAstModification> {
         return if (!module.isLibrary && (module.containsNoCodeNorVars || callgraph.unused(module)))

compiler/build.gradle

@@ -2,8 +2,7 @@ plugins {
     id 'java'
     id 'application'
     id 'org.jetbrains.kotlin.jvm'
-    id 'com.github.johnrengelman.shadow' version '7.1.2'
-    id 'io.kotest' version '0.3.9'
+    id 'com.github.johnrengelman.shadow' version '8.1.1'
     id 'com.peterabeles.gversion' version '1.10.2'
 }
 
@@ -37,10 +36,14 @@ dependencies {
     implementation "org.jetbrains.kotlin:kotlin-stdlib-jdk8"
     // implementation "org.jetbrains.kotlin:kotlin-reflect"
     implementation 'org.jetbrains.kotlinx:kotlinx-cli:0.3.6'
-    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.18"
+    implementation "com.michael-bull.kotlin-result:kotlin-result-jvm:1.1.20"
 
+    testImplementation project(':codeCore')
     testImplementation project(':intermediate')
     testImplementation 'io.kotest:kotest-runner-junit5-jvm:5.8.0'
+    testImplementation 'org.junit.jupiter:junit-jupiter:5.9.2'
+    testRuntimeOnly 'org.junit.platform:junit-platform-launcher'
+
 }
 
 configurations.all {
@@ -102,7 +105,7 @@ test {
 gversion {
     srcDir       = "src/"           // path is relative to the sub-project by default
     classPackage = "prog8.buildversion"
-    className    = "BuildVersion"
+    className    = "Version"
     language     = "kotlin"
 }
 

compiler/res/prog8lib/atari/syslib.p8

@@ -99,20 +99,14 @@ sys {
             rts     ; nothing to copy
 
 _copyshort
-            ; decrease source and target pointers so we can simply index by Y
-            lda  P8ZP_SCRATCH_W1
-            bne  +
-            dec  P8ZP_SCRATCH_W1+1
-+           dec  P8ZP_SCRATCH_W1
-            lda  P8ZP_SCRATCH_W2
-            bne  +
-            dec  P8ZP_SCRATCH_W2+1
-+           dec  P8ZP_SCRATCH_W2
-
+            dey
+            beq  +
 -           lda  (P8ZP_SCRATCH_W1),y
             sta  (P8ZP_SCRATCH_W2),y
             dey
             bne  -
++           lda  (P8ZP_SCRATCH_W1),y
+            sta  (P8ZP_SCRATCH_W2),y
             rts
 
 _longcopy
@@ -262,6 +256,22 @@ save_SCRATCH_ZPWORD2	.word  0
         }}
     }
 
+    asmsub exit2(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y) {
+        ; -- immediately exit the program with result values in the A, X and Y registers.
+        ;    TODO where to store A,X,Y as exit code?
+        %asm {{
+            jmp  exit
+        }}
+    }
+
+    asmsub exit3(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y, bool carry @Pc) {
+        ; -- immediately exit the program with result values in the A, X and Y registers, and the Carry flag in the status register.
+        ;    TODO where to store A,X,Y,Carry as exit code?
+        %asm {{
+            jmp  exit
+        }}
+    }
+
     inline asmsub progend() -> uword @AY {
         %asm {{
             lda  #<prog8_program_end

compiler/res/prog8lib/atari/textio.p8

@@ -3,7 +3,7 @@
 
 %import syslib
 %import conv
-
+%import shared_textio_functions
 
 txt {
 
@@ -14,15 +14,19 @@ const ubyte DEFAULT_HEIGHT = 24
 
 
 sub  clear_screen() {
-    txt.chrout(125)
+    chrout(125)
 }
 
 sub nl() {
-    txt.chrout('\n')
+    chrout('\n')
 }
 
 sub spc() {
-    txt.chrout(' ')
+    chrout(' ')
+}
+
+sub bell() {
+    chrout($fd)
 }
 
 sub column(ubyte col) {

compiler/res/prog8lib/c128/syslib.p8

@@ -98,14 +98,14 @@ romsub $FFE4 = GETIN() clobbers(X,Y) -> bool @Pc, ubyte @ A    ; (via 810 ($32A)
 romsub $FFE7 = CLALL() clobbers(A,X)                            ; (via 812 ($32C)) close all files
 romsub $FFEA = UDTIM() clobbers(A,X)                            ; update the software clock
 romsub $FFED = SCREEN() -> ubyte @ X, ubyte @ Y                 ; read number of screen rows and columns
-romsub $FFF0 = PLOT(ubyte col @ Y, ubyte row @ X, bool dir @ Pc) -> ubyte @ X, ubyte @ Y       ; read/set position of cursor on screen.  Use txt.plot for a 'safe' wrapper that preserves X.
+romsub $FFF0 = PLOT(ubyte col @ Y, ubyte row @ X, bool dir @ Pc) clobbers(A) -> ubyte @ X, ubyte @ Y       ; read/set position of cursor on screen.  Use txt.plot for a 'safe' wrapper that preserves X.
 romsub $FFF3 = IOBASE() -> uword @ XY                           ; read base address of I/O devices
 
 ; ---- end of C64 compatible ROM kernal routines ----
 
 ; ---- utilities -----
 
-asmsub STOP2() clobbers(X) -> ubyte @A  {
+asmsub STOP2() clobbers(X) -> bool @A  {
     ; -- check if STOP key was pressed, returns true if so.  More convenient to use than STOP() because that only sets the carry status flag.
     %asm {{
         jsr  cbm.STOP
@@ -138,6 +138,16 @@ sub CLEARST() {
     CLOSE(15)
 }
 
+asmsub kbdbuf_clear() {
+    ; -- convenience helper routine to clear the keyboard buffer
+    %asm {{
+-       jsr  GETIN
+        cmp  #0
+        bne  -
+        rts
+    }}
+}
+
 }
 
 c64 {
@@ -377,8 +387,15 @@ asmsub  cleanup_at_exit() {
         sta  $ff00          ; default bank 15
         jsr  cbm.CLRCHN		; reset i/o channels
         jsr  enable_runstop_and_charsetswitch
+_exitcodeCarry = *+1
+        lda  #0
+        lsr  a
 _exitcode = *+1
         lda  #0        ; exit code possibly modified in exit()
+_exitcodeX = *+1
+        ldx  #0
+_exitcodeY = *+1
+        ldy  #0
         rts
     }}
 }
@@ -641,20 +658,14 @@ _loop       lda  P8ZP_SCRATCH_W1
             rts     ; nothing to copy
 
 _copyshort
-            ; decrease source and target pointers so we can simply index by Y
-            lda  P8ZP_SCRATCH_W1
-            bne  +
-            dec  P8ZP_SCRATCH_W1+1
-+           dec  P8ZP_SCRATCH_W1
-            lda  P8ZP_SCRATCH_W2
-            bne  +
-            dec  P8ZP_SCRATCH_W2+1
-+           dec  P8ZP_SCRATCH_W2
-
+            dey
+            beq  +
 -           lda  (P8ZP_SCRATCH_W1),y
             sta  (P8ZP_SCRATCH_W2),y
             dey
             bne  -
++           lda  (P8ZP_SCRATCH_W1),y
+            sta  (P8ZP_SCRATCH_W2),y
             rts
 
 _longcopy
@@ -770,6 +781,33 @@ _longcopy
         }}
     }
 
+    asmsub exit2(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y) {
+        ; -- immediately exit the program with result values in the A, X and Y registers.
+        %asm {{
+            sta  cleanup_at_exit._exitcode
+            stx  cleanup_at_exit._exitcodeX
+            sty  cleanup_at_exit._exitcodeY
+            ldx  prog8_lib.orig_stackpointer
+            txs
+            jmp  cleanup_at_exit
+        }}
+    }
+
+    asmsub exit3(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y, bool carry @Pc) {
+        ; -- immediately exit the program with result values in the A, X and Y registers, and the Carry flag in the status register.
+        %asm {{
+            sta  cleanup_at_exit._exitcode
+            lda  #0
+            rol  a
+            sta  cleanup_at_exit._exitcodeCarry
+            stx  cleanup_at_exit._exitcodeX
+            sty  cleanup_at_exit._exitcodeY
+            ldx  prog8_lib.orig_stackpointer
+            txs
+            jmp  cleanup_at_exit
+        }}
+    }
+
     inline asmsub progend() -> uword @AY {
         %asm {{
             lda  #<prog8_program_end

compiler/res/prog8lib/c128/textio.p8

@@ -4,6 +4,7 @@
 
 %import syslib
 %import conv
+%import shared_textio_functions
 
 
 txt {
@@ -14,26 +15,32 @@ const ubyte DEFAULT_HEIGHT = 25
 
 
 sub  clear_screen() {
-    txt.chrout(147)
+    chrout(147)
 }
 
 sub home() {
-    txt.chrout(19)
+    chrout(19)
 }
 
 sub nl() {
-    txt.chrout('\n')
+    chrout('\n')
 }
 
 sub spc() {
-    txt.chrout(' ')
+    chrout(' ')
+}
+
+sub bell() {
+    chrout(7)
 }
 
 asmsub column(ubyte col @A) clobbers(A, X, Y) {
     ; ---- set the cursor on the given column (starting with 0) on the current line
     %asm {{
+        pha
         sec
         jsr  cbm.PLOT
+        pla
         tay
         clc
         jmp  cbm.PLOT
@@ -51,8 +58,10 @@ asmsub get_column() -> ubyte @Y {
 asmsub row(ubyte rownum @A) clobbers(A, X, Y) {
     ; ---- set the cursor on the given row (starting with 0) on the current line
     %asm {{
+        pha
         sec
         jsr  cbm.PLOT
+        pla
         tax
         clc
         jmp  cbm.PLOT

compiler/res/prog8lib/c64/floats.asm

@@ -98,6 +98,21 @@ cast_as_w_into_ay	.proc               ; also used for float 2 b
 		jmp  cast_FAC1_as_w_into_ay
 		.pend
 
+cast_as_bool_into_a	.proc
+		; -- cast float at A/Y to bool into A
+		;    clobbers X
+		jsr  MOVFM
+		jmp  cast_FAC1_as_bool_into_a
+		.pend
+
+cast_FAC1_as_bool_into_a	.proc
+		; -- cast fac1 to bool into A
+		;    clobbers X
+		jsr  SIGN
+		and  #1
+		rts
+		.pend
+
 cast_FAC1_as_uw_into_ya	.proc               ; also used for float 2 ub
 		; -- cast fac1 to uword into Y/A
 		;    clobbers X
@@ -221,7 +236,7 @@ var_fac1_greatereq_f	.proc
 		.pend
 
 var_fac1_equal_f	.proc
-		; -- are the floats numbers in FAC1 and the variable AY *not* identical?   Result in A. Clobbers X.
+		; -- are the floats numbers in FAC1 and the variable AY identical?   Result in A. Clobbers X.
 		jsr  FCOMP
 		and  #1
 		eor  #1

compiler/res/prog8lib/c64/floats.p8

@@ -1,7 +1,7 @@
 ; Prog8 definitions for floating point handling on the Commodore-64
 
 %option enable_floats, no_symbol_prefixing, ignore_unused
-%import floats_functions
+%import shared_floats_functions
 
 floats {
 	; ---- this block contains C-64 floating point related functions ----

compiler/res/prog8lib/c64/graphics.p8

@@ -73,7 +73,7 @@ graphics {
         }
 
         word @zp d = 0
-        ubyte positive_ix = true
+        bool positive_ix = true
         if dx < 0 {
             dx = -dx
             positive_ix = false
@@ -173,7 +173,7 @@ graphics {
         ubyte separate_pixels = lsb(xx) & 7
         uword pixaddr = get_y_lookup(yy) + (xx&$fff8)
 
-        if separate_pixels {
+        if separate_pixels!=0 {
             %asm {{
                 lda  p8v_pixaddr
                 sta  P8ZP_SCRATCH_W1
@@ -191,7 +191,7 @@ graphics {
             length -= 8
         }
 
-        if length {
+        if length!=0 {
             %asm {{
                 lda  p8v_length
                 and  #7

compiler/res/prog8lib/c64/syslib.p8

@@ -99,10 +99,10 @@ romsub $FFE4 = GETIN() clobbers(X,Y) -> bool @Pc, ubyte @ A    ; (via 810 ($32A)
 romsub $FFE7 = CLALL() clobbers(A,X)                            ; (via 812 ($32C)) close all files
 romsub $FFEA = UDTIM() clobbers(A,X)                            ; update the software clock
 romsub $FFED = SCREEN() -> ubyte @ X, ubyte @ Y                 ; read number of screen rows and columns
-romsub $FFF0 = PLOT(ubyte col @ Y, ubyte row @ X, bool dir @ Pc) -> ubyte @ X, ubyte @ Y       ; read/set position of cursor on screen.  Use txt.plot for a 'safe' wrapper that preserves X.
+romsub $FFF0 = PLOT(ubyte col @ Y, ubyte row @ X, bool dir @ Pc) clobbers(A) -> ubyte @ X, ubyte @ Y       ; read/set position of cursor on screen.  Use txt.plot for a 'safe' wrapper that preserves X.
 romsub $FFF3 = IOBASE() -> uword @ XY                           ; read base address of I/O devices
 
-asmsub STOP2() clobbers(X) -> ubyte @A  {
+asmsub STOP2() clobbers(X) -> bool @A  {
     ; -- check if STOP key was pressed, returns true if so.  More convenient to use than STOP() because that only sets the carry status flag.
     %asm {{
         jsr  cbm.STOP
@@ -135,6 +135,16 @@ sub CLEARST() {
     CLOSE(15)
 }
 
+asmsub kbdbuf_clear() {
+    ; -- convenience helper routine to clear the keyboard buffer
+    %asm {{
+-       jsr  GETIN
+        cmp  #0
+        bne  -
+        rts
+    }}
+}
+
 }
 
 c64 {
@@ -376,8 +386,15 @@ asmsub  cleanup_at_exit() {
         sta  $01        ; bank the kernal in
         jsr  cbm.CLRCHN		; reset i/o channels
         jsr  enable_runstop_and_charsetswitch
+_exitcodeCarry = *+1
+        lda  #0
+        lsr  a
 _exitcode = *+1
         lda  #0        ; exit code possibly modified in exit()
+_exitcodeX = *+1
+        ldx  #0
+_exitcodeY = *+1
+        ldy  #0
         rts
     }}
 }
@@ -639,20 +656,14 @@ _loop       lda  P8ZP_SCRATCH_W1
             rts     ; nothing to copy
 
 _copyshort
-            ; decrease source and target pointers so we can simply index by Y
-            lda  P8ZP_SCRATCH_W1
-            bne  +
-            dec  P8ZP_SCRATCH_W1+1
-+           dec  P8ZP_SCRATCH_W1
-            lda  P8ZP_SCRATCH_W2
-            bne  +
-            dec  P8ZP_SCRATCH_W2+1
-+           dec  P8ZP_SCRATCH_W2
-
+            dey
+            beq  +
 -           lda  (P8ZP_SCRATCH_W1),y
             sta  (P8ZP_SCRATCH_W2),y
             dey
             bne  -
++           lda  (P8ZP_SCRATCH_W1),y
+            sta  (P8ZP_SCRATCH_W2),y
             rts
 
 _longcopy
@@ -768,6 +779,33 @@ _longcopy
         }}
     }
 
+    asmsub exit2(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y) {
+        ; -- immediately exit the program with result values in the A, X and Y registers.
+        %asm {{
+            sta  cleanup_at_exit._exitcode
+            stx  cleanup_at_exit._exitcodeX
+            sty  cleanup_at_exit._exitcodeY
+            ldx  prog8_lib.orig_stackpointer
+            txs
+            jmp  cleanup_at_exit
+        }}
+    }
+
+    asmsub exit3(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y, bool carry @Pc) {
+        ; -- immediately exit the program with result values in the A, X and Y registers, and the Carry flag in the status register.
+        %asm {{
+            sta  cleanup_at_exit._exitcode
+            lda  #0
+            rol  a
+            sta  cleanup_at_exit._exitcodeCarry
+            stx  cleanup_at_exit._exitcodeX
+            sty  cleanup_at_exit._exitcodeY
+            ldx  prog8_lib.orig_stackpointer
+            txs
+            jmp  cleanup_at_exit
+        }}
+    }
+
     inline asmsub progend() -> uword @AY {
         %asm {{
             lda  #<prog8_program_end

compiler/res/prog8lib/c64/textio.p8

@@ -4,7 +4,7 @@
 
 %import syslib
 %import conv
-
+%import shared_textio_functions
 
 txt {
 
@@ -15,26 +15,38 @@ const ubyte DEFAULT_HEIGHT = 25
 
 
 sub  clear_screen() {
-    txt.chrout(147)
+    chrout(147)
 }
 
 sub home() {
-    txt.chrout(19)
+    chrout(19)
 }
 
 sub nl() {
-    txt.chrout('\n')
+    chrout('\n')
 }
 
 sub spc() {
-    txt.chrout(' ')
+    chrout(' ')
+}
+
+sub bell() {
+    ; beep
+    c64.MVOL = 11
+    c64.AD1 = %00110111
+    c64.SR1 = %00000000
+    c64.FREQ1 = 8500
+    c64.CR1 = %00010000
+    c64.CR1 = %00010001
 }
 
 asmsub column(ubyte col @A) clobbers(A, X, Y) {
     ; ---- set the cursor on the given column (starting with 0) on the current line
     %asm {{
+        pha
         sec
         jsr  cbm.PLOT
+        pla
         tay
         clc
         jmp  cbm.PLOT
@@ -52,8 +64,10 @@ asmsub get_column() -> ubyte @Y {
 asmsub row(ubyte rownum @A) clobbers(A, X, Y) {
     ; ---- set the cursor on the given row (starting with 0) on the current line
     %asm {{
+        pha
         sec
         jsr  cbm.PLOT
+        pla
         tax
         clc
         jmp  cbm.PLOT

compiler/res/prog8lib/conv.p8

@@ -8,81 +8,108 @@ conv {
 
     str  @shared string_out = "????????????????"       ; result buffer for the string conversion routines
 
-asmsub  str_ub0  (ubyte value @ A) clobbers(A,X,Y)  {
+asmsub  str_ub0  (ubyte value @ A) clobbers(X) -> str @AY  {
 	; ---- convert the ubyte in A in decimal string form, with left padding 0s (3 positions total)
 	%asm {{
-            jsr  conv.ubyte2decimal
-            sty  string_out
-            sta  string_out+1
-            stx  string_out+2
-            lda  #0
-            sta  string_out+3
-            rts
+        jsr  conv.ubyte2decimal
+        sty  string_out
+        sta  string_out+1
+        stx  string_out+2
+        lda  #0
+        sta  string_out+3
+        lda  #<string_out
+        ldy  #>string_out
+        rts
 	}}
 }
 
-asmsub  str_ub  (ubyte value @ A) clobbers(A,X,Y)  {
+asmsub  str_ub  (ubyte value @ A) clobbers(X) -> str @AY  {
 	; ---- convert the ubyte in A in decimal string form, without left padding 0s
 	%asm {{
 		ldy  #0
 		sty  P8ZP_SCRATCH_B1
-		jsr  conv.ubyte2decimal
-_output_byte_digits
-                ; hundreds?
+		jsr  ubyte2decimal     ; result in Y/A/X (100s, 10s, 1s).
+        ; hundreds?
 		cpy  #'0'
 		beq  +
-		pha
-		tya
-		ldy  P8ZP_SCRATCH_B1
-		sta  string_out,y
-		pla
-		inc  P8ZP_SCRATCH_B1
+		sty  string_out
+		sta  string_out+1
+		stx  string_out+2
+		lda  #0
+		sta  string_out+3
+		jmp  _done
 		; tens?
-+		ldy  P8ZP_SCRATCH_B1
-                cmp  #'0'
++		cmp  #'0'
 		beq  +
-		sta  string_out,y
-		iny
-+               ; ones.
-                txa
-                sta  string_out,y
-                iny
-                lda  #0
-                sta  string_out,y
-                rts
+		sta  string_out
+		stx  string_out+1
+		lda  #0
+		sta  string_out+2
+		jmp  _done
++       ; ones.
+        stx  string_out
+		lda  #0
+		sta  string_out+1
+_done   lda  #<string_out
+        ldy  #>string_out
+        rts
 	}}
 }
 
-asmsub  str_b  (byte value @ A) clobbers(A,X,Y)  {
+asmsub  str_b  (byte value @ A) clobbers(X) -> str @AY  {
 	; ---- convert the byte in A in decimal string form, without left padding 0s
 	%asm {{
-            ldy  #0
-            sty  P8ZP_SCRATCH_B1
-            cmp  #0
-            bpl  +
-            pha
-            lda  #'-'
-            sta  string_out
-            inc  P8ZP_SCRATCH_B1
-            pla
-+	    jsr  conv.byte2decimal
-            bra  str_ub._output_byte_digits
+	    ldy  #0
+        cmp  #0
+        bpl  +
+        ldy  #'-'
+        sty  string_out
+        ldy  #1
++       sty  P8ZP_SCRATCH_REG
+   	    jsr  conv.byte2decimal      ; result in Y/A/X (100s, 10s, 1s).  and in uword2decimal.decHundreds, decTens, decOnes.
+        ; hundreds?
+		cpy  #'0'
+		bne  _out_hundreds
+		ldy  P8ZP_SCRATCH_REG
+        cmp  #'0'
+        bne  _out_tens
+        beq  _out_ones
+_out_hundreds
+		ldy  P8ZP_SCRATCH_REG
+		lda  uword2decimal.decHundreds
+		sta  string_out,y
+		iny
+_out_tens
+		lda  uword2decimal.decTens
+		sta  string_out,y
+		iny
+_out_ones
+		lda  uword2decimal.decOnes
+        sta  string_out,y
+        iny
+        lda  #0
+        sta  string_out,y
+        lda  #<string_out
+        ldy  #>string_out
+   	    rts
 	}}
 }
 
-asmsub  str_ubhex  (ubyte value @ A) clobbers(A,X,Y)  {
+asmsub  str_ubhex  (ubyte value @ A) clobbers(X) -> str @AY {
 	; ---- convert the ubyte in A in hex string form
 	%asm {{
-            jsr  conv.ubyte2hex
-            sta  string_out
-            sty  string_out+1
-            lda  #0
-            sta  string_out+2
-            rts
+        jsr  conv.ubyte2hex
+        sta  string_out
+        sty  string_out+1
+        lda  #0
+        sta  string_out+2
+        lda  #<string_out
+        ldy  #>string_out
+        rts
 	}}
 }
 
-asmsub  str_ubbin  (ubyte value @ A) clobbers(A,X,Y)  {
+asmsub  str_ubbin  (ubyte value @ A) clobbers(X) -> str @AY  {
 	; ---- convert the ubyte in A in binary string form
 	%asm {{
 	    sta  P8ZP_SCRATCH_B1
@@ -90,18 +117,20 @@ asmsub  str_ubbin  (ubyte value @ A) clobbers(A,X,Y)  {
 	    sty  string_out+8
 	    ldy  #7
 -	    lsr  P8ZP_SCRATCH_B1
-            bcc  +
-            lda  #'1'
-            bne  _digit
-+           lda  #'0'
-_digit      sta  string_out,y
-            dey
+        bcc  +
+        lda  #'1'
+        bne  _digit
++       lda  #'0'
+_digit  sta  string_out,y
+        dey
 	    bpl  -
+        lda  #<string_out
+        ldy  #>string_out
 	    rts
 	}}
 }
 
-asmsub  str_uwbin  (uword value @ AY) clobbers(A,X,Y)  {
+asmsub  str_uwbin  (uword value @ AY) clobbers(X) -> str @AY  {
 	; ---- convert the uword in A/Y in binary string form
 	%asm {{
 	    sta  P8ZP_SCRATCH_REG
@@ -111,79 +140,88 @@ asmsub  str_uwbin  (uword value @ AY) clobbers(A,X,Y)  {
 	    sty  string_out+16
 	    ldy  #7
 -	    lsr  P8ZP_SCRATCH_REG
-            bcc  +
-            lda  #'1'
-            bne  _digit
-+           lda  #'0'
-_digit      sta  string_out+8,y
-            dey
+        bcc  +
+        lda  #'1'
+        bne  _digit
++       lda  #'0'
+_digit  sta  string_out+8,y
+        dey
 	    bpl  -
+        lda  #<string_out
+        ldy  #>string_out
 	    rts
 	}}
 }
 
-asmsub  str_uwhex  (uword value @ AY) clobbers(A,Y)  {
+asmsub  str_uwhex  (uword value @ AY) -> str @AY  {
 	; ---- convert the uword in A/Y in hexadecimal string form (4 digits)
 	%asm {{
-            pha
-            tya
-            jsr  conv.ubyte2hex
-            sta  string_out
-            sty  string_out+1
-            pla
-            jsr  conv.ubyte2hex
-            sta  string_out+2
-            sty  string_out+3
-            lda  #0
-            sta  string_out+4
-            rts
+        pha
+        tya
+        jsr  conv.ubyte2hex
+        sta  string_out
+        sty  string_out+1
+        pla
+        jsr  conv.ubyte2hex
+        sta  string_out+2
+        sty  string_out+3
+        lda  #0
+        sta  string_out+4
+        lda  #<string_out
+        ldy  #>string_out
+        rts
 	}}
 }
 
-asmsub  str_uw0  (uword value @ AY) clobbers(A,X,Y)  {
+asmsub  str_uw0  (uword value @ AY) clobbers(X) -> str @AY  {
 	; ---- convert the uword in A/Y in decimal string form, with left padding 0s (5 positions total)
 	%asm {{
 	    jsr  conv.uword2decimal
 	    ldy  #0
--           lda  conv.uword2decimal.decTenThousands,y
-            sta  string_out,y
-            beq  +
-            iny
-            bne  -
-+	    rts
+-       lda  conv.uword2decimal.decTenThousands,y
+        sta  string_out,y
+        beq  +
+        iny
+        bne  -
++
+        lda  #<string_out
+        ldy  #>string_out
+        rts
 	}}
 }
 
-asmsub  str_uw  (uword value @ AY) clobbers(A,X,Y)  {
+asmsub  str_uw  (uword value @ AY) clobbers(X) -> str @AY  {
 	; ---- convert the uword in A/Y in decimal string form, without left padding 0s
 	%asm {{
 	    jsr  conv.uword2decimal
 	    ldx  #0
 _output_digits
 	    ldy  #0
--           lda  conv.uword2decimal.decTenThousands,y
-            beq  _allzero
-            cmp  #'0'
-            bne  _gotdigit
-            iny
-            bne  -
+-       lda  conv.uword2decimal.decTenThousands,y
+        beq  _allzero
+        cmp  #'0'
+        bne  _gotdigit
+        iny
+        bne  -
 _gotdigit   sta  string_out,x
-            inx
-            iny
-            lda  conv.uword2decimal.decTenThousands,y
-            bne  _gotdigit
-_end        lda  #0
-            sta  string_out,x
-            rts
+        inx
+        iny
+        lda  conv.uword2decimal.decTenThousands,y
+        bne  _gotdigit
+_end    lda  #0
+        sta  string_out,x
+        lda  #<string_out
+        ldy  #>string_out
+        rts
 
 _allzero    lda  #'0'
-            sta  string_out,x
-            inx
-            bne  _end
+        sta  string_out,x
+        inx
+        bne  _end
 	}}
 }
 
-asmsub  str_w  (word value @ AY) clobbers(A,X,Y)  {
+asmsub  str_w  (word value @ AY) clobbers(X) -> str @AY  {
 	; ---- convert the (signed) word in A/Y in decimal string form, without left padding 0's
 	%asm {{
 	    cpy  #0
@@ -191,18 +229,19 @@ asmsub  str_w  (word value @ AY) clobbers(A,X,Y)  {
 	    pha
 	    lda  #'-'
 	    sta  string_out
-            tya
-            eor  #255
-            tay
-            pla
-            eor  #255
-            clc
-            adc  #1
-            bcc  +
-            iny
+        tya
+        eor  #255
+        tay
+        pla
+        eor  #255
+        clc
+        adc  #1
+        bcc  +
+        iny
 +	    jsr  conv.uword2decimal
 	    ldx  #1
 	    bne  str_uw._output_digits
+	    rts
 	}}
 }
 

compiler/res/prog8lib/cx16/bmx.p8

@@ -126,7 +126,7 @@ bmx {
         ; If you're saving the whole screen width, you can leave screenwidth at 0.
         ; Returns: success status. If false, error_message points to the error message string.
         error_message = 0
-        if compression {
+        if compression!=0 {
             error_message = "compression not supported"
             return false
         }
@@ -184,7 +184,7 @@ save_end:
         for cx16.r0L in 0 to sizeof(header)-1 {
             header[cx16.r0L] = cbm.CHRIN()
         }
-        return not cbm.READST()
+        return cbm.READST()==0
     }
 
     sub read_palette() -> bool {
@@ -197,7 +197,7 @@ save_end:
         do {
             cx16.r4L = cbm.CHRIN()
             cx16.r4H = cbm.CHRIN()
-            if cx16.r3 {
+            if cx16.r3!=0 {
                 pokew(cx16.r3, cx16.r4)             ; into memory
                 cx16.r3+=2
             } else {
@@ -206,7 +206,7 @@ save_end:
             }
             cx16.r2L--
         } until cx16.r2L==0
-        return cbm.READST()==0 or cbm.READST()&$40    ; no error or eof?
+        return cbm.READST()==0 or cbm.READST()&$40!=0    ; no error or eof?
     }
 
     sub read_bitmap_padded(ubyte vbank, uword vaddr, uword screenwidth) -> bool {
@@ -220,7 +220,7 @@ save_end:
             if_cs
                 vbank++
         }
-        return cbm.READST()==0 or cbm.READST()&$40    ; no error or eof?
+        return cbm.READST()==0 or cbm.READST()&$40!=0    ; no error or eof?
     }
 
     sub read_bitmap(ubyte vbank, uword vaddr) -> bool {
@@ -229,11 +229,11 @@ save_end:
         cx16.vaddr(vbank, vaddr, 0, 1)
         repeat height
             read_scanline(cx16.r3)
-        return cbm.READST()==0 or cbm.READST()&$40    ; no error or eof?
+        return cbm.READST()==0 or cbm.READST()&$40!=0    ; no error or eof?
     }
 
     sub read_scanline(uword size) {
-        while size {
+        while size!=0 {
             cx16.r0 = cx16.MACPTR(min(255, size) as ubyte, &cx16.VERA_DATA0, true)
             if_cs {
                 ; no MACPTR support
@@ -251,7 +251,7 @@ save_end:
         for cx16.r0L in 0 to sizeof(header)-1 {
             cbm.CHROUT(header[cx16.r0L])
         }
-        return not cbm.READST()
+        return cbm.READST()==0
     }
 
     sub write_palette() -> bool {
@@ -262,7 +262,7 @@ save_end:
         cx16.r2L = lsb(palette_entries)
         cx16.vaddr(1, $fa00+palette_start*2, 0, 1)
         do {
-            if cx16.r3 {
+            if cx16.r3!=0 {
                 cbm.CHROUT(@(cx16.r3))      ; from memory
                 cx16.r3++
                 cbm.CHROUT(@(cx16.r3))
@@ -273,7 +273,7 @@ save_end:
             }
             cx16.r2L--
         } until cx16.r2L==0
-        return not cbm.READST()
+        return cbm.READST()==0
     }
 
     sub write_bitmap(ubyte vbank, uword vaddr, uword screenwidth) -> bool {
@@ -282,19 +282,19 @@ save_end:
         cx16.vaddr(vbank, vaddr, 0, 1)
         cx16.r3 = bytes_per_scanline(width)         ; num bytes per image scanline
         cx16.r2 = 0
-        if screenwidth
+        if screenwidth!=0
             cx16.r2 = bytes_per_scanline(screenwidth-width)   ; num bytes padding per screen scanline
         repeat height {
             write_scanline(cx16.r3)
             repeat cx16.r2
                 cx16.r0L = cx16.VERA_DATA0
         }
-        return not cbm.READST()
+        return cbm.READST()==0
 
         sub write_scanline(uword size) {
-            while size {
+            while size!=0 {
                 cx16.r0L = lsb(size)
-                if msb(size)
+                if msb(size)!=0
                     cx16.r0L = 0        ; 256 bytes
                 cx16.r0 = cx16.MCIOUT(cx16.r0L, &cx16.VERA_DATA0, true)
                 if_cs {

compiler/res/prog8lib/cx16/diskio.p8

@@ -79,7 +79,7 @@ io_error:
         cbm.CLRCHN()        ; restore default i/o devices
         cbm.CLOSE(READ_IO_CHANNEL)
 
-        if status and status & $40 == 0 {            ; bit 6=end of file
+        if status!=0 and status & $40 == 0 {            ; bit 6=end of file
             txt.print("\ni/o error, status: ")
             txt.print_ub(status)
             txt.nl()
@@ -121,7 +121,7 @@ io_error:
 io_error:
         cbm.CLRCHN()
         cbm.CLOSE(READ_IO_CHANNEL)
-        if status and status & $40 == 0
+        if status!=0 and status & $40 == 0
             return 0
         return list_filename
     }
@@ -210,14 +210,14 @@ io_error:
             ubyte blocks_lsb = cbm.CHRIN()
             ubyte blocks_msb = cbm.CHRIN()
 
-            if cbm.READST()
+            if cbm.READST()!=0
                 goto close_end
 
             list_blocks = mkword(blocks_msb, blocks_lsb)
 
             ; read until the filename starts after the first "
             while cbm.CHRIN()!='\"'  {
-                if cbm.READST()
+                if cbm.READST()!=0
                     goto close_end
             }
 
@@ -240,7 +240,7 @@ io_error:
             list_filetype[0] = cx16.r15L
             list_filetype[1] = cbm.CHRIN()
             list_filetype[2] = cbm.CHRIN()
-            while cbm.CHRIN() {
+            while cbm.CHRIN()!=0 {
                 ; read the rest of the entry until the end
             }
 
@@ -248,7 +248,7 @@ io_error:
             void cbm.CHRIN()
 
             if not list_skip_disk_name {
-                if not list_pattern
+                if list_pattern==0
                     return true
                 if string.pattern_match(list_filename, list_pattern)
                     return true
@@ -308,14 +308,14 @@ close_end:
         ; -- read from the currently open file, up to the given number of bytes.
         ;    returns the actual number of bytes read.  (checks for End-of-file and error conditions)
         ;    NOTE: cannot be used to load into VRAM.  Use vload() or call cx16.MACPTR() yourself with the vera data register as address.
-        if not iteration_in_progress or not num_bytes
+        if not iteration_in_progress or num_bytes==0
             return 0
 
         reset_read_channel()
         list_blocks = 0     ; we reuse this variable for the total number of bytes read
 
         uword readsize
-        while num_bytes {
+        while num_bytes!=0 {
             readsize = 255
             if num_bytes<readsize
                 readsize = num_bytes
@@ -327,7 +327,7 @@ close_end:
             if msb(bufferpointer) == $c0
                 bufferpointer = mkword($a0, lsb(bufferpointer))  ; wrap over bank boundary
             num_bytes -= readsize
-            if cbm.READST() & $40 {
+            if cbm.READST() & $40 !=0 {
                 f_close()       ; end of file, close it
                 break
             }
@@ -341,10 +341,10 @@ byte_read_loop:         ; fallback if MACPTR isn't supported on the device
             lda  bufferpointer+1
             sta  m_in_buffer+2
         }}
-        while num_bytes {
-            if cbm.READST() {
+        while num_bytes!=0 {
+            if cbm.READST()!=0 {
                 f_close()
-                if cbm.READST() & $40    ; eof?
+                if cbm.READST() & $40 !=0    ; eof?
                     return list_blocks   ; number of bytes read
                 return 0  ; error.
             }
@@ -371,7 +371,7 @@ m_in_buffer     sta  $ffff
 
         reset_read_channel()
         uword total_read = 0
-        while not cbm.READST() {
+        while cbm.READST()==0 {
             cx16.r0 = f_read(bufferpointer, 256)
             total_read += cx16.r0
             bufferpointer += cx16.r0
@@ -444,7 +444,7 @@ _end        rts
         cbm.SETLFS(WRITE_IO_CHANNEL, drivenumber, WRITE_IO_CHANNEL)
         void cbm.OPEN()             ; open 13,8,13,"filename"
         if_cc {
-            return not cbm.READST()
+            return cbm.READST()==0
         }
         cbm.CLOSE(WRITE_IO_CHANNEL)
         f_close_w()
@@ -466,7 +466,7 @@ _end        rts
                 if cbm.READST()!=0
                     return false
             } until num_bytes==0
-            return not cbm.READST()
+            return cbm.READST()==0
 
 no_mciout:
             ; the device doesn't support MCIOUT, use a normal per-byte write loop
@@ -474,7 +474,7 @@ no_mciout:
                 cbm.CHROUT(@(bufferpointer))
                 bufferpointer++
             }
-            return not cbm.READST()
+            return cbm.READST()==0
         }
         return true
     }
@@ -493,8 +493,7 @@ no_mciout:
         str device_not_present_error = "device not present #xx"
         if cbm.READST()==128 {
             device_not_present_error[len(device_not_present_error)-2] = 0
-            conv.str_ub(drivenumber)
-            void string.copy(conv.string_out, &device_not_present_error+len(device_not_present_error)-2)
+            void string.copy(conv.str_ub(drivenumber), &device_not_present_error+len(device_not_present_error)-2)
             return device_not_present_error
         }
 
@@ -506,7 +505,7 @@ no_mciout:
             goto io_error
         void cbm.CHKIN(15)        ; use #15 as input channel
 
-        while not cbm.READST() {
+        while cbm.READST()==0 {
             cx16.r5L = cbm.CHRIN()
             if cx16.r5L=='\r' or cx16.r5L=='\n'
                 break
@@ -562,9 +561,9 @@ io_error:
         }}
 
         if_cc
-            cx16.r0L = cbm.READST()==0
+            cx16.r0L = cbm.READST()==0 as ubyte
 
-        return cx16.r0L
+        return cx16.r0L as bool
     }
 
     ; Use kernal LOAD routine to load the given program file in memory.
@@ -596,7 +595,7 @@ io_error:
         cbm.SETNAM(string.length(filenameptr), filenameptr)
         ubyte secondary = 1
         cx16.r1 = 0
-        if address_override
+        if address_override!=0
             secondary = 0
         if headerless
             secondary |= %00000010  ; activate cx16 kernal headerless load support
@@ -659,7 +658,7 @@ io_error:
         return $2000 * (cx16.getrambank() - startbank) + endaddress - startaddress
     }
 
-    asmsub vload(str name @R0, ubyte bank @A, uword startaddress @R1) clobbers(X, Y) -> ubyte @A {
+    asmsub vload(str name @R0, ubyte bank @A, uword startaddress @R1) clobbers(X, Y) -> bool @A {
         ; -- like the basic command VLOAD "filename",drivenumber,bank,address
         ;    loads a file into Vera's video memory in the given bank:address, returns success in A
         ;    the file has to have the usual 2 byte header (which will be skipped)
@@ -698,7 +697,7 @@ internal_vload:
         }}
     }
 
-    asmsub vload_raw(str name @R0, ubyte bank @A, uword startaddress @R1) clobbers(X, Y) -> ubyte @A {
+    asmsub vload_raw(str name @R0, ubyte bank @A, uword startaddress @R1) clobbers(X, Y) -> bool @A {
         ; -- like the basic command BVLOAD "filename",drivenumber,bank,address
         ;    loads a file into Vera's video memory in the given bank:address, returns success in A
         ;    the file is read fully including the first two bytes.
@@ -755,7 +754,7 @@ internal_vload:
         repeat 6 {
             void cbm.CHRIN()
         }
-        while cbm.CHRIN() {
+        while cbm.CHRIN()!=0 {
             ; skip first line (drive label)
         }
         while cbm.CHRIN()!='"' {
@@ -786,7 +785,7 @@ internal_vload:
 io_error:
         cbm.CLRCHN()
         cbm.CLOSE(READ_IO_CHANNEL)
-        if status and status & $40 == 0
+        if status!=0 and status & $40 == 0
             return 0
         if @(cx16.r12)==0 {
             cx16.r12--

compiler/res/prog8lib/cx16/floats.p8

@@ -1,7 +1,7 @@
 ; Prog8 definitions for floating point handling on the CommanderX16
 
 %option enable_floats, no_symbol_prefixing, ignore_unused
-%import floats_functions
+%import shared_floats_functions
 
 floats {
 	; ---- this block contains C-64 compatible floating point related functions ----

compiler/res/prog8lib/cx16/gfx2.p8

@@ -55,14 +55,16 @@ gfx2 {
             }
             else -> {
                 ; back to default text mode
-                cx16.r15L = cx16.VERA_DC_VIDEO & %00000111 ; retain chroma + output mode
-                cbm.CINT()
-                cx16.VERA_DC_VIDEO = (cx16.VERA_DC_VIDEO & %11111000) | cx16.r15L
+                if active_mode!=0 {
+                    cx16.r15L = cx16.VERA_DC_VIDEO & %00000111 ; retain chroma + output mode
+                    cbm.CINT()
+                    cx16.VERA_DC_VIDEO = (cx16.VERA_DC_VIDEO & %11111000) | cx16.r15L
+                }
             }
         }
 
         init_mode(mode)
-        if active_mode
+        if active_mode!=0
             clear_screen(0)
     }
 
@@ -305,17 +307,17 @@ gfx2 {
         }
 
         word @zp d = 0
-        cx16.r1L = true      ; 'positive_ix'
+        cx16.r1L = 1  ;; true      ; 'positive_ix'
         if dx < 0 {
             dx = -dx
-            cx16.r1L = false
+            cx16.r1L = 0 ;; false
         }
         word @zp dx2 = dx*2
         word @zp dy2 = dy*2
         cx16.r14 = x1       ; internal plot X
 
         if dx >= dy {
-            if cx16.r1L {
+            if cx16.r1L!=0 {
                 repeat {
                     plot(cx16.r14, y1, color)
                     if cx16.r14==x2
@@ -342,7 +344,7 @@ gfx2 {
             }
         }
         else {
-            if cx16.r1L {
+            if cx16.r1L!=0 {
                 repeat {
                     plot(cx16.r14, y1, color)
                     if y1 == y2
@@ -705,7 +707,7 @@ gfx2 {
         push_stack(xx, xx, yy, 1)
         push_stack(xx, xx, yy + 1, -1)
         word left = 0
-        while cx16.r12L {
+        while cx16.r12L!=0 {
             pop_stack()
             xx = x1
             ; possible speed optimization: if mode==1 (256c) use vera autodecrement instead of pget(), but code bloat not worth it?
@@ -725,15 +727,15 @@ gfx2 {
             xx = x1 + 1
 
             do {
-                cx16.r9 = xx
+                cx16.r9s = xx
                 ; possible speed optimization: if mode==1 (256c) use vera autoincrement instead of pget(), but code bloat not worth it?
                 while xx <= width-1 {
                     if pget(xx as uword, yy as uword) != cx16.r11L
                         break
                     xx++
                 }
-                if cx16.r9!=xx
-                    horizontal_line(cx16.r9, yy as uword, (xx as uword)-cx16.r9, cx16.r10L)
+                if cx16.r9s!=xx
+                    horizontal_line(cx16.r9, yy as uword, xx-cx16.r9s as uword, cx16.r10L)
 
                 push_stack(left, xx - 1, yy, dy)
                 if xx > x2 + 1
@@ -845,7 +847,7 @@ skip:
         when active_mode {
             1 -> {
                 ; lores 256c
-                while @(sctextptr) {
+                while @(sctextptr)!=0 {
                     chardataptr = charset_addr + (@(sctextptr) as uword)*8
                     cx16.vaddr(charset_bank, chardataptr, 1, 1)
                     repeat 8 {
@@ -874,9 +876,9 @@ skip:
                 ; hires 4c
                 ; we're going to use a few cx16 registers to make sure every variable is in zeropage in the inner loop.
                 cx16.r11L = color
-                while @(sctextptr) {
+                while @(sctextptr)!=0 {
                     chardataptr = charset_addr + (@(sctextptr) as uword)*8
-                    cx16.vaddr(charset_bank, chardataptr, 1, true)  ; for reading the chardata from Vera data channel 1
+                    cx16.vaddr(charset_bank, chardataptr, 1, 1)  ; for reading the chardata from Vera data channel 1
                     position(xx, yy)              ; only calculated once, we update vera address in the loop instead
                     cx16.VERA_ADDR_H &= $0f     ; no auto increment
                     repeat 8 {

compiler/res/prog8lib/cx16/graphics.p8

@@ -57,20 +57,20 @@ graphics {
     }
 
     sub fillrect(uword xx, uword yy, uword width, uword height) {
-        cx16.GRAPH_draw_rect(xx, yy, width, height, 0, 1)
+        cx16.GRAPH_draw_rect(xx, yy, width, height, 0, true)
     }
 
     sub rect(uword xx, uword yy, uword width, uword height) {
-        cx16.GRAPH_draw_rect(xx, yy, width, height, 0, 0)
+        cx16.GRAPH_draw_rect(xx, yy, width, height, 0, false)
     }
 
     sub horizontal_line(uword xx, uword yy, uword length) {
-        if length
+        if length!=0
             cx16.GRAPH_draw_line(xx, yy, xx+length-1, yy)
     }
 
     sub vertical_line(uword xx, uword yy, uword height) {
-        if height
+        if height!=0
             cx16.GRAPH_draw_line(xx, yy, xx, yy+height-1)
     }
 

compiler/res/prog8lib/cx16/monogfx.p8

@@ -7,7 +7,6 @@
 ; NOTE: a lot of the code here is similar or the same to that in gfx2
 ; NOTE: For sake of speed, NO BOUNDS CHECKING is performed in most routines!
 ;       You'll have to make sure yourself that you're not writing outside of bitmap boundaries!
-;
 
 monogfx {
 
@@ -16,7 +15,10 @@ monogfx {
     ; read-only control variables:
     uword width = 0
     uword height = 0
-    bool dont_stipple_flag = true            ; set to false to enable stippling mode
+    ubyte mode
+    const ubyte MODE_NORMAL  = %00000000
+    const ubyte MODE_STIPPLE = %00000001
+    const ubyte MODE_INVERT  = %00000010
 
     sub lores() {
         ; enable 320*240 bitmap mode
@@ -29,6 +31,7 @@ monogfx {
         cx16.VERA_L1_TILEBASE = 0
         width = 320
         height = 240
+        mode = MODE_NORMAL
         clear_screen(0)
     }
 
@@ -43,6 +46,7 @@ monogfx {
         cx16.VERA_L1_TILEBASE = %00000001
         width = 640
         height = 480
+        mode = MODE_NORMAL
         clear_screen(0)
     }
 
@@ -53,8 +57,11 @@ monogfx {
         cx16.VERA_DC_VIDEO = (cx16.VERA_DC_VIDEO & %11111000) | cx16.r15L
     }
 
+    sub drawmode(ubyte dm) {
+        mode = dm
+    }
+
     sub clear_screen(ubyte color) {
-        stipple(false)
         position(0, 0)
         when width {
             320 -> {
@@ -69,10 +76,6 @@ monogfx {
         position(0, 0)
     }
 
-    sub stipple(bool enable) {
-        dont_stipple_flag = not enable
-    }
-
     sub rect(uword xx, uword yy, uword rwidth, uword rheight, bool draw) {
         if rwidth==0 or rheight==0
             return
@@ -122,8 +125,9 @@ monogfx {
                 bne  -
                 sta  cx16.r0L           ; new left byte
 +
-                lda  p8v_dont_stipple_flag
-                bne  _dontstipple
+                lda  p8v_mode
+                lsr  a
+                bcc  _dontstipple
                 ; determine stipple pattern
                 lda  p8v_yy
                 and  #1
@@ -162,24 +166,36 @@ _clear
         }
 
         ubyte separate_pixels = (8-lsb(xx)) & 7
-        if separate_pixels {
-            if dont_stipple_flag {
-                position(xx,yy)
-                cx16.VERA_ADDR_H &= %00000111   ; vera auto-increment off
-                if draw
-                    cx16.VERA_DATA0 |= masked_starts[separate_pixels]
-                else
-                    cx16.VERA_DATA0 &= ~masked_starts[separate_pixels]
-                xx += separate_pixels
-            } else {
-                repeat separate_pixels {
-                    plot(xx, yy, draw)
-                    xx++
+        if separate_pixels!=0 {
+            when mode {
+                MODE_NORMAL -> {
+                    position(xx,yy)
+                    cx16.VERA_ADDR_H &= %00000111   ; vera auto-increment off
+                    if draw
+                        cx16.VERA_DATA0 |= masked_starts[separate_pixels]
+                    else
+                        cx16.VERA_DATA0 &= ~masked_starts[separate_pixels]
+                    xx += separate_pixels
+                }
+                MODE_STIPPLE -> {
+                    repeat separate_pixels {
+                        plot(xx, yy, draw)
+                        xx++
+                    }
+                }
+                MODE_INVERT -> {
+                    position(xx,yy)
+                    cx16.VERA_ADDR_H &= %00000111   ; vera auto-increment off
+                    if draw
+                        cx16.VERA_DATA0 ^= masked_starts[separate_pixels]
+                    else
+                        cx16.VERA_DATA0 &= masked_starts[separate_pixels]
+                    xx += separate_pixels
                 }
             }
             length -= separate_pixels
         }
-        if length {
+        if length!=0 {
             position(xx, yy)
             separate_pixels = lsb(length) & 7
             xx += length & $fff8
@@ -194,10 +210,28 @@ _clear
                 bne  +
                 ldy  #0     ; black
                 bra  _loop
-+               lda  p8v_dont_stipple_flag
-                beq  _stipple
-                ldy  #255       ; don't stipple
++               lda  p8v_mode
+                lsr  a
+                bcs  _stipple
+                lsr  a
+                bcs  _inverted
+                ldy  #255       ; normal drawing mode
                 bra  _loop
+
+_inverted       lda  #0
+                jsr  cx16.vaddr_clone
+_invertedloop   lda  p8v_length
+                ora  p8v_length+1
+                beq  _done
+                lda  cx16.VERA_DATA1
+                eor  #255
+                sta  cx16.VERA_DATA0
+                lda  p8v_length
+                bne  +
+                dec  p8v_length+1
++               dec  p8v_length
+                bra  _invertedloop
+
 _stipple        lda  p8v_yy
                 and  #1         ; determine stipple pattern to use
                 bne  +
@@ -216,16 +250,26 @@ _loop           lda  p8v_length
 _done
             }}
 
-            if dont_stipple_flag {
+            when mode {
+                MODE_NORMAL -> {
+                    cx16.VERA_ADDR_H &= %00000111   ; vera auto-increment off
+                    if draw
+                        cx16.VERA_DATA0 |= masked_ends[separate_pixels]
+                    else
+                        cx16.VERA_DATA0 &= ~masked_ends[separate_pixels]
+                }
+                MODE_STIPPLE -> {
+                    repeat separate_pixels {
+                        plot(xx, yy, draw)
+                        xx++
+                    }
+                }
+                MODE_INVERT -> {
                 cx16.VERA_ADDR_H &= %00000111   ; vera auto-increment off
                 if draw
-                    cx16.VERA_DATA0 |= masked_ends[separate_pixels]
+                    cx16.VERA_DATA0 ^= masked_ends[separate_pixels]
                 else
-                    cx16.VERA_DATA0 &= ~masked_ends[separate_pixels]
-            } else {
-                repeat separate_pixels {
-                    plot(xx, yy, draw)
-                    xx++
+                    cx16.VERA_DATA0 &= masked_ends[separate_pixels]
                 }
             }
         }
@@ -253,7 +297,16 @@ _done
     sub vertical_line(uword xx, uword yy, uword lheight, bool draw) {
         cx16.r15L = monogfx.plot.maskbits[xx as ubyte & 7]           ; bitmask
         if draw {
-            if dont_stipple_flag {
+            %asm {{
+                lda  p8v_mode
+                and  #p8c_MODE_INVERT
+                beq  +
+                lda  #$45       ; eor ZP modifying code
+                bne  ++
++               lda  #$05       ; ora ZP modifying code
++               sta  drawmode
+         }}
+            if mode!=MODE_STIPPLE {
                 ; draw continuous line.
                 position2(xx,yy,true)
                 if width==320
@@ -263,7 +316,7 @@ _done
                 repeat lheight {
                     %asm {{
                         lda  cx16.VERA_DATA0
-                        ora  cx16.r15L
+drawmode:               ora  cx16.r15L
                         sta  cx16.VERA_DATA1
                     }}
                 }
@@ -340,17 +393,17 @@ _done
         }
 
         word @zp d = 0
-        cx16.r1L = true      ; 'positive_ix'
+        cx16.r1L = 1 ;; true      ; 'positive_ix'
         if dx < 0 {
             dx = -dx
-            cx16.r1L = false
+            cx16.r1L = 0 ;; false
         }
         word @zp dx2 = dx*2
         word @zp dy2 = dy*2
         cx16.r14 = x1       ; internal plot X
 
         if dx >= dy {
-            if cx16.r1L {
+            if cx16.r1L!=0 {
                 repeat {
                     plot(cx16.r14, y1, draw)
                     if cx16.r14==x2
@@ -377,7 +430,7 @@ _done
             }
         }
         else {
-            if cx16.r1L {
+            if cx16.r1L!=0 {
                 repeat {
                     plot(cx16.r14, y1, draw)
                     if y1 == y2
@@ -554,12 +607,19 @@ _done
         if draw {
             ; solid color or perhaps stipple
             %asm {{
+                lda  p8v_mode
+                lsr  a
+                bcs  +
+                lsr  a
+                bcs  p8l_invert
+                bra  p8l_nostipple
++               ; stipple mode
                 lda  p8v_xx
                 eor  p8v_yy
-                ora  p8v_dont_stipple_flag
                 and  #1
             }}
             if_nz {
+nostipple:
                 prepare()
                 %asm {{
                     tsb  cx16.VERA_DATA0
@@ -572,6 +632,16 @@ _done
                 trb  cx16.VERA_DATA0
             }}
         }
+        return
+
+invert:
+        prepare()
+        %asm {{
+            lda  cx16.VERA_DATA0
+            eor  p8v_maskbits,y
+            sta  cx16.VERA_DATA0
+        }}
+        return
 
         sub prepare() {
             %asm {{
@@ -606,7 +676,7 @@ _done
         plot(xx, yy, draw)
     }
 
-    sub pget(uword @zp xx, uword yy) -> ubyte {
+    sub pget(uword @zp xx, uword yy) -> bool {
         %asm {{
             lda  p8v_xx
             and  #7
@@ -708,7 +778,7 @@ _done
             }}
             yy+=dy
         }
-        cx16.r11L = pget(xx as uword, yy as uword)        ; old_color
+        cx16.r11L = pget(xx as uword, yy as uword) as ubyte        ; old_color
         if cx16.r11L == cx16.r10L
             return
         if xx<0 or xx>width-1 or yy<0 or yy>height-1
@@ -716,16 +786,16 @@ _done
         push_stack(xx, xx, yy, 1)
         push_stack(xx, xx, yy + 1, -1)
         word left = 0
-        while cx16.r12L {
+        while cx16.r12L!=0 {
             pop_stack()
             xx = x1
             while xx >= 0 {
-                if pget(xx as uword, yy as uword) != cx16.r11L
+                if pget(xx as uword, yy as uword) as ubyte != cx16.r11L
                     break
                 xx--
             }
             if x1!=xx
-                horizontal_line(xx as uword+1, yy as uword, x1-xx as uword, cx16.r10L)
+                horizontal_line(xx as uword+1, yy as uword, x1-xx as uword, cx16.r10L as bool)
             else
                 goto skip
 
@@ -735,14 +805,14 @@ _done
             xx = x1 + 1
 
             do {
-                cx16.r9 = xx
+                cx16.r9s = xx
                 while xx <= width-1 {
-                    if pget(xx as uword, yy as uword) != cx16.r11L
+                    if pget(xx as uword, yy as uword) as ubyte != cx16.r11L
                         break
                     xx++
                 }
-                if cx16.r9!=xx
-                    horizontal_line(cx16.r9, yy as uword, (xx as uword)-cx16.r9, cx16.r10L)
+                if cx16.r9s!=xx
+                    horizontal_line(cx16.r9, yy as uword, xx-cx16.r9s as uword, cx16.r10L as bool)
 
                 push_stack(left, xx - 1, yy, dy)
                 if xx > x2 + 1
@@ -750,7 +820,7 @@ _done
 skip:
                 xx++
                 while xx <= x2 {
-                    if pget(xx as uword, yy as uword) == cx16.r11L
+                    if pget(xx as uword, yy as uword) as ubyte == cx16.r11L
                         break
                     xx++
                 }
@@ -783,7 +853,7 @@ skip:
         cx16.screen_set_charset(charset, 0)
     }
 
-    sub text(uword @zp xx, uword yy, bool draw, uword sctextptr) {
+    sub text(uword @zp xx, uword yy, bool draw, str sctextptr) {
         ; -- Write some text at the given pixel position. The text string must be in screencode encoding (not petscii!).
         ;    You must also have called text_charset() first to select and prepare the character set to use.
         uword chardataptr
@@ -791,7 +861,18 @@ skip:
         ubyte[8] @shared char_bitmap_bytes_right
 
         cx16.r3 = sctextptr
-        while @(cx16.r3) {
+        %asm {{
+            lda  p8v_mode
+            cmp  #p8c_MODE_INVERT
+            beq  +
+            lda  #$0d       ; ORA abs   modifying code
+            bne  ++
++           lda  #$4d       ; EOR abs   modifying code
++           sta  cdraw_mod1
+            sta  cdraw_mod2
+        }}
+
+        while @(cx16.r3)!=0 {
             chardataptr = charset_addr + @(cx16.r3) * $0008
             ; copy the character bitmap into RAM
             cx16.vaddr_autoincr(charset_bank, chardataptr, 0, 1)
@@ -824,7 +905,7 @@ skip:
                 %asm {{
                     ldy  #0
 -                   lda  p8v_char_bitmap_bytes_left,y
-                    ora  cx16.VERA_DATA1
+cdraw_mod1          ora  cx16.VERA_DATA1
                     sta  cx16.VERA_DATA0
                     iny
                     cpy  #8
@@ -843,14 +924,14 @@ skip:
                 }}
             }
             ; right part of shifted char
-            if lsb(xx) & 7 {
+            if lsb(xx) & 7 !=0 {
                 position2(xx+8, yy, true)
                 set_autoincrs()
                 if draw {
                     %asm {{
                         ldy  #0
 -                       lda  p8v_char_bitmap_bytes_right,y
-                        ora  cx16.VERA_DATA1
+cdraw_mod2              ora  cx16.VERA_DATA1
                         sta  cx16.VERA_DATA0
                         iny
                         cpy  #8

compiler/res/prog8lib/cx16/psg.p8

@@ -134,7 +134,7 @@ psg {
                 }
                 1 -> {
                     ; sustain
-                    if envelope_sustains[cx16.r1L] {
+                    if envelope_sustains[cx16.r1L]!=0 {
                         envelope_sustains[cx16.r1L]--
                     } else {
                         envelope_states[cx16.r1L] = 2  ; start release
@@ -143,7 +143,7 @@ psg {
                 2 -> {
                     ; release
                     cx16.r0 = envelope_volumes[cx16.r1L] - envelope_releases[cx16.r1L] * $0040
-                    if msb(cx16.r0) & %11000000 {
+                    if msb(cx16.r0) & %11000000 !=0 {
                         cx16.r0 = 0
                         envelope_releases[cx16.r1L] = 0
                     }

compiler/res/prog8lib/cx16/sprites.p8

@@ -47,7 +47,7 @@ sprites {
         sprite_reg = VERA_SPRITEREGS + spritenum*$0008
         cx16.r0L = cx16.vpeek(1, sprite_reg)
         cx16.r0H = cx16.vpeek(1, sprite_reg+1)
-        cx16.r1L = cx16.r0H & %00001000 !=0     ; bank
+        cx16.r1L = cx16.r0H & %00001000 !=0 as ubyte     ; bank
         cx16.r0 <<= 5                           ; address
     }
 
@@ -96,8 +96,8 @@ sprites {
     sub move(ubyte spritenum, word dx, word dy) {
         ; move a sprite based on its current position
         sprite_reg = VERA_SPRITEREGS + 2 + spritenum*$0008
-        cx16.r1 = mkword(cx16.vpeek(1, sprite_reg+1), cx16.vpeek(1, sprite_reg)) as word + dx
-        cx16.r2 = mkword(cx16.vpeek(1, sprite_reg+3), cx16.vpeek(1, sprite_reg+2)) as word + dy
+        cx16.r1s = mkword(cx16.vpeek(1, sprite_reg+1), cx16.vpeek(1, sprite_reg)) as word + dx
+        cx16.r2s = mkword(cx16.vpeek(1, sprite_reg+3), cx16.vpeek(1, sprite_reg+2)) as word + dy
         cx16.vpoke(1, sprite_reg, cx16.r1L)
         cx16.vpoke(1, sprite_reg+1, cx16.r1H)
         cx16.vpoke(1, sprite_reg+2, cx16.r2L)
@@ -107,7 +107,7 @@ sprites {
     sub movex(ubyte spritenum, word dx) {
         ; move a sprite horizontally based on its current position
         sprite_reg = VERA_SPRITEREGS + 2 + spritenum*$0008
-        cx16.r1 = mkword(cx16.vpeek(1, sprite_reg+1), cx16.vpeek(1, sprite_reg)) as word + dx
+        cx16.r1s = mkword(cx16.vpeek(1, sprite_reg+1), cx16.vpeek(1, sprite_reg)) as word + dx
         cx16.vpoke(1, sprite_reg, cx16.r1L)
         cx16.vpoke(1, sprite_reg+1, cx16.r1H)
     }
@@ -115,7 +115,7 @@ sprites {
     sub movey(ubyte spritenum, word dy) {
         ; move a sprite vertically based on its current position
         sprite_reg = VERA_SPRITEREGS + 4 + spritenum*$0008
-        cx16.r1 = mkword(cx16.vpeek(1, sprite_reg+1), cx16.vpeek(1, sprite_reg)) as word + dy
+        cx16.r1s = mkword(cx16.vpeek(1, sprite_reg+1), cx16.vpeek(1, sprite_reg)) as word + dy
         cx16.vpoke(1, sprite_reg, cx16.r1L)
         cx16.vpoke(1, sprite_reg+1, cx16.r1H)
     }
@@ -135,11 +135,11 @@ sprites {
     }
 
     sub flipx(ubyte spritenum, bool flipped) {
-        cx16.vpoke_mask(1, VERA_SPRITEREGS + 6 + spritenum*$0008, %11111110, flipped)
+        cx16.vpoke_mask(1, VERA_SPRITEREGS + 6 + spritenum*$0008, %11111110, flipped as ubyte)
     }
 
     sub flipy(ubyte spritenum, bool flipped) {
-        cx16.vpoke_mask(1, VERA_SPRITEREGS + 6 + spritenum*$0008, %11111101, flipped<<1)
+        cx16.vpoke_mask(1, VERA_SPRITEREGS + 6 + spritenum*$0008, %11111101, (flipped as ubyte)<<1)
     }
 
     sub set_palette_offset(ubyte spritenum, ubyte offset) {

compiler/res/prog8lib/cx16/syslib.p8

@@ -48,13 +48,13 @@ romsub $FFE4 = GETIN() clobbers(X,Y) -> bool @Pc, ubyte @ A    ; (via 810 ($32A)
 romsub $FFE7 = CLALL() clobbers(A,X)                            ; (via 812 ($32C)) close all files
 romsub $FFEA = UDTIM() clobbers(A,X)                            ; update the software clock
 romsub $FFED = SCREEN() -> ubyte @ X, ubyte @ Y                 ; read number of screen rows and columns
-romsub $FFF0 = PLOT(ubyte col @ Y, ubyte row @ X, bool dir @ Pc) -> ubyte @ X, ubyte @ Y       ; read/set position of cursor on screen.  Use txt.plot for a 'safe' wrapper that preserves X.
+romsub $FFF0 = PLOT(ubyte col @ Y, ubyte row @ X, bool dir @ Pc) clobbers(A) -> ubyte @ X, ubyte @ Y       ; read/set position of cursor on screen.  Use txt.plot for a 'safe' wrapper that preserves X.
 romsub $FFF3 = IOBASE() -> uword @ XY                           ; read base address of I/O devices
 
 ; ---- utility
 
-asmsub STOP2() clobbers(X) -> ubyte @A  {
-    ; -- check if STOP key was pressed, returns true if so.  More convenient to use than STOP() because that only sets the carry status flag.
+asmsub STOP2() clobbers(X) -> bool @A  {
+    ; -- check if STOP key was pressed, returns true if so.  More convenient to use than STOP() because that only sets the zero status flag.
     %asm {{
         jsr  cbm.STOP
         beq  +
@@ -90,6 +90,16 @@ sub CLEARST() {
     CLOSE(15)
 }
 
+asmsub kbdbuf_clear() {
+    ; -- convenience helper routine to clear the keyboard buffer
+    %asm {{
+-       jsr  GETIN
+        cmp  #0
+        bne  -
+        rts
+    }}
+}
+
 }
 
 cx16 {
@@ -436,7 +446,7 @@ romsub $ff4d = clock_set_date_time(uword yearmonth @R0, uword dayhours @R1, uwor
 romsub $ff50 = clock_get_date_time()  clobbers(A, X, Y)  -> uword @R0, uword @R1, uword @R2, uword @R3   ; result registers see clock_set_date_time()
 
 ; keyboard, mouse, joystick
-; note: also see the kbdbuf_clear() helper routine below!
+; note: also see the cbm.kbdbuf_clear() helper routine
 romsub $febd = kbdbuf_peek() -> ubyte @A, ubyte @X     ; key in A, queue length in X
 romsub $febd = kbdbuf_peek2() -> uword @AX             ; alternative to above to not have the hassle to deal with multiple return values
 romsub $fec0 = kbdbuf_get_modifiers() -> ubyte @A
@@ -534,15 +544,6 @@ asmsub get_screen_mode() -> byte @A, byte @X, byte @Y {
     }}
 }
 
-asmsub kbdbuf_clear() {
-    ; -- convenience helper routine to clear the keyboard buffer
-    %asm {{
--       jsr  cbm.GETIN
-        bne  -
-        rts
-    }}
-}
-
 asmsub mouse_config2(byte shape @A) clobbers (A, X, Y) {
     ; -- convenience wrapper function that handles the screen resolution for mouse_config() for you
     %asm {{
@@ -556,7 +557,7 @@ asmsub mouse_config2(byte shape @A) clobbers (A, X, Y) {
 
 asmsub mouse_pos() clobbers(X) -> ubyte @A {
     ; -- short wrapper around mouse_get() kernal routine:
-    ; -- gets the position of the mouse cursor in cx16.r0 and cx16.r1 (x/y coordinate), returns mouse button status.
+    ; -- gets the position of the mouse cursor in cx16.r0 and cx16.r1 (x/y coordinate), returns mouse button status in A.
     %asm {{
         ldx  #cx16.r0
         jmp  cx16.mouse_get
@@ -1213,6 +1214,20 @@ _continue   iny
         }}
     }
 
+    sub reset_system() {
+        ; Soft-reset the system back to initial power-on Basic prompt.
+        sys.reset_system()
+    }
+
+    sub poweroff_system() {
+        ; use the SMC to shutdown the computer
+        void cx16.i2c_write_byte($42, $01, $00)
+    }
+
+    sub set_led_brightness(ubyte brightness) {
+        void cx16.i2c_write_byte($42, $05, brightness)
+    }
+
 }
 
 sys {
@@ -1289,8 +1304,15 @@ asmsub  cleanup_at_exit() {
         sta  $01        ; rom bank 4 (basic)
         stz  $2d        ; hack to reset machine code monitor bank to 0
         jsr  cbm.CLRCHN		; reset i/o channels
+_exitcodeCarry = *+1
+        lda  #0
+        lsr  a
 _exitcode = *+1
         lda  #0        ; exit code possibly modified in exit()
+_exitcodeX = *+1
+        ldx  #0
+_exitcodeY = *+1
+        ldy  #0
         rts
     }}
 }
@@ -1405,13 +1427,13 @@ asmsub  set_rasterline(uword line @AY) {
     asmsub reset_system() {
         ; Soft-reset the system back to initial power-on Basic prompt.
         ; We do this via the SMC so that a true reset is performed that also resets the Vera fully.
+        ; (note: this is an asmsub on purpose! don't change into a normal sub)
         %asm {{
             sei
             ldx #$42
             ldy #2
             lda #0
-            jsr  cx16.i2c_write_byte
-            bra  *
+            jmp  cx16.i2c_write_byte
         }}
     }
 
@@ -1492,20 +1514,14 @@ _loop       lda  P8ZP_SCRATCH_W1
             rts     ; nothing to copy
 
 _copyshort
-            ; decrease source and target pointers so we can simply index by Y
-            lda  cx16.r0
-            bne  +
-            dec  cx16.r0+1
-+           dec  cx16.r0
-            lda  cx16.r1
-            bne  +
-            dec  cx16.r1+1
-+           dec  cx16.r1
-
+            dey
+            beq  +
 -           lda  (cx16.r0),y
             sta  (cx16.r1),y
             dey
             bne  -
++           lda  (cx16.r0),y
+            sta  (cx16.r1),y
             rts
 
 _longcopy
@@ -1661,6 +1677,33 @@ save_SCRATCH_ZPWORD2	.word  0
         }}
     }
 
+    asmsub exit2(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y) {
+        ; -- immediately exit the program with result values in the A, X and Y registers.
+        %asm {{
+            sta  cleanup_at_exit._exitcode
+            stx  cleanup_at_exit._exitcodeX
+            sty  cleanup_at_exit._exitcodeY
+            ldx  prog8_lib.orig_stackpointer
+            txs
+            jmp  cleanup_at_exit
+        }}
+    }
+
+    asmsub exit3(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y, bool carry @Pc) {
+        ; -- immediately exit the program with result values in the A, X and Y registers, and the Carry flag in the status register.
+        %asm {{
+            sta  cleanup_at_exit._exitcode
+            lda  #0
+            rol  a
+            sta  cleanup_at_exit._exitcodeCarry
+            stx  cleanup_at_exit._exitcodeX
+            sty  cleanup_at_exit._exitcodeY
+            ldx  prog8_lib.orig_stackpointer
+            txs
+            jmp  cleanup_at_exit
+        }}
+    }
+
     inline asmsub progend() -> uword @AY {
         %asm {{
             lda  #<prog8_program_end

compiler/res/prog8lib/cx16/textio.p8

@@ -4,7 +4,7 @@
 
 %import syslib
 %import conv
-
+%import shared_textio_functions
 
 txt {
 
@@ -18,26 +18,32 @@ const uword VERA_TEXTMATRIX_ADDR = $b000
 
 
 sub clear_screen() {
-    txt.chrout(147)
+    chrout(147)
 }
 
 sub home() {
-    txt.chrout(19)
+    chrout(19)
 }
 
 sub nl() {
-    txt.chrout('\n')
+    chrout('\n')
 }
 
 sub spc() {
-    txt.chrout(' ')
+    chrout(' ')
+}
+
+sub bell() {
+    chrout(7)
 }
 
 asmsub column(ubyte col @A) clobbers(A, X, Y) {
     ; ---- set the cursor on the given column (starting with 0) on the current line
     %asm {{
+        pha
         sec
         jsr  cbm.PLOT
+        pla
         tay
         clc
         jmp  cbm.PLOT
@@ -54,8 +60,10 @@ asmsub get_column() -> ubyte @Y {
 asmsub row(ubyte rownum @A) clobbers(A, X, Y) {
     ; ---- set the cursor on the given row (starting with 0) on the current line
     %asm {{
+        pha
         sec
         jsr  cbm.PLOT
+        pla
         tax
         clc
         jmp  cbm.PLOT

compiler/res/prog8lib/cx16/verafx.p8

@@ -1,7 +1,7 @@
 ; Somewhat experimental Vera FX support.
 ; Docs:
-; https://github.com/X16Community/x16-docs/blob/master/VERA%20FX%20Reference.md
-; https://docs.google.com/document/d/1q34uWOiM3Be2pnaHRVgSdHySI-qsiQWPTo_gfE54PTg/edit
+; https://github.com/X16Community/x16-docs/blob/master/X16%20Reference%20-%2010%20-%20VERA%20FX%20Reference.md
+; https://docs.google.com/document/d/1q34uWOiM3Be2pnaHRVgSdHySI-qsiQWPTo_gfE54PTg
 
 verafx {
     %option no_symbol_prefixing, ignore_unused
@@ -15,10 +15,10 @@ verafx {
             ; Vera version number is valid.
             ; Vera fx is available on Vera version 0.3.1 and later,
             ; so no need to even check VERA_DC_VER1, which contains 0 (or higher)
-            cx16.r1L = mkword(cx16.VERA_DC_VER2, cx16.VERA_DC_VER3) >= $0301
+            cx16.r1L = mkword(cx16.VERA_DC_VER2, cx16.VERA_DC_VER3) >= $0301 as ubyte
         }
         cx16.VERA_CTRL = cx16.r0L
-        return cx16.r1L
+        return cx16.r1L as bool
     }
 
     sub clear(ubyte vbank, uword vaddr, ubyte data, uword num_longwords) {

compiler/res/prog8lib/diskio.p8

@@ -68,7 +68,7 @@ io_error:
         cbm.CLRCHN()        ; restore default i/o devices
         cbm.CLOSE(READ_IO_CHANNEL)
 
-        if status and status & $40 == 0 {            ; bit 6=end of file
+        if status!=0 and status & $40 == 0 {            ; bit 6=end of file
             txt.print("\ni/o error, status: ")
             txt.print_ub(status)
             txt.nl()
@@ -83,7 +83,7 @@ io_error:
 
         cbm.SETNAM(1, "$")
         cbm.SETLFS(READ_IO_CHANNEL, drivenumber, 0)
-        ubyte okay = false
+        bool okay = false
         void cbm.OPEN()          ; open 12,8,0,"$"
         if_cs
             goto io_error
@@ -201,14 +201,14 @@ io_error:
             ubyte blocks_lsb = cbm.CHRIN()
             ubyte blocks_msb = cbm.CHRIN()
 
-            if cbm.READST()
+            if cbm.READST()!=0
                 goto close_end
 
             list_blocks = mkword(blocks_msb, blocks_lsb)
 
             ; read until the filename starts after the first "
             while cbm.CHRIN()!='\"'  {
-                if cbm.READST()
+                if cbm.READST()!=0
                     goto close_end
             }
 
@@ -231,7 +231,7 @@ io_error:
             list_filetype[0] = cx16.r15L
             list_filetype[1] = cbm.CHRIN()
             list_filetype[2] = cbm.CHRIN()
-            while cbm.CHRIN() {
+            while cbm.CHRIN()!=0 {
                 ; read the rest of the entry until the end
             }
 
@@ -239,7 +239,7 @@ io_error:
             void cbm.CHRIN()
 
             if not list_skip_disk_name {
-                if not list_pattern
+                if list_pattern==0
                     return true
                 if string.pattern_match(list_filename, list_pattern)
                     return true
@@ -297,7 +297,7 @@ close_end:
     sub f_read(uword bufferpointer, uword num_bytes) -> uword {
         ; -- read from the currently open file, up to the given number of bytes.
         ;    returns the actual number of bytes read.  (checks for End-of-file and error conditions)
-        if not iteration_in_progress or not num_bytes
+        if not iteration_in_progress or num_bytes==0
             return 0
 
         reset_read_channel()
@@ -309,10 +309,10 @@ close_end:
             lda  bufferpointer+1
             sta  m_in_buffer+2
         }}
-        while num_bytes {
-            if cbm.READST() {
+        while num_bytes!=0 {
+            if cbm.READST()!=0 {
                 f_close()
-                if cbm.READST() & $40    ; eof?
+                if cbm.READST() & $40 !=0    ; eof?
                     return list_blocks   ; number of bytes read
                 return 0  ; error.
             }
@@ -337,7 +337,7 @@ m_in_buffer     sta  $ffff
 
         reset_read_channel()
         uword total_read = 0
-        while not cbm.READST() {
+        while cbm.READST()==0 {
             cx16.r0 = f_read(bufferpointer, 256)
             total_read += cx16.r0
             bufferpointer += cx16.r0
@@ -396,7 +396,7 @@ _end        rts
         cbm.SETLFS(WRITE_IO_CHANNEL, drivenumber, 1)
         void cbm.OPEN()             ; open 13,8,1,"filename"
         if_cc
-            return not cbm.READST()
+            return cbm.READST()==0
         cbm.CLOSE(WRITE_IO_CHANNEL)
         f_close_w()
         return false
@@ -410,7 +410,7 @@ _end        rts
                 cbm.CHROUT(@(bufferpointer))
                 bufferpointer++
             }
-            return not cbm.READST()
+            return cbm.READST()==0
         }
         return true
     }
@@ -429,8 +429,7 @@ _end        rts
         str device_not_present_error = "device not present #xx"
         if cbm.READST()==128 {
             device_not_present_error[len(device_not_present_error)-2] = 0
-            conv.str_ub(drivenumber)
-            void string.copy(conv.string_out, &device_not_present_error+len(device_not_present_error)-2)
+            void string.copy(conv.str_ub(drivenumber), &device_not_present_error+len(device_not_present_error)-2)
             return device_not_present_error
         }
         uword messageptr = &list_filename
@@ -441,7 +440,7 @@ _end        rts
             goto io_error
         void cbm.CHKIN(15)        ; use #15 as input channel
 
-        while not cbm.READST() {
+        while cbm.READST()==0 {
             cx16.r5L = cbm.CHRIN()
             if cx16.r5L=='\r' or cx16.r5L=='\n'
                 break
@@ -481,9 +480,9 @@ io_error:
         }}
 
         if_cc
-            cx16.r0L = cbm.READST()==0
+            cx16.r0L = cbm.READST()==0 as ubyte
 
-        return cx16.r0L
+        return cx16.r0L as bool
     }
 
     ; Use kernal LOAD routine to load the given program file in memory.
@@ -496,7 +495,7 @@ io_error:
         cbm.SETNAM(string.length(filenameptr), filenameptr)
         ubyte secondary = 1
         cx16.r1 = 0
-        if address_override
+        if address_override!=0
             secondary = 0
         cbm.SETLFS(1, drivenumber, secondary)
         %asm {{

compiler/res/prog8lib/math.p8

@@ -75,7 +75,7 @@ _sinecosR8	.char  trunc(127.0 * sin(range(180+45) * rad(360.0/180.0)))
         }}
     }
 
-    asmsub rnd() -> ubyte @A {
+    asmsub rnd() clobbers(Y) -> ubyte @A {
         %asm {{
             jmp  math.randbyte
         }}
@@ -87,6 +87,24 @@ _sinecosR8	.char  trunc(127.0 * sin(range(180+45) * rad(360.0/180.0)))
         }}
     }
 
+    sub randrange(ubyte n) -> ubyte {
+        ; -- return random number uniformly distributed from 0 to n-1 (compensates for divisibility bias)
+        cx16.r0H = 255 / n * n
+        do {
+            cx16.r0L = math.rnd()
+        } until cx16.r0L < cx16.r0H
+        return cx16.r0L % n
+    }
+
+    sub randrangew(uword n) -> uword {
+        ; -- return random number uniformly distributed from 0 to n-1 (compensates for divisibility bias)
+        cx16.r1 = 65535 / n * n
+        do {
+            cx16.r0 = math.rndw()
+        } until cx16.r0 < cx16.r1
+        return cx16.r0 % n
+    }
+
     asmsub rndseed(uword seed1 @AY, uword seed2 @R0) clobbers(A,Y) {
         ; -- set new pseudo RNG's seed values. Defaults are: $00c2, $1137
         %asm {{
@@ -486,47 +504,137 @@ log2_tab
 
     sub crc16(uword data, uword length) -> uword {
         ; calculates the CRC16 (XMODEM) checksum of the buffer.
-        cx16.r1 = data  ; make sure pointer is in zp (on cx16)
-        cx16.r0 = 0  ; the crc value
-        repeat length {
-            cx16.r0H ^= @(cx16.r1)
-            repeat 8 {
-                if cx16.r0H & $80 {
-                    cx16.r0 <<= 1
-                    cx16.r0 ^= $1021
-                }
-                else
-                    cx16.r0<<=1
-            }
-            cx16.r1++
+        ; There are also "streaming" crc16_start/update/end routines below, that allow you to calculate crc16 for data that doesn't fit in a single memory block.
+        crc16_start()
+        cx16.r13 = data
+        cx16.r14 = data+length
+        while cx16.r13!=cx16.r14 {
+            crc16_update(@(cx16.r13))
+            cx16.r13++
         }
-        return cx16.r0
+        return crc16_end()
+    }
+
+    sub crc16_start() {
+        ; start the "streaming" crc16
+        ; note: tracks the crc16 checksum in cx16.r15!
+        ;       if your code uses that, it must save/restore it before calling this routine
+        cx16.r15 = 0
+    }
+
+    asmsub crc16_update(ubyte value @A) {
+        ; update the "streaming" crc16 with next byte value
+        ; note: tracks the crc16 checksum in cx16.r15!
+        ;       if your code uses that, it must save/restore it before calling this routine
+        %asm {{
+            eor  cx16.r15H
+            sta  cx16.r15H
+            ldy  #8
+-           lda  cx16.r15H
+            asl  cx16.r15L
+            rol  cx16.r15H
+            and  #$80
+            beq  +
+            lda  cx16.r15H
+            eor  #$10
+            sta  cx16.r15H
+            lda  cx16.r15L
+            eor  #$21
+            sta  cx16.r15L
++           dey
+            bne  -
+            rts
+        }}
+; orignal prog8 code was:
+;        cx16.r15H ^= value
+;        repeat 8 {
+;            if cx16.r15H & $80 !=0 {
+;                cx16.r15 <<=1
+;                cx16.r15 ^= $1021
+;            } else
+;                cx16.r15<<=1
+;        }
+    }
+
+    sub crc16_end() -> uword {
+        ; finalize the "streaming" crc16, returns resulting crc16 value
+        return cx16.r15
     }
 
     sub crc32(uword data, uword length) {
         ; Calculates the CRC-32 (POSIX) checksum of the buffer.
         ; because prog8 doesn't have 32 bits integers, we have to split up the calculation over 2 words.
-        ; result stored in cx16.r0 (low word) and cx16.r1 (high word)
-        cx16.r2 = data  ; make sure pointer is in zp (on cx16)
-        cx16.r1 = 0
-        cx16.r0 = 0
-        repeat length {
-            cx16.r1H ^= @(cx16.r2)
-            repeat 8 {
-                if cx16.r1H & $80 {
-                    cx16.r0 <<= 1
-                    rol(cx16.r1)
-                    cx16.r1 ^= $04c1
-                    cx16.r0 ^= $1db7
-                }
-                else {
-                    cx16.r0 <<= 1
-                    rol(cx16.r1)
-                }
-            }
-            cx16.r2++
+        ; result stored in cx16.r14 (low word) and cx16.r15 (high word)
+        ; There are also "streaming" crc32_start/update/end routines below, that allow you to calculate crc32 for data that doesn't fit in a single memory block.
+        crc32_start()
+        cx16.r12 = data
+        cx16.r13 = data+length
+        while cx16.r12!=cx16.r13 {
+            crc32_update(@(cx16.r12))
+            cx16.r12++
         }
-        cx16.r1 ^= $ffff
-        cx16.r0 ^= $ffff
+        crc32_end()
+    }
+
+    sub crc32_start() {
+        ; start the "streaming" crc32
+        ; note: tracks the crc32 checksum in cx16.r14 and cx16.r15!
+        ;       if your code uses these, it must save/restore them before calling this routine
+        cx16.r14 = cx16.r15 = 0
+    }
+
+    asmsub crc32_update(ubyte value @A) {
+        ; update the "streaming" crc32 with next byte value
+        ; note: tracks the crc32 checksum in cx16.r14 and cx16.r15!
+        ;       if your code uses these, it must save/restore them before calling this routine
+        %asm {{
+            eor  cx16.r15H
+            sta  cx16.r15H
+            ldy  #8
+-           lda  cx16.r15H
+            asl  cx16.r14L
+            rol  cx16.r14H
+            rol  cx16.r15L
+            rol  cx16.r15H
+            and  #$80
+            beq  +
+            lda  cx16.r15H
+            eor  #$04
+            sta  cx16.r15H
+            lda  cx16.r15L
+            eor  #$c1
+            sta  cx16.r15L
+            lda  cx16.r14H
+            eor  #$1d
+            sta  cx16.r14H
+            lda  cx16.r14L
+            eor  #$b7
+            sta  cx16.r14L
++           dey
+            bne  -
+            rts
+        }}
+; original prog8 code:
+;        cx16.r15H ^= value
+;        repeat 8 {
+;            if cx16.r15H & $80 !=0 {
+;                cx16.r14 <<= 1
+;                rol(cx16.r15)
+;                cx16.r15 ^= $04c1
+;                cx16.r14 ^= $1db7
+;            }
+;            else {
+;                cx16.r14 <<= 1
+;                rol(cx16.r15)
+;            }
+;        }
+
+    }
+
+    sub crc32_end() {
+        ; finalize the "streaming" crc32
+        ; result stored in cx16.r14 (low word) and cx16.r15 (high word)
+        cx16.r15 ^= $ffff
+        cx16.r14 ^= $ffff
     }
 }

compiler/res/prog8lib/pet32/syslib.p8

@@ -34,7 +34,7 @@ romsub $FFE4 = GETIN() clobbers(X,Y) -> bool @Pc, ubyte @ A     ; get a characte
 romsub $FFE7 = CLALL() clobbers(A,X)                            ; close all files
 romsub $FFEA = UDTIM() clobbers(A,X)                            ; update the software clock
 
-asmsub STOP2() clobbers(X) -> ubyte @A  {
+asmsub STOP2() clobbers(X) -> bool @A  {
     ; -- check if STOP key was pressed, returns true if so.  More convenient to use than STOP() because that only sets the carry status flag.
     %asm {{
         jsr  cbm.STOP
@@ -75,6 +75,15 @@ asmsub RDTIM16() clobbers(X) -> uword @AY {
     }}
 }
 
+asmsub kbdbuf_clear() {
+    ; -- convenience helper routine to clear the keyboard buffer
+    %asm {{
+-       jsr  GETIN
+        cmp  #0
+        bne  -
+        rts
+    }}
+}
 
 }
 
@@ -110,8 +119,15 @@ asmsub  init_system_phase2()  {
 asmsub  cleanup_at_exit() {
     ; executed when the main subroutine does rts
     %asm {{
+_exitcodeCarry = *+1
+        lda  #0
+        lsr  a
 _exitcode = *+1
         lda  #0        ; exit code possibly modified in exit()
+_exitcodeX = *+1
+        ldx  #0
+_exitcodeY = *+1
+        ldy  #0
         rts
     }}
 }
@@ -194,20 +210,14 @@ _loop       lda  P8ZP_SCRATCH_W1
             rts     ; nothing to copy
 
 _copyshort
-            ; decrease source and target pointers so we can simply index by Y
-            lda  P8ZP_SCRATCH_W1
-            bne  +
-            dec  P8ZP_SCRATCH_W1+1
-+           dec  P8ZP_SCRATCH_W1
-            lda  P8ZP_SCRATCH_W2
-            bne  +
-            dec  P8ZP_SCRATCH_W2+1
-+           dec  P8ZP_SCRATCH_W2
-
+            dey
+            beq  +
 -           lda  (P8ZP_SCRATCH_W1),y
             sta  (P8ZP_SCRATCH_W2),y
             dey
             bne  -
++           lda  (P8ZP_SCRATCH_W1),y
+            sta  (P8ZP_SCRATCH_W2),y
             rts
 
 _longcopy
@@ -357,6 +367,33 @@ save_SCRATCH_ZPWORD2	.word  0
         }}
     }
 
+    asmsub exit2(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y) {
+        ; -- immediately exit the program with result values in the A, X and Y registers.
+        %asm {{
+            sta  cleanup_at_exit._exitcode
+            stx  cleanup_at_exit._exitcodeX
+            sty  cleanup_at_exit._exitcodeY
+            ldx  prog8_lib.orig_stackpointer
+            txs
+            jmp  cleanup_at_exit
+        }}
+    }
+
+    asmsub exit3(ubyte resulta @A, ubyte resultx @X, ubyte resulty @Y, bool carry @Pc) {
+        ; -- immediately exit the program with result values in the A, X and Y registers, and the Carry flag in the status register.
+        %asm {{
+            sta  cleanup_at_exit._exitcode
+            lda  #0
+            rol  a
+            sta  cleanup_at_exit._exitcodeCarry
+            stx  cleanup_at_exit._exitcodeX
+            sty  cleanup_at_exit._exitcodeY
+            ldx  prog8_lib.orig_stackpointer
+            txs
+            jmp  cleanup_at_exit
+        }}
+    }
+
     inline asmsub progend() -> uword @AY {
         %asm {{
             lda  #<prog8_program_end

compiler/res/prog8lib/pet32/textio.p8

@@ -4,7 +4,7 @@
 
 %import syslib
 %import conv
-
+%import shared_textio_functions
 
 txt {
     %option no_symbol_prefixing, ignore_unused
@@ -14,58 +14,23 @@ const ubyte DEFAULT_HEIGHT = 25
 
 
 sub  clear_screen() {
-    txt.chrout(147)
+    chrout(147)
 }
 
 sub home() {
-    txt.chrout(19)
+    chrout(19)
 }
 
 sub nl() {
-    txt.chrout('\n')
+    chrout('\n')
 }
 
 sub spc() {
-    txt.chrout(' ')
+    chrout(' ')
 }
 
-
-
-asmsub get_column() -> ubyte @Y {
-    %asm {{
-        sec
-        jmp cbm.PLOT
-    }}
-}
-
-asmsub row(ubyte rownum @A) clobbers(A, X, Y) {
-    ; ---- set the cursor on the given row (starting with 0) on the current line
-    %asm {{
-        sec
-        jsr  cbm.PLOT
-        tax
-        clc
-        jmp  cbm.PLOT
-    }}
-}
-
-asmsub get_row() -> ubyte @X {
-    %asm {{
-        sec
-        jmp cbm.PLOT
-    }}
-}
-
-sub get_cursor(uword colptr, uword rowptr) {
-    %asm {{
-        sec
-        jsr cbm.PLOT
-        tya
-        ldy #$00
-        sta (colptr),y
-        txa
-        sta (rowptr),y
-    }}
+sub bell() {
+    chrout(7)
 }
 
 
@@ -73,7 +38,7 @@ asmsub  fill_screen (ubyte character @ A, ubyte color @ Y) clobbers(A)  {
 	; ---- fill the character screen with the given fill character. color is ignored on PET
 	%asm {{
 		jmp  clear_screenchars
-        }}
+    }}
 
 }
 

compiler/res/prog8lib/prog8_lib.asm

@@ -330,3 +330,102 @@ containment_wordarray	.proc
 _found		lda  #1
 		rts
 		.pend
+
+
+arraycopy_split_to_normal_words .proc
+	; P8ZP_SCRATCH_W1 = start of lsb array
+	; P8ZP_SCRATCH_W2 = start of msb array
+	; AY = start of normal word target array
+	; X = number of elements to copy
+                sta  _modlsb+1
+                sty  _modlsb+2
+                clc
+                adc  #1
+                bne  +
+                iny
++               sta  _modmsb+1
+                sty  _modmsb+2
+                ldy  #0
+-               lda  (P8ZP_SCRATCH_W1),y
+_modlsb         sta  $ffff       ; modified lsb store
+                lda  _modlsb+1
+                clc
+                adc  #2
+                sta  _modlsb+1
+                bcc  +
+                inc  _modlsb+2
++               lda  (P8ZP_SCRATCH_W2),y
+_modmsb         sta  $ffff       ; modified msb store
+                lda  _modmsb+1
+                clc
+                adc  #2
+                sta  _modmsb+1
+                bcc  +
+                inc  _modmsb+2
++               iny
+                dex
+                bne  -
+		rts
+		.pend
+
+
+arraycopy_normal_to_split_words .proc
+	; P8ZP_SCRATCH_W1 = start of target lsb array
+	; P8ZP_SCRATCH_W2 = start of target msb array
+	; AY = start of normal word source array
+	; X = number of elements to copy
+		sta  _modsrclsb+1
+		sty  _modsrclsb+2
+		clc
+		adc  #1
+		bne  +
+		iny
++		sta  _modsrcmsb+1
+		sty  _modsrcmsb+2
+		ldy  #0
+_modsrclsb      lda  $ffff      ; modified lsb read
+		sta  (P8ZP_SCRATCH_W1),y
+		lda  _modsrclsb+1
+		clc
+		adc  #2
+		sta  _modsrclsb+1
+		bcc  +
+		inc  _modsrclsb+2
++
+_modsrcmsb      lda  $ffff      ; modnfied msb read
+		sta  (P8ZP_SCRATCH_W2),y
+		lda  _modsrcmsb+1
+		clc
+		adc  #2
+		sta  _modsrcmsb+1
+		bcc  +
+		inc  _modsrcmsb+2
++		iny
+		dex
+		bne  _modsrclsb
+		rts
+		.pend
+
+memcopy_small   .proc
+		; copy up to a single page (256 bytes) of memory.
+		; note: only works for NON-OVERLAPPING memory regions!
+		; P8ZP_SCRATCH_W1 = from address
+		; P8ZP_SCRATCH_W2 = destination address
+		; Y = number of bytes to copy  (where 0 means 256)
+		cpy  #0
+		beq  _fullpage
+		dey
+		beq  _lastbyte
+_loop           lda  (P8ZP_SCRATCH_W1),y
+                sta  (P8ZP_SCRATCH_W2),y
+                dey
+                bne  _loop
+_lastbyte       lda  (P8ZP_SCRATCH_W1),y
+                sta  (P8ZP_SCRATCH_W2),y
+                rts
+_fullpage       lda  (P8ZP_SCRATCH_W1),y
+                sta  (P8ZP_SCRATCH_W2),y
+                dey
+                bne  _fullpage
+                rts
+		.pend

compiler/res/prog8lib/shared_string_functions.p8

@@ -0,0 +1,119 @@
+string {
+    ; the string functions shared across compiler targets
+    %option merge, no_symbol_prefixing, ignore_unused
+
+    sub strip(str s) {
+        ; -- gets rid of whitespace and other non-visible characters at the edges of the string
+        rstrip(s)
+        lstrip(s)
+    }
+
+    sub rstrip(str s) {
+        ; -- gets rid of whitespace and other non-visible characters at the end of the string
+        if s[0]==0
+            return
+        cx16.r0L = string.length(s)
+        do {
+            cx16.r0L--
+            cx16.r1L = s[cx16.r0L]
+        } until cx16.r0L==0 or string.isprint(cx16.r1L) and not string.isspace(cx16.r1L)
+        s[cx16.r0L+1] = 0
+    }
+
+    sub lstrip(str s) {
+        ; -- gets rid of whitespace and other non-visible characters at the start of the string (destructive)
+        cx16.r0 = lstripped(s)
+        if cx16.r0 != s
+            void string.copy(cx16.r0, s)
+    }
+
+    sub lstripped(str s) -> str {
+        ; -- returns pointer to first non-whitespace and non-visible character at the start of the string (non-destructive lstrip)
+        if s[0]==0
+            return s
+        cx16.r0L = 255
+        do {
+            cx16.r0L++
+            cx16.r1L = s[cx16.r0L]
+        } until cx16.r1L==0 or string.isprint(cx16.r1L) and not string.isspace(cx16.r1L)
+        return s+cx16.r0L
+    }
+
+    sub trim(str s) {
+        ; -- gets rid of whitespace characters at the edges of the string
+        rtrim(s)
+        ltrim(s)
+    }
+
+    sub rtrim(str s) {
+        ; -- gets rid of whitespace characters at the end of the string
+        if s[0]==0
+            return
+        cx16.r0L = string.length(s)
+        do {
+            cx16.r0L--
+            cx16.r1L = s[cx16.r0L]
+        } until cx16.r0L==0 or not string.isspace(cx16.r1L)
+        s[cx16.r0L+1] = 0
+    }
+
+    sub ltrim(str s) {
+        ; -- gets rid of whitespace characters at the start of the string (destructive)
+        cx16.r0 = ltrimmed(s)
+        if cx16.r0 != s
+            void string.copy(cx16.r0, s)
+    }
+
+    sub ltrimmed(str s) -> str {
+        ; -- return pointer to first non-whitespace character at the start of the string (non-destructive ltrim)
+        if s[0]==0
+            return s
+        cx16.r0L = 255
+        do {
+            cx16.r0L++
+            cx16.r1L = s[cx16.r0L]
+        } until not string.isspace(cx16.r1L)
+        return s+cx16.r0L
+    }
+
+    sub startswith(str st, str prefix) -> bool {
+        ubyte prefix_len = string.length(prefix)
+        ubyte str_len = string.length(st)
+        if prefix_len > str_len
+            return false
+        cx16.r9L = st[prefix_len]
+        st[prefix_len] = 0
+        cx16.r9H = string.compare(st, prefix) as ubyte
+        st[prefix_len] = cx16.r9L
+        return cx16.r9H==0
+    }
+
+    sub endswith(str st, str suffix) -> bool {
+        ubyte suffix_len = string.length(suffix)
+        ubyte str_len = string.length(st)
+        if suffix_len > str_len
+            return false
+        return string.compare(st + str_len - suffix_len, suffix) == 0
+    }
+
+    sub findstr(str haystack, str needle) -> ubyte {
+        ; searches for needle in haystack.
+        ; returns index in haystack where it first occurs, and Carry set,
+        ; or if needle doesn't occur in haystack it returns Carry clear and 255 (an invalid index.)
+        cx16.r2L = string.length(haystack)
+        cx16.r3L = string.length(needle)
+        if cx16.r3L <= cx16.r2L {
+            cx16.r2L = cx16.r2L-cx16.r3L+1
+            cx16.r3 = haystack
+            repeat cx16.r2L {
+                if string.startswith(cx16.r3, needle) {
+                    sys.set_carry()
+                    return cx16.r3-haystack as ubyte
+                }
+                cx16.r3++
+            }
+        }
+        sys.clear_carry()
+        return 255
+    }
+}

compiler/res/prog8lib/shared_textio_functions.p8

@@ -0,0 +1,46 @@
+txt {
+    ; the textio functions shared across compiler targets
+    %option merge, no_symbol_prefixing, ignore_unused
+
+    asmsub petscii2scr(ubyte petscii_char @A) -> ubyte @A {
+        ; -- convert petscii character to screencode
+        %asm {{
+            sta  P8ZP_SCRATCH_REG
+            lsr  a
+            lsr  a
+            lsr  a
+            lsr  a
+            lsr  a
+            tax
+            lda  _offsets,x
+            eor  P8ZP_SCRATCH_REG
+            rts
+_offsets    .byte  128, 0, 64, 32, 64, 192, 128, 128
+        }}
+    }
+
+    asmsub petscii2scr_str(str petscii_string @AY) {
+        ; -- convert petscii string to screencodes string
+        %asm {{
+            sta  P8ZP_SCRATCH_W1
+            sty  P8ZP_SCRATCH_W1+1
+            ldy  #0
+-           lda  (P8ZP_SCRATCH_W1),y
+            beq  +
+            jsr  petscii2scr
+            sta  (P8ZP_SCRATCH_W1),y
+            iny
+            bne  -
++           rts
+        }}
+}
+
+
+    sub print_bool(bool value) {
+        if value
+            txt.print("true")
+        else
+            txt.print("false")
+    }
+
+}

compiler/res/prog8lib/string.p8

@@ -1,5 +1,7 @@
 ; 0-terminated string manipulation routines.
 
+%import shared_string_functions
+
 string {
     %option no_symbol_prefixing, ignore_unused
 
@@ -127,7 +129,7 @@ _startloop	dey
 
     asmsub find(uword string @AY, ubyte character @X) -> ubyte @A, bool @Pc {
         ; Locates the first position of the given character in the string,
-        ; returns Carry set if found + index in A, or A=0 + Carry clear if not found.
+        ; returns Carry set if found + index in A, or Carry clear if not found (and A will be 255, an invalid index).
         %asm {{
 		; need to copy the the cx16 virtual registers to zeropage to make this run on C64...
 		sta  P8ZP_SCRATCH_W1
@@ -140,7 +142,7 @@ _startloop	dey
 		beq  _found
 		iny
 		bne  -
-_notfound	lda  #0
+_notfound	lda  #255
         clc
 		rts
 _found	tya
@@ -275,27 +277,7 @@ _done       rts
         }}
     }
 
-    sub startswith(str st, str prefix) -> bool {
-        ubyte prefix_len = length(prefix)
-        ubyte str_len = length(st)
-        if prefix_len > str_len
-            return false
-        cx16.r9L = st[prefix_len]
-        st[prefix_len] = 0
-        cx16.r9H = compare(st, prefix) as ubyte
-        st[prefix_len] = cx16.r9L
-        return cx16.r9H==0
-    }
-
-    sub endswith(str st, str suffix) -> bool {
-        ubyte suffix_len = length(suffix)
-        ubyte str_len = length(st)
-        if suffix_len > str_len
-            return false
-        return compare(st + str_len - suffix_len, suffix) == 0
-    }
-
-    asmsub pattern_match(str string @AY, str pattern @R0) clobbers(Y) -> ubyte @A {
+    asmsub pattern_match(str string @AY, str pattern @R0) clobbers(Y) -> bool @A {
 		%asm {{
 ; pattern matching of a string.
 ; Input:  cx16.r0:  A NUL-terminated, <255-length pattern
@@ -364,7 +346,6 @@ fail    clc             ; yes, no match found, return with c=0
 		}}
 	}
 
-
     asmsub hash(str string @R0) -> ubyte @A {
         ; experimental 8 bit hashing function.
         ; hash(-1)=179;  clear carry;  hash(i) = ROL hash(i-1)  XOR  string[i]

compiler/res/prog8lib/virtual/conv.p8

@@ -8,7 +8,7 @@ conv {
 
     str  string_out = "????????????????"       ; result buffer for the string conversion routines
 
-sub  str_ub0(ubyte value)   {
+sub  str_ub0(ubyte value) -> str {
     ; ---- convert the ubyte in A in decimal string form, with left padding 0s (3 positions total)
     ubyte hundreds = value / 100
     value -= hundreds*100
@@ -18,14 +18,16 @@ sub  str_ub0(ubyte value)   {
     string_out[1] = tens+'0'
     string_out[2] = value+'0'
     string_out[3] = 0
+    return string_out
 }
 
-sub  str_ub(ubyte value) {
+sub  str_ub(ubyte value) -> str {
     ; ---- convert the ubyte in A in decimal string form, without left padding 0s
     internal_str_ub(value, string_out)
+    return string_out
 }
 
-sub  str_b(byte value) {
+sub  str_b(byte value) -> str {
     ; ---- convert the byte in A in decimal string form, without left padding 0s
     uword out_ptr = &string_out
     if value<0 {
@@ -34,6 +36,7 @@ sub  str_b(byte value) {
         value = -value
     }
     internal_str_ub(value as ubyte, out_ptr)
+    return string_out
 }
 
 sub internal_str_ub(ubyte value, uword out_ptr) {
@@ -41,9 +44,9 @@ sub internal_str_ub(ubyte value, uword out_ptr) {
     value -= hundreds*100
     ubyte tens = value / 10
     value -= tens*10
-    if hundreds
+    if hundreds!=0
         goto output_hundreds
-    if tens
+    if tens!=0
         goto output_tens
     goto output_ones
 output_hundreds:
@@ -60,14 +63,15 @@ output_ones:
 
 str hex_digits = "0123456789abcdef"
 
-sub  str_ubhex  (ubyte value)  {
+sub  str_ubhex  (ubyte value) -> str {
     ; ---- convert the ubyte in A in hex string form
     string_out[0] = hex_digits[value>>4]
     string_out[1] = hex_digits[value&15]
     string_out[2] = 0
+    return string_out
 }
 
-sub  str_ubbin  (ubyte value)  {
+sub  str_ubbin  (ubyte value) -> str {
     ; ---- convert the ubyte in A in binary string form
     uword out_ptr = &string_out
     repeat 8 {
@@ -79,9 +83,10 @@ sub  str_ubbin  (ubyte value)  {
         out_ptr++
     }
     @(out_ptr) = 0
+    return string_out
 }
 
-sub  str_uwbin  (uword value) {
+sub  str_uwbin  (uword value) -> str {
     ; ---- convert the uword in A/Y in binary string form
     uword out_ptr = &string_out
     repeat 16 {
@@ -93,9 +98,10 @@ sub  str_uwbin  (uword value) {
         out_ptr++
     }
     @(out_ptr) = 0
+    return string_out
 }
 
-sub  str_uwhex  (uword value) {
+sub  str_uwhex  (uword value) -> str {
     ; ---- convert the uword in A/Y in hexadecimal string form (4 digits)
     ubyte bits = msb(value)
     string_out[0] = hex_digits[bits>>4]
@@ -104,9 +110,10 @@ sub  str_uwhex  (uword value) {
     string_out[2] = hex_digits[bits>>4]
     string_out[3] = hex_digits[bits&15]
     string_out[4] = 0
+    return string_out
 }
 
-sub  str_uw0  (uword value) {
+sub  str_uw0  (uword value) -> str {
     ; ---- convert the uword in A/Y in decimal string form, with left padding 0s (5 positions total)
     uword value2 = value/10
     ubyte digits = value-value2*10 as ubyte
@@ -124,14 +131,16 @@ sub  str_uw0  (uword value) {
     string_out[3] = tens+'0'
     string_out[4] = digits+'0'
     string_out[5] = 0
+    return string_out
 }
 
-sub  str_uw  (uword value)  {
+sub  str_uw  (uword value) -> str {
     ; ---- convert the uword in A/Y in decimal string form, without left padding 0s
     internal_str_uw(value, string_out)
+    return string_out
 }
 
-sub  str_w  (word value)  {
+sub  str_w  (word value) -> str {
     ; ---- convert the (signed) word in A/Y in decimal string form, without left padding 0's
     uword out_ptr = &string_out
     if value<0 {
@@ -140,6 +149,7 @@ sub  str_w  (word value)  {
         value = -value
     }
     internal_str_uw(value as uword, out_ptr)
+    return string_out
 }
 
 sub internal_str_uw(uword value, uword out_ptr) {
@@ -153,13 +163,13 @@ sub internal_str_uw(uword value, uword out_ptr) {
     ubyte thousands = value4-value5*10 as ubyte
     uword value6 = value5/10
     ubyte tenthousands = value5-value6*10 as ubyte
-    if tenthousands
+    if tenthousands!=0
         goto output_tenthousands
-    if thousands
+    if thousands!=0
         goto output_thousands
-    if hundreds
+    if hundreds!=0
         goto output_hundreds
-    if tens
+    if tens!=0
         goto output_tens
     goto output_ones
 output_tenthousands:

compiler/res/prog8lib/virtual/diskio.p8

@@ -1 +1,214 @@
-; there is no diskio yet for the VM target.
+; file I/O routines. (partially implemented)
+
+%import textio
+%import syslib
+
+diskio {
+    %option no_symbol_prefixing, ignore_unused
+
+    sub directory() -> bool {
+        ; -- Prints the directory contents to the screen. Returns success.
+        %ir {{
+            loadm.w r65534,diskio.load.filenameptr
+            loadm.w r65535,diskio.load.address_override
+            syscall 61 (): r0.b
+            returnr.b r0
+        }}
+    }
+
+    sub list_filenames(uword pattern_ptr, uword filenames_buffer, uword filenames_buf_size) -> ubyte {
+        ; -- fill the provided buffer with the names of the files on the disk (until buffer is full).
+        ;    Files in the buffer are separated by a 0 byte. You can provide an optional pattern to match against.
+        ;    After the last filename one additional 0 byte is placed to indicate the end of the list.
+        ;    Returns number of files (it skips 'dir' entries i.e. subdirectories).
+        ;    Also sets carry on exit: Carry clear = all files returned, Carry set = directory has more files that didn't fit in the buffer.
+        txt.print("@TODO: list_filenames\n")
+        sys.clear_carry()
+        return 0
+    }
+
+    ; ----- iterative file lister functions (uses the read io channel) -----
+
+    sub lf_start_list(uword pattern_ptr) -> bool {
+        ; -- start an iterative file listing with optional pattern matching.
+        ;    note: only a single iteration loop can be active at a time!
+        txt.print("@TODO: lf_start_list\n")
+        return false
+    }
+
+    sub lf_next_entry() -> bool {
+        ; -- retrieve the next entry from an iterative file listing session.
+        ;    results will be found in list_blocks, list_filename, and list_filetype.
+        ;    if it returns false though, there are no more entries (or an error occurred).
+        txt.print("@TODO: lf_next_entry\n")
+        return false
+    }
+
+    sub lf_end_list() {
+        txt.print("@TODO: lf_end_list\n")
+    }
+
+
+    ; ----- iterative file loader functions (uses the input io channel) -----
+
+    sub f_open(uword filenameptr) -> bool {
+        ; -- open a file for iterative reading with f_read
+        ;    note: only a single iteration loop can be active at a time!
+        ;    Returns true if the file is successfully opened and readable.
+        ;    No need to check status(), unlike f_open_w() !
+        ;    NOTE: the default input isn't yet set to this logical file, you must use reset_read_channel() to do this,
+        ;          if you're going to read from it yourself instead of using f_read()!
+
+        txt.print("@TODO: f_open\n")
+        return false
+    }
+
+    sub f_read(uword bufferpointer, uword num_bytes) -> uword {
+        ; -- read from the currently open file, up to the given number of bytes.
+        ;    returns the actual number of bytes read.  (checks for End-of-file and error conditions)
+        txt.print("@TODO: f_read\n")
+        return 0
+    }
+
+    sub f_read_all(uword bufferpointer) -> uword {
+        ; -- read the full contents of the file, returns number of bytes read.
+        txt.print("@TODO: f_read_all\n")
+        return 0
+    }
+
+    sub f_readline(uword bufptr) -> ubyte {
+        ; Routine to read text lines from a text file. Lines must be less than 255 characters.
+        ; Reads characters from the input file UNTIL a newline or return character (or EOF).
+        ; The line read will be 0-terminated in the buffer (and not contain the end of line character).
+        ; The length of the line is returned in Y. Note that an empty line is okay and is length 0!
+        ; I/O error status should be checked by the caller itself via READST() routine.
+        txt.print("@TODO: f_readline\n")
+        return 0
+    }
+
+    sub f_close() {
+        ; -- end an iterative file loading session (close channels).
+        txt.print("@TODO: f_close\n")
+    }
+
+
+    ; ----- iterative file writing functions (uses write io channel) -----
+
+    sub f_open_w(uword filenameptr) -> bool {
+        ; -- open a file for iterative writing with f_write
+        ;    WARNING: returns true if the open command was received by the device,
+        ;    but this can still mean the file wasn't successfully opened for writing!
+        ;    (for example, if it already exists). This is different than f_open()!
+        ;    To be 100% sure if this call was successful, you have to use status()
+        ;    and check the drive's status message!
+        txt.print("@TODO: f_open_w\n")
+        return false
+    }
+
+    sub f_write(uword bufferpointer, uword num_bytes) -> bool {
+        ; -- write the given number of bytes to the currently open file
+        txt.print("@TODO: f_write\n")
+        return false
+    }
+
+    sub f_close_w() {
+        ; -- end an iterative file writing session (close channels).
+        txt.print("@TODO: f_close_w\n")
+    }
+
+
+    ; ---- other functions ----
+
+    sub chdir(str path) {
+        ; -- change current directory.
+        txt.print("@TODO: chdir\n")
+    }
+
+    sub mkdir(str name) {
+        ; -- make a new subdirectory.
+        txt.print("@TODO: mkdir\n")
+    }
+
+    sub rmdir(str name) {
+        ; -- remove a subdirectory.
+        txt.print("@TODO: rmdir\n")
+    }
+
+    sub curdir() -> uword {
+        ; return current directory name or 0 if error
+        txt.print("@TODO: curdir\n")
+        return 0
+    }
+
+    sub status() -> str {
+        ; -- retrieve the disk drive's current status message
+        return "unknown"
+    }
+
+    sub save(uword filenameptr, uword start_address, uword savesize) -> bool {
+        %ir {{
+            load.b r65532,0
+            loadm.w r65533,diskio.save.filenameptr
+            loadm.w r65534,diskio.save.start_address
+            loadm.w r65535,diskio.save.savesize
+            syscall 58 (r65532.b, r65533.w, r65534.w, r65535.w): r0.b
+            returnr.b r0
+        }}
+    }
+
+    ; like save() but omits the 2 byte prg header.
+    sub save_raw(uword filenameptr, uword startaddress, uword savesize) -> bool {
+        %ir {{
+            load.b r65532,1
+            loadm.w r65533,diskio.save.filenameptr
+            loadm.w r65534,diskio.save.start_address
+            loadm.w r65535,diskio.save.savesize
+            syscall 58 (r65532.b, r65533.w, r65534.w, r65535.w): r0.b
+            returnr.b r0
+        }}
+    }
+
+    ; Use kernal LOAD routine to load the given program file in memory.
+    ; This is similar to Basic's  LOAD "filename",drive  /  LOAD "filename",drive,1
+    ; If you don't give an address_override, the location in memory is taken from the 2-byte file header.
+    ; If you specify a custom address_override, the first 2 bytes in the file are ignored
+    ; and the rest is loaded at the given location in memory.
+    ; Returns the end load address+1 if successful or 0 if a load error occurred.
+    sub load(uword filenameptr, uword address_override) -> uword {
+        %ir {{
+            loadm.w r65534,diskio.load.filenameptr
+            loadm.w r65535,diskio.load.address_override
+            syscall 56 (r65534.w, r65535.w): r0.w
+            returnr.w r0
+        }}
+    }
+
+    ; Identical to load(), but DOES INCLUDE the first 2 bytes in the file.
+    ; No program header is assumed in the file. Everything is loaded.
+    ; See comments on load() for more details.
+    sub load_raw(uword filenameptr, uword start_address) -> uword {
+        %ir {{
+            loadm.w r65534,diskio.load_raw.filenameptr
+            loadm.w r65535,diskio.load_raw.start_address
+            syscall 57 (r65534.w, r65535.w): r0.w
+            returnr.w r0
+        }}
+    }
+
+    sub delete(uword filenameptr) {
+        ; -- delete a file on the drive
+        %ir {{
+            loadm.w r65535,diskio.delete.filenameptr
+            syscall 59 (r65535.w)
+        }}
+    }
+
+    sub rename(uword oldfileptr, uword newfileptr) {
+        ; -- rename a file on the drive
+        %ir {{
+            loadm.w r65534,diskio.rename.oldfileptr
+            loadm.w r65535,diskio.rename.newfileptr
+            syscall 60 (r65534.w, r65535.w)
+        }}
+    }
+}

compiler/res/prog8lib/virtual/math.p8

@@ -176,6 +176,24 @@ math {
         }}
     }
 
+    sub randrange(ubyte n) -> ubyte {
+        ; -- return random number uniformly distributed from 0 to n-1 (compensates for divisibility bias)
+        cx16.r0H = 255 / n * n
+        do {
+            cx16.r0L = math.rnd()
+        } until cx16.r0L < cx16.r0H
+        return cx16.r0L % n
+    }
+
+    sub randrangew(uword n) -> uword {
+        ; -- return random number uniformly distributed from 0 to n-1 (compensates for divisibility bias)
+        cx16.r1 = 65535 / n * n
+        do {
+            cx16.r0 = math.rndw()
+        } until cx16.r0 < cx16.r1
+        return cx16.r0 % n
+    }
+
     sub rndseed(uword seed1, uword seed2) {
         ; -- reset the pseudo RNG's seed values. Defaults are: $a55a, $7653.
         %ir {{
@@ -190,7 +208,7 @@ math {
         ubyte result = 7
         ubyte compare = $80
         repeat {
-            if value&compare
+            if value&compare!=0
                 return result
             result--
             if_z
@@ -203,7 +221,7 @@ math {
         ubyte result = 15
         uword compare = $8000
         repeat {
-            if value&compare
+            if value&compare!=0
                 return result
             result--
             if_z
@@ -295,43 +313,87 @@ math {
 
     sub crc16(uword data, uword length) -> uword {
         ; calculates the CRC16 (XMODEM) checksum of the buffer.
-        cx16.r0 = 0  ; the crc value
-        repeat length {
-            cx16.r0H ^= @(data)
-            repeat 8 {
-                if cx16.r0H & $80
-                    cx16.r0 = (cx16.r0<<1)^$1021
-                else
-                    cx16.r0<<=1
-            }
-            data++
+        ; There are also "streaming" crc16_start/update/end routines below, that allow you to calculate crc32 for data that doesn't fit in a single memory block.
+        crc16_start()
+        cx16.r13 = data
+        cx16.r14 = data+length
+        while cx16.r13!=cx16.r14 {
+            crc16_update(@(cx16.r13))
+            cx16.r13++
         }
-        return cx16.r0
+        return crc16_end()
+    }
+
+    sub crc16_start() {
+        ; start the "streaming" crc16
+        ; note: tracks the crc16 checksum in cx16.r15!
+        ;       if your code uses that, it must save/restore it before calling this routine
+        cx16.r15 = 0
+    }
+
+    sub crc16_update(ubyte value) {
+        ; update the "streaming" crc16 with next byte value
+        ; note: tracks the crc16 checksum in cx16.r15!
+        ;       if your code uses that, it must save/restore it before calling this routine
+        cx16.r15H ^= value
+        repeat 8 {
+            if cx16.r15H & $80 !=0
+                cx16.r15 = (cx16.r15<<1)^$1021
+            else
+                cx16.r15<<=1
+        }
+    }
+
+    sub crc16_end() -> uword {
+        ; finalize the "streaming" crc16, returns resulting crc16 value
+        return cx16.r15
     }
 
     sub crc32(uword data, uword length) {
         ; Calculates the CRC-32 (POSIX) checksum of the buffer.
         ; because prog8 doesn't have 32 bits integers, we have to split up the calculation over 2 words.
-        ; result stored in cx16.r0 (low word) and cx16.r1 (high word)
-        cx16.r1 = 0
-        cx16.r0 = 0
-        repeat length {
-            cx16.r1H ^= @(data)
-            repeat 8 {
-                if cx16.r1H & $80 {
-                    cx16.r0 <<= 1
-                    rol(cx16.r1)
-                    cx16.r1 ^= $04c1
-                    cx16.r0 ^= $1db7
-                }
-                else {
-                    cx16.r0 <<= 1
-                    rol(cx16.r1)
-                }
-            }
-            data++
+        ; result stored in cx16.r14 (low word) and cx16.r15 (high word)
+        ; There are also "streaming" crc32_start/update/end routines below, that allow you to calculate crc32 for data that doesn't fit in a single memory block.
+        crc32_start()
+        cx16.r12 = data
+        cx16.r13 = data+length
+        while cx16.r12!=cx16.r13 {
+            crc32_update(@(cx16.r12))
+            cx16.r12++
         }
-        cx16.r1 ^= $ffff
-        cx16.r0 ^= $ffff
+        crc32_end()
+    }
+
+    sub crc32_start() {
+        ; start the "streaming" crc32
+        ; note: tracks the crc32 checksum in cx16.r14 and cx16.r15!
+        ;       if your code uses these, it must save/restore them before calling this routine
+        cx16.r14 = cx16.r15 = 0
+    }
+
+    sub crc32_update(ubyte value) {
+        ; update the "streaming" crc32 with next byte value
+        ; note: tracks the crc32 checksum in cx16.r14 and cx16.r15!
+        ;       if your code uses these, it must save/restore them before calling this routine
+        cx16.r15H ^= value
+        repeat 8 {
+            if cx16.r15H & $80 !=0 {
+                cx16.r14 <<= 1
+                rol(cx16.r15)
+                cx16.r15 ^= $04c1
+                cx16.r14 ^= $1db7
+            }
+            else {
+                cx16.r14 <<= 1
+                rol(cx16.r15)
+            }
+        }
+    }
+
+    sub crc32_end() {
+        ; finalize the "streaming" crc32
+        ; result stored in cx16.r14 (low word) and cx16.r15 (high word)
+        cx16.r15 ^= $ffff
+        cx16.r14 ^= $ffff
     }
 }

compiler/res/prog8lib/virtual/monogfx.p8

@@ -12,13 +12,17 @@ monogfx {
     ; read-only control variables:
     uword width = 0
     uword height = 0
-    bool dont_stipple_flag = true            ; set to false to enable stippling mode
+    ubyte mode
+    const ubyte MODE_NORMAL  = %00000000
+    const ubyte MODE_STIPPLE = %00000001
+    const ubyte MODE_INVERT  = %00000010
 
     sub lores() {
         ; enable 320*240 bitmap mode
         sys.gfx_enable(0)
         width = 320
         height = 240
+        mode = MODE_NORMAL
         clear_screen(0)
     }
 
@@ -27,6 +31,7 @@ monogfx {
         sys.gfx_enable(1)
         width = 640
         height = 480
+        mode = MODE_NORMAL
         clear_screen(0)
     }
 
@@ -34,15 +39,14 @@ monogfx {
         ; back to normal text mode
     }
 
-    sub clear_screen(ubyte color) {
-        stipple(false)
-        if color
-            color=255
-        sys.gfx_clear(color)
+    sub drawmode(ubyte dm) {
+        mode = dm
     }
 
-    sub stipple(bool enable) {
-        dont_stipple_flag = not enable
+    sub clear_screen(ubyte color) {
+        if color!=0
+            color=255
+        sys.gfx_clear(color)
     }
 
     sub rect(uword xx, uword yy, uword rwidth, uword rheight, bool draw) {
@@ -126,17 +130,17 @@ monogfx {
         }
 
         word @zp d = 0
-        cx16.r1L = true      ; 'positive_ix'
+        cx16.r1L = 1  ; true      ; 'positive_ix'
         if dx < 0 {
             dx = -dx
-            cx16.r1L = false
+            cx16.r1L = 0 ; false
         }
         word @zp dx2 = dx*2
         word @zp dy2 = dy*2
         cx16.r14 = x1       ; internal plot X
 
         if dx >= dy {
-            if cx16.r1L {
+            if cx16.r1L!=0 {
                 repeat {
                     plot(cx16.r14, y1, draw)
                     if cx16.r14==x2
@@ -163,7 +167,7 @@ monogfx {
             }
         }
         else {
-            if cx16.r1L {
+            if cx16.r1L!=0 {
                 repeat {
                     plot(cx16.r14, y1, draw)
                     if y1 == y2
@@ -337,13 +341,19 @@ monogfx {
 
     sub plot(uword @zp xx, uword @zp yy, bool @zp draw) {
         if draw {
-            if dont_stipple_flag
-                sys.gfx_plot(xx, yy, 255)
-            else {
-                if (xx ^ yy)&1
+            when mode {
+                MODE_NORMAL -> {
                     sys.gfx_plot(xx, yy, 255)
-                else
-                    sys.gfx_plot(xx, yy, 0)
+                }
+                MODE_STIPPLE -> {
+                    if (xx ^ yy)&1 !=0
+                        sys.gfx_plot(xx, yy, 255)
+                    else
+                        sys.gfx_plot(xx, yy, 0)
+                }
+                MODE_INVERT -> {
+                    sys.gfx_plot(xx, yy, 255 ^ sys.gfx_getpixel(xx, yy))
+                }
             }
         }
         else
@@ -359,8 +369,8 @@ monogfx {
         plot(xx, yy, draw)
     }
 
-    sub pget(uword @zp xx, uword yy) -> ubyte {
-        return sys.gfx_getpixel(xx, yy)
+    sub pget(uword @zp xx, uword yy) -> bool {
+        return sys.gfx_getpixel(xx, yy) as bool
     }
 
     sub fill(uword x, uword y, bool draw) {
@@ -399,7 +409,7 @@ monogfx {
             dy = stack_dy[cx16.r12L]
             yy+=dy
         }
-        cx16.r11L = pget(xx as uword, yy as uword)        ; old_color
+        cx16.r11L = pget(xx as uword, yy as uword) as ubyte       ; old_color
         if cx16.r11L == cx16.r10L
             return
         if xx<0 or xx>width-1 or yy<0 or yy>height-1
@@ -407,16 +417,16 @@ monogfx {
         push_stack(xx, xx, yy, 1)
         push_stack(xx, xx, yy + 1, -1)
         word left = 0
-        while cx16.r12L {
+        while cx16.r12L!=0 {
             pop_stack()
             xx = x1
             while xx >= 0 {
-                if pget(xx as uword, yy as uword) != cx16.r11L
+                if pget(xx as uword, yy as uword) as ubyte != cx16.r11L
                     break
                 xx--
             }
             if x1!=xx
-                horizontal_line(xx as uword+1, yy as uword, x1-xx as uword, cx16.r10L)
+                horizontal_line(xx as uword+1, yy as uword, x1-xx as uword, cx16.r10L as bool)
             else
                 goto skip
 
@@ -426,14 +436,14 @@ monogfx {
             xx = x1 + 1
 
             do {
-                cx16.r9 = xx
+                cx16.r9 = xx as uword
                 while xx <= width-1 {
-                    if pget(xx as uword, yy as uword) != cx16.r11L
+                    if pget(xx as uword, yy as uword) as ubyte != cx16.r11L
                         break
                     xx++
                 }
                 if cx16.r9!=xx
-                    horizontal_line(cx16.r9, yy as uword, (xx as uword)-cx16.r9, cx16.r10L)
+                    horizontal_line(cx16.r9, yy as uword, (xx as uword)-cx16.r9, cx16.r10L as bool)
 
                 push_stack(left, xx - 1, yy, dy)
                 if xx > x2 + 1
@@ -441,7 +451,7 @@ monogfx {
 skip:
                 xx++
                 while xx <= x2 {
-                    if pget(xx as uword, yy as uword) == cx16.r11L
+                    if pget(xx as uword, yy as uword) as ubyte == cx16.r11L
                         break
                     xx++
                 }
@@ -457,7 +467,7 @@ skip:
         ; TODO vm bitmap charset
     }
 
-    sub text(uword @zp xx, uword yy, bool draw, uword sctextptr) {
+    sub text(uword @zp xx, uword yy, bool draw, str sctextptr) {
         ; -- Write some text at the given pixel position. The text string must be in screencode encoding (not petscii!).
         ;    You must also have called text_charset() first to select and prepare the character set to use.
         ; TODO vm bitmap charset