Files
Toy/source/toy_compiler.c

1306 lines
49 KiB
C

#include "toy_compiler.h"
#include "toy_memory.h"
#include "toy_literal.h"
#include "toy_literal_array.h"
#include "toy_literal_dictionary.h"
#include "toy_console_colors.h"
#include <stdio.h>
#include <string.h>
void Toy_initCompiler(Toy_Compiler* compiler) {
Toy_initLiteralArray(&compiler->literalCache);
compiler->bytecode = NULL;
compiler->capacity = 0;
compiler->count = 0;
compiler->panic = false;
}
//separated out, so it can be recursive
static int writeLiteralTypeToCache(Toy_LiteralArray* literalCache, Toy_Literal literal) {
bool shouldFree = false;
//if it's a compound type, recurse and store the results
if (TOY_AS_TYPE(literal).typeOf == TOY_LITERAL_ARRAY || TOY_AS_TYPE(literal).typeOf == TOY_LITERAL_DICTIONARY) {
//I don't like storing types in an array, but it's the easiest and most straight forward method
Toy_LiteralArray* store = TOY_ALLOCATE(Toy_LiteralArray, 1);
Toy_initLiteralArray(store);
//store the base literal in the store
Toy_pushLiteralArray(store, literal);
for (int i = 0; i < TOY_AS_TYPE(literal).count; i++) {
//write the values to the cache, and the indexes to the store
int subIndex = writeLiteralTypeToCache(literalCache, ((Toy_Literal*)(TOY_AS_TYPE(literal).subtypes))[i]);
Toy_Literal lit = TOY_TO_INTEGER_LITERAL(subIndex);
Toy_pushLiteralArray(store, lit);
Toy_freeLiteral(lit);
}
//push the store to the cache, tweaking the type
shouldFree = true;
literal = TOY_TO_ARRAY_LITERAL(store);
literal.type = TOY_LITERAL_TYPE_INTERMEDIATE; //NOTE: tweaking the type usually isn't a good idea
}
//optimisation: check if exactly this literal array exists
int index = Toy_findLiteralIndex(literalCache, literal);
if (index < 0) {
index = Toy_pushLiteralArray(literalCache, literal);
}
if (shouldFree) {
Toy_freeLiteral(literal);
}
return index;
}
static int writeNodeCompoundToCache(Toy_Compiler* compiler, Toy_ASTNode* node) {
int index = -1;
//for both, stored as an array
Toy_LiteralArray* store = TOY_ALLOCATE(Toy_LiteralArray, 1);
Toy_initLiteralArray(store);
//emit an array or a dictionary definition
if (node->compound.literalType == TOY_LITERAL_DICTIONARY) {
//ensure each literal key and value are in the cache, individually
for (int i = 0; i < node->compound.count; i++) {
//keys
switch(node->compound.nodes[i].pair.left->type) {
case TOY_AST_NODE_LITERAL: {
//keys are literals
int key = Toy_findLiteralIndex(&compiler->literalCache, node->compound.nodes[i].pair.left->atomic.literal);
if (key < 0) {
key = Toy_pushLiteralArray(&compiler->literalCache, node->compound.nodes[i].pair.left->atomic.literal);
}
Toy_Literal literal = TOY_TO_INTEGER_LITERAL(key);
Toy_pushLiteralArray(store, literal);
Toy_freeLiteral(literal);
}
break;
case TOY_AST_NODE_COMPOUND: {
int key = writeNodeCompoundToCache(compiler, node->compound.nodes[i].pair.left);
Toy_Literal literal = TOY_TO_INTEGER_LITERAL(key);
Toy_pushLiteralArray(store, literal);
Toy_freeLiteral(literal);
}
break;
default:
fprintf(stderr, TOY_CC_ERROR "[internal] Unrecognized key node type in writeNodeCompoundToCache()\n" TOY_CC_RESET);
return -1;
}
//values
switch(node->compound.nodes[i].pair.right->type) {
case TOY_AST_NODE_LITERAL: {
//values are literals
int val = Toy_findLiteralIndex(&compiler->literalCache, node->compound.nodes[i].pair.right->atomic.literal);
if (val < 0) {
val = Toy_pushLiteralArray(&compiler->literalCache, node->compound.nodes[i].pair.right->atomic.literal);
}
Toy_Literal literal = TOY_TO_INTEGER_LITERAL(val);
Toy_pushLiteralArray(store, literal);
Toy_freeLiteral(literal);
}
break;
case TOY_AST_NODE_COMPOUND: {
int val = writeNodeCompoundToCache(compiler, node->compound.nodes[i].pair.right);
Toy_Literal literal = TOY_TO_INTEGER_LITERAL(val);
Toy_pushLiteralArray(store, literal);
Toy_freeLiteral(literal);
}
break;
default:
fprintf(stderr, TOY_CC_ERROR "[internal] Unrecognized value node type in writeNodeCompoundToCache()\n" TOY_CC_RESET);
return -1;
}
}
//push the store to the cache, with instructions about how pack it
Toy_Literal literal = TOY_TO_DICTIONARY_LITERAL((Toy_LiteralDictionary*)store); //cast from array to dict, because it's intermediate
literal.type = TOY_LITERAL_DICTIONARY_INTERMEDIATE; //god damn it - nested in a dictionary
index = Toy_pushLiteralArray(&compiler->literalCache, literal);
Toy_freeLiteral(literal);
}
else if (node->compound.literalType == TOY_LITERAL_ARRAY) {
//ensure each literal value is in the cache, individually
for (int i = 0; i < node->compound.count; i++) {
switch(node->compound.nodes[i].type) {
case TOY_AST_NODE_LITERAL: {
//values
int val = Toy_findLiteralIndex(&compiler->literalCache, node->compound.nodes[i].atomic.literal);
if (val < 0) {
val = Toy_pushLiteralArray(&compiler->literalCache, node->compound.nodes[i].atomic.literal);
}
Toy_Literal literal = TOY_TO_INTEGER_LITERAL(val);
Toy_pushLiteralArray(store, literal);
Toy_freeLiteral(literal);
}
break;
case TOY_AST_NODE_COMPOUND: {
int val = writeNodeCompoundToCache(compiler, &node->compound.nodes[i]);
Toy_Literal literal = TOY_TO_INTEGER_LITERAL(val);
index = Toy_pushLiteralArray(store, literal);
Toy_freeLiteral(literal);
}
break;
default:
fprintf(stderr, TOY_CC_ERROR "[internal] Unrecognized node type in writeNodeCompoundToCache()\n" TOY_CC_RESET);
return -1;
}
}
//push the store to the cache, with instructions about how pack it
Toy_Literal literal = TOY_TO_ARRAY_LITERAL(store);
literal.type = TOY_LITERAL_ARRAY_INTERMEDIATE; //god damn it - nested in an array
index = Toy_pushLiteralArray(&compiler->literalCache, literal);
Toy_freeLiteral(literal);
}
else {
fprintf(stderr, TOY_CC_ERROR "[internal] Unrecognized compound type in writeNodeCompoundToCache()\n" TOY_CC_RESET);
return -1;
}
return index;
}
static int writeNodeCollectionToCache(Toy_Compiler* compiler, Toy_ASTNode* node) {
Toy_LiteralArray* store = TOY_ALLOCATE(Toy_LiteralArray, 1);
Toy_initLiteralArray(store);
//ensure each literal value is in the cache, individually
for (int i = 0; i < node->fnCollection.count; i++) {
switch(node->fnCollection.nodes[i].type) {
case TOY_AST_NODE_VAR_DECL: {
//write each piece of the declaration to the cache
int identifierIndex = Toy_pushLiteralArray(&compiler->literalCache, node->fnCollection.nodes[i].varDecl.identifier); //store without duplication optimisation
int typeIndex = writeLiteralTypeToCache(&compiler->literalCache, node->fnCollection.nodes[i].varDecl.typeLiteral);
Toy_Literal identifierLiteral = TOY_TO_INTEGER_LITERAL(identifierIndex);
Toy_pushLiteralArray(store, identifierLiteral);
Toy_freeLiteral(identifierLiteral);
Toy_Literal typeLiteral = TOY_TO_INTEGER_LITERAL(typeIndex);
Toy_pushLiteralArray(store, typeLiteral);
Toy_freeLiteral(typeLiteral);
}
break;
case TOY_AST_NODE_LITERAL: {
//write each piece of the declaration to the cache
int typeIndex = writeLiteralTypeToCache(&compiler->literalCache, node->fnCollection.nodes[i].atomic.literal);
Toy_Literal typeLiteral = TOY_TO_INTEGER_LITERAL(typeIndex);
Toy_pushLiteralArray(store, typeLiteral);
Toy_freeLiteral(typeLiteral);
}
break;
default:
fprintf(stderr, TOY_CC_ERROR "[internal] Unrecognized node type in writeNodeCollectionToCache()\n" TOY_CC_RESET);
return -1;
}
}
//store the store
Toy_Literal literal = TOY_TO_ARRAY_LITERAL(store);
int storeIndex = Toy_pushLiteralArray(&compiler->literalCache, literal);
Toy_freeLiteral(literal);
return storeIndex;
}
static int writeLiteralToCompiler(Toy_Compiler* compiler, Toy_Literal literal) {
//get the index
int index = Toy_findLiteralIndex(&compiler->literalCache, literal);
if (index < 0) {
if (TOY_IS_TYPE(literal)) {
//check for the type literal as value
index = writeLiteralTypeToCache(&compiler->literalCache, literal);
}
else {
index = Toy_pushLiteralArray(&compiler->literalCache, literal);
}
}
//push the literal to the bytecode
if (index >= 256) {
//push a "long" index
compiler->bytecode[compiler->count++] = TOY_OP_LITERAL_LONG; //1 byte
memcpy(compiler->bytecode + compiler->count, &index, sizeof(unsigned short)); //2 bytes
compiler->count += sizeof(unsigned short);
}
else {
//push the index
compiler->bytecode[compiler->count++] = TOY_OP_LITERAL; //1 byte
compiler->bytecode[compiler->count++] = (unsigned char)index; //1 byte
}
return index;
}
//NOTE: jumpOfsets are included, because function arg and return indexes are embedded in the code body i.e. need to include their sizes in the jump
//NOTE: rootNode should NOT include groupings and blocks
static Toy_Opcode Toy_writeCompilerWithJumps(Toy_Compiler* compiler, Toy_ASTNode* node, void* breakAddressesPtr, void* continueAddressesPtr, int jumpOffsets, Toy_ASTNode* rootNode) {
//grow if the bytecode space is too small
if (compiler->count + 32 > compiler->capacity) {
int oldCapacity = compiler->capacity;
compiler->capacity = TOY_GROW_CAPACITY_FAST(oldCapacity);
compiler->bytecode = TOY_GROW_ARRAY(unsigned char, compiler->bytecode, oldCapacity, compiler->capacity);
}
//determine node type
switch(node->type) {
case TOY_AST_NODE_ERROR: {
fprintf(stderr, TOY_CC_ERROR "[internal] TOY_AST_NODEERROR encountered in Toy_writeCompilerWithJumps()\n" TOY_CC_RESET);
compiler->bytecode[compiler->count++] = TOY_OP_EOF; //1 byte
}
break;
case TOY_AST_NODE_LITERAL: {
writeLiteralToCompiler(compiler, node->atomic.literal);
}
break;
case TOY_AST_NODE_UNARY: {
//pass to the child node, then embed the unary command (print, negate, etc.)
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->unary.child, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
compiler->bytecode[compiler->count++] = (unsigned char)node->unary.opcode; //1 byte
}
break;
//all infixes come here
case TOY_AST_NODE_BINARY: {
//pass to the child nodes, then embed the binary command (math, etc.)
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->binary.left, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
//special case for when indexing and assigning
if (override != TOY_OP_EOF && node->binary.opcode >= TOY_OP_VAR_ASSIGN && node->binary.opcode <= TOY_OP_VAR_MODULO_ASSIGN) {
Toy_writeCompilerWithJumps(compiler, node->binary.right, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
//Special case if there's an index on both sides of the sign, just set it as indexing
if (node->binary.left->type == TOY_AST_NODE_BINARY && node->binary.right->type == TOY_AST_NODE_BINARY && node->binary.left->binary.opcode == TOY_OP_INDEX && node->binary.right->binary.opcode == TOY_OP_INDEX) {
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_INDEX;
}
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_INDEX_ASSIGN; //1 byte WARNING: enum trickery
compiler->bytecode[compiler->count++] = (unsigned char)node->binary.opcode; //1 byte
return TOY_OP_EOF;
}
//compensate for... yikes
if (override != TOY_OP_EOF) {
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//return this if...
Toy_Opcode ret = Toy_writeCompilerWithJumps(compiler, node->binary.right, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (node->binary.opcode == TOY_OP_INDEX && rootNode->type == TOY_AST_NODE_BINARY && (rootNode->binary.opcode >= TOY_OP_VAR_ASSIGN && rootNode->binary.opcode <= TOY_OP_VAR_MODULO_ASSIGN) && rootNode->binary.right != node) { //range-based check for assignment type; make sure the index is on the left of the assignment symbol
return TOY_OP_INDEX_ASSIGN_INTERMEDIATE;
}
//loopy logic - if opcode == index or dot
if (node->binary.opcode == TOY_OP_INDEX || node->binary.opcode == TOY_OP_DOT) {
return node->binary.opcode;
}
if (ret != TOY_OP_EOF && (node->binary.opcode == TOY_OP_VAR_ASSIGN || node->binary.opcode == TOY_OP_AND || node->binary.opcode == TOY_OP_OR || (node->binary.opcode >= TOY_OP_COMPARE_EQUAL && node->binary.opcode <= TOY_OP_INVERT))) {
compiler->bytecode[compiler->count++] = (unsigned char)ret; //1 byte
ret = TOY_OP_EOF; //untangle in this case
}
compiler->bytecode[compiler->count++] = (unsigned char)node->binary.opcode; //1 byte
return ret;
}
break;
case TOY_AST_NODE_TERNARY: {
// TODO: a ?: b;
//process the condition
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->ternary.condition, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//cache the point to insert the jump distance at
compiler->bytecode[compiler->count++] = TOY_OP_IF_FALSE_JUMP; //1 byte
int jumpToElse = compiler->count;
compiler->count += sizeof(unsigned short); //2 bytes
//write the then path
override = Toy_writeCompilerWithJumps(compiler, node->pathIf.thenPath, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
int jumpToEnd = 0;
//insert jump to end
compiler->bytecode[compiler->count++] = TOY_OP_JUMP; //1 byte
jumpToEnd = compiler->count;
compiler->count += sizeof(unsigned short); //2 bytes
//update the jumpToElse to point here
unsigned short tmpVal = compiler->count + jumpOffsets;
memcpy(compiler->bytecode + jumpToElse, &tmpVal, sizeof(tmpVal)); //2 bytes
//write the else path
Toy_Opcode override2 = Toy_writeCompilerWithJumps(compiler, node->pathIf.elsePath, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override2 != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override2; //1 byte
}
//update the jumpToEnd to point here
tmpVal = compiler->count + jumpOffsets;
memcpy(compiler->bytecode + jumpToEnd, &tmpVal, sizeof(tmpVal)); //2 bytes
}
break;
case TOY_AST_NODE_GROUPING: {
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_GROUPING_BEGIN; //1 byte
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->grouping.child, breakAddressesPtr, continueAddressesPtr, jumpOffsets, node->grouping.child);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_GROUPING_END; //1 byte
}
break;
case TOY_AST_NODE_BLOCK: {
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_SCOPE_BEGIN; //1 byte
for (int i = 0; i < node->block.count; i++) {
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, &(node->block.nodes[i]), breakAddressesPtr, continueAddressesPtr, jumpOffsets, &(node->block.nodes[i]));
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
}
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_SCOPE_END; //1 byte
}
break;
case TOY_AST_NODE_COMPOUND: {
int index = writeNodeCompoundToCache(compiler, node);
if (index < 0) {
compiler->panic = true;
return TOY_OP_EOF;
}
//push the node opcode to the bytecode
if (index >= 256) {
//push a "long" index
compiler->bytecode[compiler->count++] = TOY_OP_LITERAL_LONG; //1 byte
memcpy(compiler->bytecode + compiler->count, &index, sizeof(unsigned short));
compiler->count += sizeof(unsigned short);
}
else {
//push the index
compiler->bytecode[compiler->count++] = TOY_OP_LITERAL; //1 byte
compiler->bytecode[compiler->count++] = (unsigned char)index; //1 byte
}
}
break;
case TOY_AST_NODE_PAIR:
fprintf(stderr, TOY_CC_ERROR "[internal] TOY_AST_NODEPAIR encountered in Toy_writeCompilerWithJumps()\n" TOY_CC_RESET);
compiler->bytecode[compiler->count++] = TOY_OP_EOF; //1 byte
break;
case TOY_AST_NODE_VAR_DECL: {
//first, embed the expression (leaves it on the stack)
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->varDecl.expression, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//write each piece of the declaration to the bytecode
int identifierIndex = Toy_findLiteralIndex(&compiler->literalCache, node->varDecl.identifier);
if (identifierIndex < 0) {
identifierIndex = Toy_pushLiteralArray(&compiler->literalCache, node->varDecl.identifier);
}
int typeIndex = writeLiteralTypeToCache(&compiler->literalCache, node->varDecl.typeLiteral);
//embed the info into the bytecode
if (identifierIndex >= 256 || typeIndex >= 256) {
//push a "long" declaration
compiler->bytecode[compiler->count++] = TOY_OP_VAR_DECL_LONG; //1 byte
*((unsigned short*)(compiler->bytecode + compiler->count)) = (unsigned short)identifierIndex; //2 bytes
compiler->count += sizeof(unsigned short);
*((unsigned short*)(compiler->bytecode + compiler->count)) = (unsigned short)typeIndex; //2 bytes
compiler->count += sizeof(unsigned short);
}
else {
//push a declaration
compiler->bytecode[compiler->count++] = TOY_OP_VAR_DECL; //1 byte
compiler->bytecode[compiler->count++] = (unsigned char)identifierIndex; //1 byte
compiler->bytecode[compiler->count++] = (unsigned char)typeIndex; //1 byte
}
}
break;
case TOY_AST_NODE_FN_DECL: {
//run a compiler over the function
Toy_Compiler* fnCompiler = TOY_ALLOCATE(Toy_Compiler, 1);
Toy_initCompiler(fnCompiler);
Toy_writeCompiler(fnCompiler, node->fnDecl.arguments); //can be empty, but not NULL
Toy_writeCompiler(fnCompiler, node->fnDecl.returns); //can be empty, but not NULL
Toy_Opcode override = Toy_writeCompilerWithJumps(fnCompiler, node->fnDecl.block, NULL, NULL, -4, rootNode); //can be empty, but not NULL
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//adopt the panic state if anything happened
if (fnCompiler->panic) {
compiler->panic = true;
}
//create the function in the literal cache (by storing the compiler object)
Toy_Literal fnLiteral = TOY_TO_FUNCTION_LITERAL(fnCompiler, 0);
fnLiteral.type = TOY_LITERAL_FUNCTION_INTERMEDIATE; //NOTE: changing type
//push the name
int identifierIndex = Toy_findLiteralIndex(&compiler->literalCache, node->fnDecl.identifier);
if (identifierIndex < 0) {
identifierIndex = Toy_pushLiteralArray(&compiler->literalCache, node->fnDecl.identifier);
}
//push to function (functions are never equal)
int fnIndex = Toy_pushLiteralArray(&compiler->literalCache, fnLiteral);
//embed the info into the bytecode
if (identifierIndex >= 256 || fnIndex >= 256) {
//push a "long" declaration
compiler->bytecode[compiler->count++] = TOY_OP_FN_DECL_LONG; //1 byte
*((unsigned short*)(compiler->bytecode + compiler->count)) = (unsigned short)identifierIndex; //2 bytes
compiler->count += sizeof(unsigned short);
*((unsigned short*)(compiler->bytecode + compiler->count)) = (unsigned short)fnIndex; //2 bytes
compiler->count += sizeof(unsigned short);
}
else {
//push a declaration
compiler->bytecode[compiler->count++] = TOY_OP_FN_DECL; //1 byte
compiler->bytecode[compiler->count++] = (unsigned char)identifierIndex; //1 byte
compiler->bytecode[compiler->count++] = (unsigned char)fnIndex; //1 byte
}
}
break;
case TOY_AST_NODE_FN_COLLECTION: {
//embed these in the bytecode...
unsigned short index = (unsigned short)writeNodeCollectionToCache(compiler, node);
if (index == (unsigned short)-1) {
compiler->panic = true;
return TOY_OP_EOF;
}
memcpy(compiler->bytecode + compiler->count, &index, sizeof(index));
compiler->count += sizeof(unsigned short);
}
break;
case TOY_AST_NODE_FN_CALL: {
//NOTE: assume the function definition/name is above us
for (int i = 0; i < node->fnCall.arguments->fnCollection.count; i++) { //reverse order, to count from the beginning in the interpreter
//sub-calls
if (node->fnCall.arguments->fnCollection.nodes[i].type != TOY_AST_NODE_LITERAL) {
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, &node->fnCall.arguments->fnCollection.nodes[i], breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
continue;
}
//write each argument to the bytecode
int argumentsIndex = Toy_findLiteralIndex(&compiler->literalCache, node->fnCall.arguments->fnCollection.nodes[i].atomic.literal);
if (argumentsIndex < 0) {
argumentsIndex = Toy_pushLiteralArray(&compiler->literalCache, node->fnCall.arguments->fnCollection.nodes[i].atomic.literal);
}
//push the node opcode to the bytecode
if (argumentsIndex >= 256) {
//push a "long" index
compiler->bytecode[compiler->count++] = TOY_OP_LITERAL_LONG; //1 byte
*((unsigned short*)(compiler->bytecode + compiler->count)) = (unsigned short)argumentsIndex; //2 bytes
compiler->count += sizeof(unsigned short);
}
else {
//push the index
compiler->bytecode[compiler->count++] = TOY_OP_LITERAL; //1 byte
compiler->bytecode[compiler->count++] = (unsigned char)argumentsIndex; //1 byte
}
}
//push the argument COUNT to the top of the stack
Toy_Literal argumentsCountLiteral = TOY_TO_INTEGER_LITERAL(node->fnCall.argumentCount); //argumentCount is set elsewhere to support dot operator
int argumentsCountIndex = Toy_findLiteralIndex(&compiler->literalCache, argumentsCountLiteral);
if (argumentsCountIndex < 0) {
argumentsCountIndex = Toy_pushLiteralArray(&compiler->literalCache, argumentsCountLiteral);
}
Toy_freeLiteral(argumentsCountLiteral);
if (argumentsCountIndex >= 256) {
//push a "long" index
compiler->bytecode[compiler->count++] = TOY_OP_LITERAL_LONG; //1 byte
*((unsigned short*)(compiler->bytecode + compiler->count)) = (unsigned short)argumentsCountIndex; //2 bytes
compiler->count += sizeof(unsigned short);
}
else {
//push the index
compiler->bytecode[compiler->count++] = TOY_OP_LITERAL; //1 byte
compiler->bytecode[compiler->count++] = (unsigned char)argumentsCountIndex; //1 byte
}
//call the function
//DO NOT call the collection, this is done in binary
}
break;
case TOY_AST_NODE_IF: {
//process the condition
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->pathIf.condition, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//cache the point to insert the jump distance at
compiler->bytecode[compiler->count++] = TOY_OP_IF_FALSE_JUMP; //1 byte
int jumpToElse = compiler->count;
compiler->count += sizeof(unsigned short); //2 bytes
//write the then path
override = Toy_writeCompilerWithJumps(compiler, node->pathIf.thenPath, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
int jumpToEnd = 0;
if (node->pathIf.elsePath) {
//insert jump to end
compiler->bytecode[compiler->count++] = TOY_OP_JUMP; //1 byte
jumpToEnd = compiler->count;
compiler->count += sizeof(unsigned short); //2 bytes
}
//update the jumpToElse to point here
unsigned short tmpVal = compiler->count + jumpOffsets;
memcpy(compiler->bytecode + jumpToElse, &tmpVal, sizeof(tmpVal)); //2 bytes
if (node->pathIf.elsePath) {
//if there's an else path, write it and
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->pathIf.elsePath, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//update the jumpToEnd to point here
tmpVal = compiler->count + jumpOffsets;
memcpy(compiler->bytecode + jumpToEnd, &tmpVal, sizeof(tmpVal)); //2 bytes
}
}
break;
case TOY_AST_NODE_WHILE: {
//for breaks and continues
Toy_LiteralArray breakAddresses;
Toy_LiteralArray continueAddresses;
Toy_initLiteralArray(&breakAddresses);
Toy_initLiteralArray(&continueAddresses);
//cache the jump point
unsigned short jumpToStart = compiler->count;
//process the condition
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->pathWhile.condition, &breakAddresses, &continueAddresses, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//if false, jump to end
compiler->bytecode[compiler->count++] = TOY_OP_IF_FALSE_JUMP; //1 byte
unsigned short jumpToEnd = compiler->count;
compiler->count += sizeof(unsigned short); //2 bytes
//write the body
override = Toy_writeCompilerWithJumps(compiler, node->pathWhile.thenPath, &breakAddresses, &continueAddresses, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//jump to condition
compiler->bytecode[compiler->count++] = TOY_OP_JUMP; //1 byte
unsigned short tmpVal = jumpToStart + jumpOffsets;
memcpy(compiler->bytecode + compiler->count, &tmpVal, sizeof(tmpVal));
compiler->count += sizeof(unsigned short); //2 bytes
//jump from condition
tmpVal = compiler->count + jumpOffsets;
memcpy(compiler->bytecode + jumpToEnd, &tmpVal, sizeof(tmpVal));
//set the breaks and continues
for (int i = 0; i < breakAddresses.count; i++) {
int point = TOY_AS_INTEGER(breakAddresses.literals[i]);
tmpVal = compiler->count + jumpOffsets;
memcpy(compiler->bytecode + point, &tmpVal, sizeof(tmpVal));
}
for (int i = 0; i < continueAddresses.count; i++) {
int point = TOY_AS_INTEGER(continueAddresses.literals[i]);
tmpVal = jumpToStart + jumpOffsets;
memcpy(compiler->bytecode + point, &tmpVal, sizeof(tmpVal));
}
//clear the stack after use
compiler->bytecode[compiler->count++] = TOY_OP_POP_STACK; //1 byte
//cleanup
Toy_freeLiteralArray(&breakAddresses);
Toy_freeLiteralArray(&continueAddresses);
}
break;
case TOY_AST_NODE_FOR: {
//for breaks and continues
Toy_LiteralArray breakAddresses;
Toy_LiteralArray continueAddresses;
Toy_initLiteralArray(&breakAddresses);
Toy_initLiteralArray(&continueAddresses);
compiler->bytecode[compiler->count++] = TOY_OP_SCOPE_BEGIN; //1 byte
//initial setup
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->pathFor.preClause, &breakAddresses, &continueAddresses, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//conditional
unsigned short jumpToStart = compiler->count;
override = Toy_writeCompilerWithJumps(compiler, node->pathFor.condition, &breakAddresses, &continueAddresses, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
//if false jump to end
compiler->bytecode[compiler->count++] = TOY_OP_IF_FALSE_JUMP; //1 byte
unsigned short jumpToEnd = compiler->count;
compiler->count += sizeof(unsigned short); //2 bytes
//write the body
compiler->bytecode[compiler->count++] = TOY_OP_SCOPE_BEGIN; //1 byte
override = Toy_writeCompilerWithJumps(compiler, node->pathFor.thenPath, &breakAddresses, &continueAddresses, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
compiler->bytecode[compiler->count++] = TOY_OP_SCOPE_END; //1 byte
//for-breaks actually jump to the bottom
int jumpToIncrement = compiler->count;
//evaluate third clause, restart
override = Toy_writeCompilerWithJumps(compiler, node->pathFor.postClause, &breakAddresses, &continueAddresses, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
compiler->bytecode[compiler->count++] = TOY_OP_JUMP; //1 byte
unsigned short tmpVal = jumpToStart + jumpOffsets;
memcpy(compiler->bytecode + compiler->count, &tmpVal, sizeof(tmpVal));
compiler->count += sizeof(unsigned short); //2 bytes
tmpVal = compiler->count + jumpOffsets;
memcpy(compiler->bytecode + jumpToEnd, &tmpVal, sizeof(tmpVal));
compiler->bytecode[compiler->count++] = TOY_OP_SCOPE_END; //1 byte
//set the breaks and continues
for (int i = 0; i < breakAddresses.count; i++) {
int point = TOY_AS_INTEGER(breakAddresses.literals[i]);
tmpVal = compiler->count + jumpOffsets;
memcpy(compiler->bytecode + point, &tmpVal, sizeof(tmpVal));
}
for (int i = 0; i < continueAddresses.count; i++) {
int point = TOY_AS_INTEGER(continueAddresses.literals[i]);
tmpVal = jumpToIncrement + jumpOffsets;
memcpy(compiler->bytecode + point, &tmpVal, sizeof(tmpVal));
}
//clear the stack after use
compiler->bytecode[compiler->count++] = TOY_OP_POP_STACK; //1 byte
//cleanup
Toy_freeLiteralArray(&breakAddresses);
Toy_freeLiteralArray(&continueAddresses);
}
break;
case TOY_AST_NODE_BREAK: {
if (!breakAddressesPtr) {
fprintf(stderr, TOY_CC_ERROR "[internal] Can't place a break statement here\n" TOY_CC_RESET);
break;
}
//insert into bytecode
compiler->bytecode[compiler->count++] = TOY_OP_JUMP; //1 byte
//push to the breakAddresses array
Toy_Literal literal = TOY_TO_INTEGER_LITERAL(compiler->count);
Toy_pushLiteralArray((Toy_LiteralArray*)breakAddressesPtr, literal);
Toy_freeLiteral(literal);
compiler->count += sizeof(unsigned short); //2 bytes
}
break;
case TOY_AST_NODE_CONTINUE: {
if (!continueAddressesPtr) {
fprintf(stderr, TOY_CC_ERROR "[internal] Can't place a continue statement here\n" TOY_CC_RESET);
break;
}
//insert into bytecode
compiler->bytecode[compiler->count++] = TOY_OP_JUMP; //1 byte
//push to the continueAddresses array
Toy_Literal literal = TOY_TO_INTEGER_LITERAL(compiler->count);
Toy_pushLiteralArray((Toy_LiteralArray*)continueAddressesPtr, literal);
Toy_freeLiteral(literal);
compiler->count += sizeof(unsigned short); //2 bytes
}
break;
case TOY_AST_NODE_FN_RETURN: {
//read each returned literal onto the stack, and return the number of values to return
for (int i = 0; i < node->returns.returns->fnCollection.count; i++) {
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, &node->returns.returns->fnCollection.nodes[i], breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
}
//push the return, with the number of literals
compiler->bytecode[compiler->count++] = TOY_OP_FN_RETURN; //1 byte
memcpy(compiler->bytecode + compiler->count, &node->returns.returns->fnCollection.count, sizeof(unsigned short));
compiler->count += sizeof(unsigned short);
}
break;
case TOY_AST_NODE_PREFIX_INCREMENT: {
//push the literal to the stack (twice: add + assign)
writeLiteralToCompiler(compiler, node->prefixIncrement.identifier);
writeLiteralToCompiler(compiler, node->prefixIncrement.identifier);
//push the increment / decrement
Toy_Literal increment = TOY_TO_INTEGER_LITERAL(1);
writeLiteralToCompiler(compiler, increment);
//push the add opcode
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_ADDITION; //1 byte
//push the assign
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_VAR_ASSIGN; //1 byte
//leave the result on the stack
writeLiteralToCompiler(compiler, node->prefixIncrement.identifier);
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_LITERAL_RAW; //1 byte
}
break;
case TOY_AST_NODE_PREFIX_DECREMENT: {
//push the literal to the stack (twice: add + assign)
writeLiteralToCompiler(compiler, node->prefixDecrement.identifier);
writeLiteralToCompiler(compiler, node->prefixDecrement.identifier);
//push the increment / decrement
Toy_Literal increment = TOY_TO_INTEGER_LITERAL(1);
writeLiteralToCompiler(compiler, increment);
//push the subtract opcode
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_SUBTRACTION; //1 byte
//push the assign
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_VAR_ASSIGN; //1 byte
//leave the result on the stack
writeLiteralToCompiler(compiler, node->prefixDecrement.identifier);
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_LITERAL_RAW; //1 byte
}
break;
case TOY_AST_NODE_POSTFIX_INCREMENT: {
//push the identifier's VALUE to the stack
writeLiteralToCompiler(compiler, node->postfixIncrement.identifier);
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_LITERAL_RAW; //1 byte
//push the identifier (twice: add + assign)
writeLiteralToCompiler(compiler, node->postfixIncrement.identifier);
writeLiteralToCompiler(compiler, node->postfixIncrement.identifier);
//push the increment / decrement
Toy_Literal increment = TOY_TO_INTEGER_LITERAL(1);
writeLiteralToCompiler(compiler, increment);
//push the add opcode
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_ADDITION; //1 byte
//push the assign
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_VAR_ASSIGN; //1 byte
}
break;
case TOY_AST_NODE_POSTFIX_DECREMENT: {
//push the identifier's VALUE to the stack
writeLiteralToCompiler(compiler, node->postfixDecrement.identifier);
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_LITERAL_RAW; //1 byte
//push the identifier (twice: add + assign)
writeLiteralToCompiler(compiler, node->postfixDecrement.identifier);
writeLiteralToCompiler(compiler, node->postfixDecrement.identifier);
//push the increment / decrement
Toy_Literal increment = TOY_TO_INTEGER_LITERAL(1);
writeLiteralToCompiler(compiler, increment);
//push the subtract opcode
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_SUBTRACTION; //1 byte
//push the assign
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_VAR_ASSIGN; //1 byte
}
break;
case TOY_AST_NODE_IMPORT: {
//push the identifier, and the alias
writeLiteralToCompiler(compiler, node->import.identifier);
writeLiteralToCompiler(compiler, node->import.alias);
//push the import opcode
compiler->bytecode[compiler->count++] = (unsigned char)TOY_OP_IMPORT; //1 byte
}
break;
case TOY_AST_NODE_INDEX: {
//pass to the child nodes, then embed the opcode
//first
if (!node->index.first) {
writeLiteralToCompiler(compiler, TOY_TO_NULL_LITERAL);
}
else {
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->index.first, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
}
//second
if (!node->index.second) {
writeLiteralToCompiler(compiler, TOY_TO_NULL_LITERAL);
}
else {
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->index.second, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
}
//third
if (!node->index.third) {
writeLiteralToCompiler(compiler, TOY_TO_NULL_LITERAL);
}
else {
Toy_Opcode override = Toy_writeCompilerWithJumps(compiler, node->index.third, breakAddressesPtr, continueAddressesPtr, jumpOffsets, rootNode);
if (override != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)override; //1 byte
}
}
// compiler->bytecode[compiler->count++] = (unsigned char)OP_INDEX; //1 byte
return TOY_OP_INDEX_ASSIGN; //override binary's instruction IF it is assign
}
break;
case TOY_AST_NODE_PASS: {
return TOY_OP_PASS;
}
break;
}
return TOY_OP_EOF;
}
void Toy_writeCompiler(Toy_Compiler* compiler, Toy_ASTNode* node) {
Toy_Opcode op = Toy_writeCompilerWithJumps(compiler, node, NULL, NULL, 0, node); //pass in "node" as the root node
if (op != TOY_OP_EOF) {//compensate for indexing & dot notation being screwy
compiler->bytecode[compiler->count++] = (unsigned char)op; //1 byte
}
//TODO: could free up AST Nodes
}
void Toy_freeCompiler(Toy_Compiler* compiler) {
Toy_freeLiteralArray(&compiler->literalCache);
TOY_FREE_ARRAY(unsigned char, compiler->bytecode, compiler->capacity);
compiler->bytecode = NULL;
compiler->capacity = 0;
compiler->count = 0;
compiler->panic = false;
}
static void emitByte(unsigned char** collationPtr, int* capacityPtr, int* countPtr, unsigned char byte) {
//grow the array
if (*countPtr + 1 > *capacityPtr) {
int oldCapacity = *capacityPtr;
*capacityPtr = TOY_GROW_CAPACITY(*capacityPtr);
*collationPtr = TOY_GROW_ARRAY(unsigned char, *collationPtr, oldCapacity, *capacityPtr);
}
//append to the collation
(*collationPtr)[(*countPtr)++] = byte;
}
static void Toy_emitShort(unsigned char** collationPtr, int* capacityPtr, int* countPtr, unsigned short bytes) {
char* ptr = (char*)&bytes;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
ptr++;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
}
static void emitInt(unsigned char** collationPtr, int* capacityPtr, int* countPtr, int bytes) {
char* ptr = (char*)&bytes;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
ptr++;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
ptr++;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
ptr++;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
}
static void emitFloat(unsigned char** collationPtr, int* capacityPtr, int* countPtr, float bytes) {
char* ptr = (char*)&bytes;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
ptr++;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
ptr++;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
ptr++;
emitByte(collationPtr, capacityPtr, countPtr, *ptr);
}
//return the result
static unsigned char* collateCompilerHeaderOpt(Toy_Compiler* compiler, size_t* size, bool embedHeader) {
if (compiler->panic) {
fprintf(stderr, TOY_CC_ERROR "[internal] Can't collate a panicked compiler\n" TOY_CC_RESET);
return NULL;
}
int capacity = TOY_GROW_CAPACITY(0);
int count = 0;
unsigned char* collation = TOY_ALLOCATE(unsigned char, capacity);
//for the function-section at the end of the main-collation
int fnIndex = 0; //counts up for each fn
int fnCapacity = TOY_GROW_CAPACITY(0);
int fnCount = 0;
unsigned char* fnCollation = TOY_ALLOCATE(unsigned char, fnCapacity);
if (embedHeader) {
//embed the header with version information
emitByte(&collation, &capacity, &count, TOY_VERSION_MAJOR);
emitByte(&collation, &capacity, &count, TOY_VERSION_MINOR);
emitByte(&collation, &capacity, &count, TOY_VERSION_PATCH);
//embed the build info
if ((int)strlen(TOY_VERSION_BUILD) + count + 1 > capacity) {
int oldCapacity = capacity;
capacity = strlen(TOY_VERSION_BUILD) + count + 1; //full header size
collation = TOY_GROW_ARRAY(unsigned char, collation, oldCapacity, capacity);
}
memcpy(&collation[count], TOY_VERSION_BUILD, strlen(TOY_VERSION_BUILD));
count += strlen(TOY_VERSION_BUILD);
collation[count++] = '\0'; //terminate the build string
emitByte(&collation, &capacity, &count, TOY_OP_SECTION_END); //terminate header
}
//embed the data section (first short is the number of literals)
Toy_emitShort(&collation, &capacity, &count, compiler->literalCache.count);
//emit each literal by type
for (int i = 0; i < compiler->literalCache.count; i++) {
//literal Opcode
// emitShort(&collation, &capacity, &count, OP_LITERAL); //This isn't needed
//literal type, followed by literal value
switch(compiler->literalCache.literals[i].type) {
case TOY_LITERAL_NULL:
emitByte(&collation, &capacity, &count, TOY_LITERAL_NULL);
//null has no following value
break;
case TOY_LITERAL_BOOLEAN:
emitByte(&collation, &capacity, &count, TOY_LITERAL_BOOLEAN);
emitByte(&collation, &capacity, &count, TOY_AS_BOOLEAN(compiler->literalCache.literals[i]));
break;
case TOY_LITERAL_INTEGER:
emitByte(&collation, &capacity, &count, TOY_LITERAL_INTEGER);
emitInt(&collation, &capacity, &count, TOY_AS_INTEGER(compiler->literalCache.literals[i]));
break;
case TOY_LITERAL_FLOAT:
emitByte(&collation, &capacity, &count, TOY_LITERAL_FLOAT);
emitFloat(&collation, &capacity, &count, TOY_AS_FLOAT(compiler->literalCache.literals[i]));
break;
case TOY_LITERAL_STRING: {
emitByte(&collation, &capacity, &count, TOY_LITERAL_STRING);
Toy_Literal str = compiler->literalCache.literals[i];
for (int c = 0; c < (int)Toy_lengthRefString(TOY_AS_STRING(str)); c++) {
emitByte(&collation, &capacity, &count, Toy_toCString(TOY_AS_STRING(str))[c]);
}
emitByte(&collation, &capacity, &count, '\0'); //terminate the string
}
break;
case TOY_LITERAL_ARRAY: {
emitByte(&collation, &capacity, &count, TOY_LITERAL_ARRAY);
Toy_LiteralArray* ptr = TOY_AS_ARRAY(compiler->literalCache.literals[i]);
//length of the array, as a short
Toy_emitShort(&collation, &capacity, &count, ptr->count);
//each element of the array
for (int i = 0; i < ptr->count; i++) {
Toy_emitShort(&collation, &capacity, &count, (unsigned short)TOY_AS_INTEGER(ptr->literals[i])); //shorts representing the indexes of the values
}
}
break;
case TOY_LITERAL_ARRAY_INTERMEDIATE: {
emitByte(&collation, &capacity, &count, TOY_LITERAL_ARRAY_INTERMEDIATE);
Toy_LiteralArray* ptr = TOY_AS_ARRAY(compiler->literalCache.literals[i]);
//length of the array, as a short
Toy_emitShort(&collation, &capacity, &count, ptr->count);
//each element of the array
for (int i = 0; i < ptr->count; i++) {
Toy_emitShort(&collation, &capacity, &count, (unsigned short)TOY_AS_INTEGER(ptr->literals[i])); //shorts representing the indexes of the values
}
}
break;
case TOY_LITERAL_DICTIONARY: {
emitByte(&collation, &capacity, &count, TOY_LITERAL_DICTIONARY);
Toy_LiteralArray* ptr = TOY_AS_ARRAY(compiler->literalCache.literals[i]); //used an array for storage above
//length of the array, as a short
Toy_emitShort(&collation, &capacity, &count, ptr->count); //count is the array size, NOT the dictionary size
//each element of the array
for (int i = 0; i < ptr->count; i++) {
Toy_emitShort(&collation, &capacity, &count, (unsigned short)TOY_AS_INTEGER(ptr->literals[i])); //shorts representing the indexes of the values
}
}
break;
case TOY_LITERAL_DICTIONARY_INTERMEDIATE: {
emitByte(&collation, &capacity, &count, TOY_LITERAL_DICTIONARY_INTERMEDIATE);
Toy_LiteralArray* ptr = TOY_AS_ARRAY(compiler->literalCache.literals[i]); //used an array for storage above
//length of the array, as a short
Toy_emitShort(&collation, &capacity, &count, ptr->count); //count is the array size, NOT the dictionary size
//each element of the array
for (int i = 0; i < ptr->count; i++) {
Toy_emitShort(&collation, &capacity, &count, (unsigned short)TOY_AS_INTEGER(ptr->literals[i])); //shorts representing the indexes of the values
}
}
break;
case TOY_LITERAL_FUNCTION_INTERMEDIATE: {
//extract the compiler
Toy_Literal fn = compiler->literalCache.literals[i];
void* fnCompiler = TOY_AS_FUNCTION(fn).inner.bytecode; //store the compiler here for now
//collate the function into bytecode (without header)
size_t size = 0;
unsigned char* bytes = collateCompilerHeaderOpt((Toy_Compiler*)fnCompiler, &size, false);
//emit how long this section is, +1 for ending mark
Toy_emitShort(&fnCollation, &fnCapacity, &fnCount, (unsigned short)size + 1);
//write the fn to the fn collation
for (size_t i = 0; i < size; i++) {
emitByte(&fnCollation, &fnCapacity, &fnCount, bytes[i]);
}
emitByte(&fnCollation, &fnCapacity, &fnCount, TOY_OP_FN_END); //for marking the correct end-point of the function
//embed the reference to the function implementation into the current collation (to be extracted later)
emitByte(&collation, &capacity, &count, TOY_LITERAL_FUNCTION);
Toy_emitShort(&collation, &capacity, &count, (unsigned short)(fnIndex++));
Toy_freeCompiler((Toy_Compiler*)fnCompiler);
TOY_FREE(compiler, fnCompiler);
TOY_FREE_ARRAY(unsigned char, bytes, size);
}
break;
case TOY_LITERAL_IDENTIFIER: {
emitByte(&collation, &capacity, &count, TOY_LITERAL_IDENTIFIER);
Toy_Literal identifier = compiler->literalCache.literals[i];
for (int c = 0; c < (int)Toy_lengthRefString(TOY_AS_IDENTIFIER(identifier)); c++) {
emitByte(&collation, &capacity, &count, Toy_toCString(TOY_AS_IDENTIFIER(identifier))[c]);
}
emitByte(&collation, &capacity, &count, '\0'); //terminate the string
}
break;
case TOY_LITERAL_TYPE: {
//push a raw type
emitByte(&collation, &capacity, &count, TOY_LITERAL_TYPE);
Toy_Literal typeLiteral = compiler->literalCache.literals[i];
//what type this literal represents
emitByte(&collation, &capacity, &count, TOY_AS_TYPE(typeLiteral).typeOf);
emitByte(&collation, &capacity, &count, TOY_AS_TYPE(typeLiteral).constant); //if it's constant
}
break;
case TOY_LITERAL_TYPE_INTERMEDIATE: {
emitByte(&collation, &capacity, &count, TOY_LITERAL_TYPE_INTERMEDIATE);
Toy_LiteralArray* ptr = TOY_AS_ARRAY(compiler->literalCache.literals[i]); //used an array for storage above
//the base literal
Toy_Literal typeLiteral = Toy_copyLiteral(ptr->literals[0]);
//what type this literal represents
emitByte(&collation, &capacity, &count, TOY_AS_TYPE(typeLiteral).typeOf);
emitByte(&collation, &capacity, &count, TOY_AS_TYPE(typeLiteral).constant); //if it's constant
//each element of the array, If they exist, representing sub-types already in the cache
if (TOY_AS_TYPE(typeLiteral).typeOf == TOY_LITERAL_ARRAY || TOY_AS_TYPE(typeLiteral).typeOf == TOY_LITERAL_DICTIONARY) {
//the type will represent how many to expect in the array
for (int i = 1; i < ptr->count; i++) {
Toy_emitShort(&collation, &capacity, &count, (unsigned short)TOY_AS_INTEGER(ptr->literals[i])); //shorts representing the indexes of the types
}
}
Toy_freeLiteral(typeLiteral);
}
break;
case TOY_LITERAL_INDEX_BLANK:
emitByte(&collation, &capacity, &count, TOY_LITERAL_INDEX_BLANK);
//blank has no following value
break;
default:
fprintf(stderr, TOY_CC_ERROR "[internal] Unknown literal type encountered within literal cache: %d\n" TOY_CC_RESET, compiler->literalCache.literals[i].type);
return NULL;
}
}
emitByte(&collation, &capacity, &count, TOY_OP_SECTION_END); //terminate data
//embed the function section (beginning with function count, size)
Toy_emitShort(&collation, &capacity, &count, fnIndex);
Toy_emitShort(&collation, &capacity, &count, fnCount);
for (int i = 0; i < fnCount; i++) {
emitByte(&collation, &capacity, &count, fnCollation[i]);
}
emitByte(&collation, &capacity, &count, TOY_OP_SECTION_END); //terminate function section
TOY_FREE_ARRAY(unsigned char, fnCollation, fnCapacity); //clear the function stuff
//code section
for (int i = 0; i < compiler->count; i++) {
emitByte(&collation, &capacity, &count, compiler->bytecode[i]);
}
emitByte(&collation, &capacity, &count, TOY_OP_SECTION_END); //terminate code
emitByte(&collation, &capacity, &count, TOY_OP_EOF); //terminate bytecode
//finalize
collation = TOY_SHRINK_ARRAY(unsigned char, collation, capacity, count);
*size = count;
return collation;
}
//the whole point of the compiler is to alter bytecode, so leave it as non-const
unsigned char* Toy_collateCompiler(Toy_Compiler* compiler, size_t* size) {
return collateCompilerHeaderOpt(compiler, size, true);
}