/*
 * *****************************************************************************
 *
 * SPDX-License-Identifier: BSD-2-Clause
 *
 * Copyright (c) 2018-2021 Gavin D. Howard and contributors.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 * * Redistributions of source code must retain the above copyright notice, this
 *   list of conditions and the following disclaimer.
 *
 * * Redistributions in binary form must reproduce the above copyright notice,
 *   this list of conditions and the following disclaimer in the documentation
 *   and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 *
 * *****************************************************************************
 *
 * The parser for dc.
 *
 */

#if DC_ENABLED

#include <assert.h>
#include <stdlib.h>
#include <string.h>
#include <setjmp.h>

#include <dc.h>
#include <program.h>
#include <vm.h>

/**
 * Parses a register. The lexer should have already lexed the true name of the
 * register, per extended registers and such.
 * @param p    The parser.
 * @param var  True if the parser is for a variable, false otherwise.
 */
static void dc_parse_register(BcParse *p, bool var) {

	bc_lex_next(&p->l);
	if (p->l.t != BC_LEX_NAME) bc_parse_err(p, BC_ERR_PARSE_TOKEN);

	bc_parse_pushName(p, p->l.str.v, var);
}

/**
 * Parses a dc string.
 * @param p  The parser.
 */
static inline void dc_parse_string(BcParse *p) {
	bc_parse_addString(p);
	bc_lex_next(&p->l);
}

/**
 * Parses a token that requires a memory operation, like load or store.
 * @param p      The parser.
 * @param inst   The instruction to push for the memory operation.
 * @param name   Whether the load or store is to a variable or array, and not to
 *               a global.
 * @param store  True if the operation is a store, false otherwise.
 */
static void dc_parse_mem(BcParse *p, uchar inst, bool name, bool store) {

	// Push the instruction.
	bc_parse_push(p, inst);

	// Parse the register if necessary.
	if (name) dc_parse_register(p, inst != BC_INST_ARRAY_ELEM);

	// Stores use the bc assign infrastructure, but they need to do a swap
	// first.
	if (store) {
		bc_parse_push(p, BC_INST_SWAP);
		bc_parse_push(p, BC_INST_ASSIGN_NO_VAL);
	}

	bc_lex_next(&p->l);
}

/**
 * Parses a conditional execution instruction.
 * @param p     The parser.
 * @param inst  The instruction for the condition.
 */
static void dc_parse_cond(BcParse *p, uchar inst) {

	// Push the instruction for the condition and the conditional execution.
	bc_parse_push(p, inst);
	bc_parse_push(p, BC_INST_EXEC_COND);

	// Parse the register.
	dc_parse_register(p, true);

	bc_lex_next(&p->l);

	// If the next token is an else, parse the else.
	if (p->l.t == BC_LEX_KW_ELSE) {
		dc_parse_register(p, true);
		bc_lex_next(&p->l);
	}
	// Otherwise, push a marker for no else.
	else bc_parse_pushIndex(p, SIZE_MAX);
}

/**
 * Parses a token for dc.
 * @param p      The parser.
 * @param t      The token to parse.
 * @param flags  The flags that say what is allowed or not.
 */
static void dc_parse_token(BcParse *p, BcLexType t, uint8_t flags) {

	uchar inst;
	bool assign, get_token = false;

	switch (t) {

		case BC_LEX_OP_REL_EQ:
		case BC_LEX_OP_REL_LE:
		case BC_LEX_OP_REL_GE:
		case BC_LEX_OP_REL_NE:
		case BC_LEX_OP_REL_LT:
		case BC_LEX_OP_REL_GT:
		{
			inst = (uchar) (t - BC_LEX_OP_REL_EQ + BC_INST_REL_EQ);
			dc_parse_cond(p, inst);
			break;
		}

		case BC_LEX_SCOLON:
		case BC_LEX_COLON:
		{
			dc_parse_mem(p, BC_INST_ARRAY_ELEM, true, t == BC_LEX_COLON);
			break;
		}

		case BC_LEX_STR:
		{
			dc_parse_string(p);
			break;
		}

		case BC_LEX_NEG:
		{
			// This tells us whether or not the neg is for a command or at the
			// beginning of a number. If it's a command, push it. Otherwise,
			// fallthrough and parse the number.
			if (dc_lex_negCommand(&p->l)) {
				bc_parse_push(p, BC_INST_NEG);
				get_token = true;
				break;
			}

			bc_lex_next(&p->l);
		}
		// Fallthrough.
		BC_FALLTHROUGH

		case BC_LEX_NUMBER:
		{
			bc_parse_number(p);

			// Push the negative instruction if we fell through from above.
			if (t == BC_LEX_NEG) bc_parse_push(p, BC_INST_NEG);
			get_token = true;

			break;
		}

		case BC_LEX_KW_READ:
		{
			// Make sure the read is not recursive.
			if (BC_ERR(flags & BC_PARSE_NOREAD))
				bc_parse_err(p, BC_ERR_EXEC_REC_READ);
			else bc_parse_push(p, BC_INST_READ);

			get_token = true;

			break;
		}

		case BC_LEX_OP_ASSIGN:
		case BC_LEX_STORE_PUSH:
		{
			assign = t == BC_LEX_OP_ASSIGN;
			inst = assign ? BC_INST_VAR : BC_INST_PUSH_TO_VAR;
			dc_parse_mem(p, inst, true, assign);
			break;
		}

		case BC_LEX_LOAD:
		case BC_LEX_LOAD_POP:
		{
			inst = t == BC_LEX_LOAD_POP ? BC_INST_PUSH_VAR : BC_INST_LOAD;
			dc_parse_mem(p, inst, true, false);
			break;
		}

		case BC_LEX_REG_STACK_LEVEL:
		{
			dc_parse_mem(p, BC_INST_REG_STACK_LEN, true, false);
			break;
		}

		case BC_LEX_STORE_IBASE:
		case BC_LEX_STORE_OBASE:
		case BC_LEX_STORE_SCALE:
#if BC_ENABLE_EXTRA_MATH
		case BC_LEX_STORE_SEED:
#endif // BC_ENABLE_EXTRA_MATH
		{
			inst = (uchar) (t - BC_LEX_STORE_IBASE + BC_INST_IBASE);
			dc_parse_mem(p, inst, false, true);
			break;
		}

		case BC_LEX_ARRAY_LENGTH:
		{
			// Need to push the array first, based on how length is implemented.
			bc_parse_push(p, BC_INST_ARRAY);
			dc_parse_register(p, false);

			bc_parse_push(p, BC_INST_LENGTH);

			get_token = true;

			break;
		}

		default:
		{
			// All other tokens should be taken care of by the caller, or they
			// actually *are* invalid.
			bc_parse_err(p, BC_ERR_PARSE_TOKEN);
		}
	}

	if (get_token) bc_lex_next(&p->l);
}

void dc_parse_expr(BcParse *p, uint8_t flags) {

	BcInst inst;
	BcLexType t;
	bool need_expr, have_expr = false;

	need_expr = ((flags & BC_PARSE_NOREAD) != 0);

	// dc can just keep parsing forever basically, unlike bc, which has to have
	// a whole bunch of complicated nonsense because its language was horribly
	// designed.

	// While we don't have EOF...
	while ((t = p->l.t) != BC_LEX_EOF) {

		// Eat newline.
		if (t == BC_LEX_NLINE) {
			bc_lex_next(&p->l);
			continue;
		}

		// Get the instruction that corresponds to the token.
		inst = dc_parse_insts[t];

		// If the instruction is invalid, that means we have to do some harder
		// parsing. So if not invalid, just push the instruction; otherwise,
		// parse the token.
		if (inst != BC_INST_INVALID) {
			bc_parse_push(p, inst);
			bc_lex_next(&p->l);
		}
		else dc_parse_token(p, t, flags);

		have_expr = true;
	}

	// If we don't have an expression and need one, barf. Otherwise, just push a
	// BC_INST_POP_EXEC if we have EOF and BC_PARSE_NOCALL, which dc uses to
	// indicate that it is executing a string.
	if (BC_ERR(need_expr && !have_expr)) bc_err(BC_ERR_EXEC_READ_EXPR);
	else if (p->l.t == BC_LEX_EOF && (flags & BC_PARSE_NOCALL))
		bc_parse_push(p, BC_INST_POP_EXEC);
}

void dc_parse_parse(BcParse *p) {

	assert(p != NULL);

	BC_SETJMP(exit);

	// If we have EOF, someone called this function one too many times.
	// Otherwise, parse.
	if (BC_ERR(p->l.t == BC_LEX_EOF)) bc_parse_err(p, BC_ERR_PARSE_EOF);
	else dc_parse_expr(p, 0);

exit:

	BC_SIG_MAYLOCK;

	// Need to reset if there was an error.
	if (BC_SIG_EXC) bc_parse_reset(p);

	BC_LONGJMP_CONT;
}
#endif // DC_ENABLED