mermaid-go/pkg/parser/bpmn.go

// Package parser provides BPMN parsing
package parser

import (
	"fmt"
	"strings"

	"mermaid-go/pkg/ast"
	"mermaid-go/pkg/lexer"
)

// BPMNParser implements BPMN parsing
type BPMNParser struct {
	tokens  []lexer.Token
	current int
	diagram *ast.BPMNDiagram
}

// NewBPMNParser creates a new BPMN parser
func NewBPMNParser() *BPMNParser {
	return &BPMNParser{
		diagram: ast.NewBPMNDiagram(),
	}
}

// Parse parses BPMN syntax
func (p *BPMNParser) Parse(input string) (*ast.BPMNDiagram, error) {
	// Tokenize
	l := lexer.NewLexer(input)
	tokens, err := l.Tokenize()
	if err != nil {
		return nil, fmt.Errorf("lexical analysis failed: %w", err)
	}

	// Filter tokens
	p.tokens = lexer.FilterTokens(tokens)
	p.current = 0
	p.diagram = ast.NewBPMNDiagram()

	// Parse document
	err = p.parseDocument()
	if err != nil {
		return nil, fmt.Errorf("syntax analysis failed: %w", err)
	}

	return p.diagram, nil
}

// parseDocument parses the BPMN document
func (p *BPMNParser) parseDocument() error {
	// Expect bpmn
	if !p.check(lexer.TokenID) || p.peek().Value != "bpmn" {
		return p.error("expected 'bpmn'")
	}
	p.advance()

	// Parse statements
	for !p.isAtEnd() {
		if err := p.parseStatement(); err != nil {
			return err
		}
	}

	return nil
}

// parseStatement parses individual BPMN statements
func (p *BPMNParser) parseStatement() error {
	if p.isAtEnd() {
		return nil
	}

	switch {
	case p.check(lexer.TokenNewline):
		p.advance() // Skip newlines
		return nil
	case p.checkKeyword("title"):
		return p.parseTitle()
	case p.checkKeyword("pool"):
		return p.parsePool()
	case p.checkKeyword("lane"):
		return p.parseLane()
	case p.check(lexer.TokenID):
		// Element definition or flow
		return p.parseElementOrFlow()
	default:
		token := p.peek()
		return p.error(fmt.Sprintf("unexpected token: %s", token.Value))
	}
}

// parseTitle parses title statements
func (p *BPMNParser) parseTitle() error {
	p.advance() // consume 'title'

	var titleParts []string
	for !p.check(lexer.TokenNewline) && !p.isAtEnd() {
		titleParts = append(titleParts, p.advance().Value)
	}

	if len(titleParts) > 0 {
		title := strings.TrimSpace(strings.Join(titleParts, " "))
		p.diagram.Title = &title
	}

	return nil
}

// parsePool parses pool statements
func (p *BPMNParser) parsePool() error {
	p.advance() // consume 'pool'

	if !p.check(lexer.TokenID) {
		return p.error("expected pool ID")
	}

	poolID := p.advance().Value

	pool := &ast.BPMNPool{
		ID:    poolID,
		Name:  poolID,
		Lanes: make([]string, 0),
	}

	// Parse pool properties
	for !p.check(lexer.TokenNewline) && !p.isAtEnd() {
		if p.check(lexer.TokenOpenBracket) {
			p.advance() // consume '['

			// Parse pool name
			var nameParts []string
			for !p.check(lexer.TokenCloseBracket) && !p.isAtEnd() {
				nameParts = append(nameParts, p.advance().Value)
			}

			if len(nameParts) > 0 {
				pool.Name = strings.TrimSpace(strings.Join(nameParts, " "))
			}

			if p.check(lexer.TokenCloseBracket) {
				p.advance() // consume ']'
			}
		} else {
			p.advance() // consume unknown token
		}
	}

	p.diagram.Pools = append(p.diagram.Pools, pool)
	return nil
}

// parseLane parses lane statements
func (p *BPMNParser) parseLane() error {
	p.advance() // consume 'lane'

	if !p.check(lexer.TokenID) {
		return p.error("expected lane ID")
	}

	laneID := p.advance().Value

	lane := &ast.BPMNLane{
		ID:       laneID,
		Name:     laneID,
		Elements: make([]string, 0),
	}

	// Parse lane properties
	for !p.check(lexer.TokenNewline) && !p.isAtEnd() {
		if p.check(lexer.TokenOpenBracket) {
			p.advance() // consume '['

			// Parse lane name
			var nameParts []string
			for !p.check(lexer.TokenCloseBracket) && !p.isAtEnd() {
				nameParts = append(nameParts, p.advance().Value)
			}

			if len(nameParts) > 0 {
				lane.Name = strings.TrimSpace(strings.Join(nameParts, " "))
			}

			if p.check(lexer.TokenCloseBracket) {
				p.advance() // consume ']'
			}
		} else if p.checkKeyword("in") {
			p.advance() // consume 'in'
			if p.check(lexer.TokenID) {
				lane.Pool = p.advance().Value
			}
		} else {
			p.advance() // consume unknown token
		}
	}

	p.diagram.Lanes = append(p.diagram.Lanes, lane)
	return nil
}

// parseElementOrFlow parses element definition or flow
func (p *BPMNParser) parseElementOrFlow() error {
	elementID := p.advance().Value

	// Check if this is a flow (has arrow indicators)
	if p.checkFlow() {
		return p.parseFlow(elementID)
	}

	// Otherwise, it's an element definition
	element := &ast.BPMNElement{
		ID:         elementID,
		Name:       elementID,
		Type:       p.inferElementType(elementID),
		Properties: make(map[string]any),
		CssClasses: make([]string, 0),
	}

	// Parse element properties
	for !p.check(lexer.TokenNewline) && !p.isAtEnd() {
		if p.check(lexer.TokenOpenBracket) {
			p.advance() // consume '['

			// Parse element name
			var nameParts []string
			for !p.check(lexer.TokenCloseBracket) && !p.isAtEnd() {
				nameParts = append(nameParts, p.advance().Value)
			}

			if len(nameParts) > 0 {
				element.Name = strings.TrimSpace(strings.Join(nameParts, " "))
			}

			if p.check(lexer.TokenCloseBracket) {
				p.advance() // consume ']'
			}
		} else if p.check(lexer.TokenOpenParen) {
			// Parse element type
			p.advance() // consume '('

			if p.check(lexer.TokenID) {
				typeStr := p.advance().Value
				element.Type = ast.BPMNElementType(typeStr)
			}

			if p.check(lexer.TokenCloseParen) {
				p.advance() // consume ')'
			}
		} else {
			p.advance() // consume unknown token
		}
	}

	p.diagram.AddElement(element)
	return nil
}

// parseFlow parses flow connections
func (p *BPMNParser) parseFlow(fromID string) error {
	// Parse flow type and direction
	flowType := ast.BPMNFlowSequence // default

	// Skip flow indicators
	for p.checkFlow() {
		token := p.advance()
		if token.Value == "-->" {
			flowType = ast.BPMNFlowSequence
		} else if token.Value == "-.>" {
			flowType = ast.BPMNFlowMessage
		}
	}

	// Parse target element
	if !p.check(lexer.TokenID) {
		return p.error("expected target element ID")
	}

	toID := p.advance().Value

	flow := &ast.BPMNFlow{
		ID:         fmt.Sprintf("%s_to_%s", fromID, toID),
		From:       fromID,
		To:         toID,
		Type:       flowType,
		Properties: make(map[string]any),
	}

	// Parse flow label
	if p.check(lexer.TokenColon) {
		p.advance() // consume ':'

		var labelParts []string
		for !p.check(lexer.TokenNewline) && !p.isAtEnd() {
			labelParts = append(labelParts, p.advance().Value)
		}

		if len(labelParts) > 0 {
			label := strings.TrimSpace(strings.Join(labelParts, " "))
			flow.Name = &label
		}
	}

	p.diagram.AddFlow(flow)
	return nil
}

// checkFlow checks if current position looks like a flow
func (p *BPMNParser) checkFlow() bool {
	if p.isAtEnd() {
		return false
	}
	token := p.peek()
	return token.Type == lexer.TokenArrowSolid ||
		token.Type == lexer.TokenArrowDotted ||
		token.Type == lexer.TokenMinus
}

// inferElementType infers BPMN element type from ID
func (p *BPMNParser) inferElementType(id string) ast.BPMNElementType {
	lowerID := strings.ToLower(id)

	// Infer type from common naming patterns
	if strings.Contains(lowerID, "start") {
		return ast.BPMNElementStartEvent
	}
	if strings.Contains(lowerID, "end") {
		return ast.BPMNElementEndEvent
	}
	if strings.Contains(lowerID, "gateway") || strings.Contains(lowerID, "decision") {
		return ast.BPMNElementExclusiveGateway
	}
	if strings.Contains(lowerID, "task") {
		return ast.BPMNElementTask
	}
	if strings.Contains(lowerID, "user") {
		return ast.BPMNElementUserTask
	}
	if strings.Contains(lowerID, "service") {
		return ast.BPMNElementServiceTask
	}

	// Default to task
	return ast.BPMNElementTask
}

// Helper methods
func (p *BPMNParser) check(tokenType lexer.TokenType) bool {
	if p.isAtEnd() {
		return false
	}
	return p.peek().Type == tokenType
}

func (p *BPMNParser) checkKeyword(keyword string) bool {
	if p.isAtEnd() {
		return false
	}
	token := p.peek()
	return token.Type == lexer.TokenID && strings.ToLower(token.Value) == strings.ToLower(keyword)
}

func (p *BPMNParser) advance() lexer.Token {
	if !p.isAtEnd() {
		p.current++
	}
	return p.previous()
}

func (p *BPMNParser) isAtEnd() bool {
	return p.current >= len(p.tokens) || p.peek().Type == lexer.TokenEOF
}

func (p *BPMNParser) peek() lexer.Token {
	if p.current >= len(p.tokens) {
		return lexer.Token{Type: lexer.TokenEOF}
	}
	return p.tokens[p.current]
}

func (p *BPMNParser) previous() lexer.Token {
	if p.current <= 0 {
		return lexer.Token{Type: lexer.TokenEOF}
	}
	return p.tokens[p.current-1]
}

func (p *BPMNParser) error(message string) error {
	token := p.peek()
	return fmt.Errorf("parse error at line %d, column %d: %s (got %s)",
		token.Line, token.Column, message, token.Type.String())
}