resume/internal/parser/parser.go

package parser

import (
	"archive/zip"
	"bytes"
	"encoding/xml"
	"fmt"
	"io"
	"path/filepath"
	"resume-platform/pkg/utils"
	"strings"
)

type Parser interface {
	Parse(fileName string, reader io.Reader) (string, error)
}

func NewParser(fileName string) (Parser, error) {
	ext := strings.ToLower(filepath.Ext(fileName))
	switch ext {
	case ".pdf":
		return &PDFParser{}, nil
	case ".docx":
		return &DOCXParser{}, nil
	case ".xlsx":
		return &ExcelParser{}, nil
	case ".md", ".markdown":
		return &MarkdownParser{}, nil
	case ".txt":
		return &TextParser{}, nil
	default:
		return &TextParser{}, nil
	}
}

func ParseFile(fileName string, reader io.Reader) (string, error) {
	data, err := io.ReadAll(reader)
	if err != nil {
		return "", err
	}
	
	parser, err := NewParser(fileName)
	if err != nil {
		return "", err
	}
	
	text, err := parser.Parse(fileName, bytes.NewReader(data))
	if err != nil {
		return "", err
	}
	
	if text == "" && len(data) > 0 {
		text = extractTextFromBinary(data)
	}
	
	return text, nil
}

func extractTextFromBinary(data []byte) string {
	content := string(data)
	var builder strings.Builder
	
	for _, r := range content {
		if r >= 0x20 && r <= 0x7E || r == '\n' || r == '\r' || r == '\t' ||
		   (r >= 0x4E00 && r <= 0x9FFF) || (r >= 0x3000 && r <= 0x303F) ||
		   (r >= 0xFF00 && r <= 0xFFEF) {
			builder.WriteRune(r)
		} else if r > 0 {
			builder.WriteRune(' ')
		}
	}
	
	return cleanText(builder.String())
}

type TextParser struct{}

func (p *TextParser) Parse(fileName string, reader io.Reader) (string, error) {
	data, err := io.ReadAll(reader)
	if err != nil {
		return "", err
	}
	return cleanText(string(data)), nil
}

type MarkdownParser struct{}

func (p *MarkdownParser) Parse(fileName string, reader io.Reader) (string, error) {
	data, err := io.ReadAll(reader)
	if err != nil {
		return "", err
	}
	return cleanText(string(data)), nil
}

type PDFParser struct{}

func (p *PDFParser) Parse(fileName string, reader io.Reader) (string, error) {
	data, err := io.ReadAll(reader)
	if err != nil {
		return "", err
	}

	content := string(data)
	if !strings.HasPrefix(content, "%PDF-") {
		return "", fmt.Errorf("not a valid PDF file")
	}

	text := extractTextFromPDF(content)
	return cleanText(text), nil
}

func extractTextFromPDF(content string) string {
	var builder strings.Builder
	inTextBlock := false
	inStream := false
	lines := strings.Split(content, "\n")

	for _, line := range lines {
		line = strings.TrimSpace(line)
		
		if strings.Contains(line, "/Length") && !inStream {
			continue
		}
		
		if strings.HasPrefix(line, "stream") {
			inStream = true
			continue
		}
		
		if strings.HasPrefix(line, "endstream") {
			inStream = false
			continue
		}
		
		if strings.Contains(line, "BT") {
			inTextBlock = true
			continue
		}
		
		if strings.Contains(line, "ET") {
			inTextBlock = false
			continue
		}
		
		if inTextBlock && !inStream {
			if strings.HasPrefix(line, "(") && strings.HasSuffix(line, ")") {
				text := line[1 : len(line)-1]
				text = strings.ReplaceAll(text, "\\n", "\n")
				text = strings.ReplaceAll(text, "\\r", "\r")
				text = strings.ReplaceAll(text, "\\t", "\t")
				text = strings.ReplaceAll(text, "\\(", "(")
				text = strings.ReplaceAll(text, "\\)", ")")
				text = strings.ReplaceAll(text, "\\/", "/")
				builder.WriteString(text)
				builder.WriteString(" ")
			} else if strings.HasPrefix(line, "[") && strings.HasSuffix(line, "]") {
				content := line[1 : len(line)-1]
				parts := strings.Fields(content)
				for _, part := range parts {
					if strings.HasPrefix(part, "(") && strings.HasSuffix(part, ")") {
						text := part[1 : len(part)-1]
						text = strings.ReplaceAll(text, "\\n", "\n")
						text = strings.ReplaceAll(text, "\\r", "\r")
						text = strings.ReplaceAll(text, "\\t", "\t")
						builder.WriteString(text)
						builder.WriteString(" ")
					}
				}
			}
		}
	}

	return strings.TrimSpace(builder.String())
}

type DOCXParser struct{}

func (p *DOCXParser) Parse(fileName string, reader io.Reader) (string, error) {
	data, err := io.ReadAll(reader)
	if err != nil {
		return "", err
	}

	fmt.Printf("DEBUG DOCXParser: file size: %d bytes\n", len(data))

	zipReader, err := zip.NewReader(bytes.NewReader(data), int64(len(data)))
	if err != nil {
		return "", fmt.Errorf("failed to open DOCX as zip: %w", err)
	}

	fmt.Printf("DEBUG DOCXParser: zip contains %d files\n", len(zipReader.File))

	var text strings.Builder
	foundDocument := false

	for _, file := range zipReader.File {
		fmt.Printf("DEBUG DOCXParser: zip file: %s (size: %d)\n", file.Name, file.UncompressedSize64)

		if file.Name == "word/document.xml" {
			foundDocument = true
			f, err := file.Open()
			if err != nil {
				return "", fmt.Errorf("failed to open document.xml: %w", err)
			}
			defer f.Close()

			xmlData, err := io.ReadAll(f)
			if err != nil {
				return "", fmt.Errorf("failed to read document.xml: %w", err)
			}

			fmt.Printf("DEBUG DOCXParser: document.xml size: %d bytes\n", len(xmlData))
			
			parsedText := parseDOCXXML(string(xmlData))
			fmt.Printf("DEBUG DOCXParser: parsed text from document.xml: %d chars\n", len(parsedText))
			
			text.WriteString(parsedText)
		} else if strings.HasPrefix(file.Name, "word/header") && strings.HasSuffix(file.Name, ".xml") {
			f, err := file.Open()
			if err != nil {
				continue
			}
			defer f.Close()
			xmlData, _ := io.ReadAll(f)
			text.WriteString(parseDOCXXML(string(xmlData)))
			text.WriteString("\n")
		} else if strings.HasPrefix(file.Name, "word/footer") && strings.HasSuffix(file.Name, ".xml") {
			f, err := file.Open()
			if err != nil {
				continue
			}
			defer f.Close()
			xmlData, _ := io.ReadAll(f)
			text.WriteString(parseDOCXXML(string(xmlData)))
			text.WriteString("\n")
		}
	}

	if !foundDocument {
		fmt.Println("WARN DOCXParser: document.xml not found in zip")
		return cleanText(extractTextFromBinary(data)), nil
	}

	result := cleanText(text.String())
	fmt.Printf("DEBUG DOCXParser: final result length: %d chars\n", len(result))
	
	if len(result) > 0 {
		fmt.Printf("DEBUG DOCXParser: first 500 chars:\n%s\n", result[:utils.Min(len(result), 500)])
	}

	return result, nil
}

func parseDOCXXML(xmlContent string) string {
	fmt.Printf("DEBUG parseDOCXXML: input length: %d chars\n", len(xmlContent))

	xmlContent = removeNamespacePrefixes(xmlContent)

	allTexts := extractTextContent(xmlContent)

	var builder strings.Builder
	for _, text := range allTexts {
		text = strings.TrimSpace(text)
		if text != "" {
			builder.WriteString(text)
			builder.WriteString("\n")
		}
	}

	result := cleanText(builder.String())
	fmt.Printf("DEBUG parseDOCXXML: extracted %d text elements, result length: %d\n", len(allTexts), len(result))

	if len(result) > 0 {
		fmt.Printf("DEBUG parseDOCXXML: first 500 chars:\n%s\n", result[:utils.Min(len(result), 500)])
	}

	if result == "" {
		fmt.Println("WARN: Structured parsing returned empty, using fallback")
		return cleanText(extractPlainTextFromXML(xmlContent))
	}

	return result
}

func extractTextContent(xmlContent string) []string {
	var texts []string
	
	inTag := false
	currentText := ""
	
	for _, r := range xmlContent {
		if r == '<' {
			inTag = true
			if currentText != "" {
				texts = append(texts, currentText)
				currentText = ""
			}
			continue
		}
		if r == '>' {
			inTag = false
			continue
		}
		if !inTag {
			currentText += string(r)
		}
	}
	
	if currentText != "" {
		texts = append(texts, currentText)
	}
	
	return texts
}

func removeNamespacePrefixes(xmlContent string) string {
	prefixes := []string{"w:", "r:", "wp:", "wpc:", "wps:", "wpg:", "wpi:", "wne:", "mc:", "o:", "v:", "w10:", "wpc:", "wpg:", "wps:", "a:", "m:"}
	for _, prefix := range prefixes {
		xmlContent = strings.ReplaceAll(xmlContent, "<"+prefix, "<")
		xmlContent = strings.ReplaceAll(xmlContent, "</"+prefix, "</")
		xmlContent = strings.ReplaceAll(xmlContent, " "+prefix, " ")
	}
	return xmlContent
}

func extractPlainTextFromXML(xmlContent string) string {
	var builder strings.Builder
	inTag := false
	
	for _, r := range xmlContent {
		if r == '<' {
			inTag = true
			continue
		}
		if r == '>' {
			inTag = false
			continue
		}
		if !inTag {
			builder.WriteRune(r)
		}
	}
	
	return builder.String()
}

type ExcelParser struct{}

func (p *ExcelParser) Parse(fileName string, reader io.Reader) (string, error) {
	data, err := io.ReadAll(reader)
	if err != nil {
		return "", err
	}

	zipReader, err := zip.NewReader(bytes.NewReader(data), int64(len(data)))
	if err != nil {
		return "", fmt.Errorf("failed to open XLSX as zip: %w", err)
	}

	var text strings.Builder
	for _, file := range zipReader.File {
		if strings.HasPrefix(file.Name, "xl/worksheets/") && strings.HasSuffix(file.Name, ".xml") {
			sheetName := strings.TrimPrefix(file.Name, "xl/worksheets/sheet")
			sheetName = strings.TrimSuffix(sheetName, ".xml")
			text.WriteString("===" + sheetName + "===\n\n")

			f, err := file.Open()
			if err != nil {
				return "", fmt.Errorf("failed to open worksheet: %w", err)
			}
			defer f.Close()

			xmlData, err := io.ReadAll(f)
			if err != nil {
				return "", fmt.Errorf("failed to read worksheet: %w", err)
			}

			text.WriteString(parseXLSXXML(string(xmlData)))
			text.WriteString("\n\n")
		}
	}

	return cleanText(text.String()), nil
}

func parseXLSXXML(xmlContent string) string {
	type Cell struct {
		Text string `xml:",chardata"`
	}
	type Row struct {
		Cells []Cell `xml:"c"`
	}
	type SheetData struct {
		Rows []Row `xml:"row"`
	}
	type Worksheet struct {
		SheetData SheetData `xml:"sheetData"`
	}

	var ws Worksheet
	err := xml.Unmarshal([]byte(xmlContent), &ws)
	if err != nil {
		return ""
	}

	var builder strings.Builder
	for _, row := range ws.SheetData.Rows {
		var cells []string
		for _, cell := range row.Cells {
			text := strings.TrimSpace(cell.Text)
			if text != "" {
				cells = append(cells, text)
			}
		}
		if len(cells) > 0 {
			builder.WriteString(strings.Join(cells, "\t"))
			builder.WriteString("\n")
		}
	}

	return strings.TrimSpace(builder.String())
}

func cleanText(text string) string {
	return utils.CleanText(text)
}