<repositories>
    <repository>
        <id>e-iceblue</id>
        <url>https://repo.e-iceblue.cn/repository/maven-public/</url>
    </repository>
</repositories>
<dependencies>
    <dependency>
        <groupId>e-iceblue</groupId>
        <artifactId>spire.pdf</artifactId>
        <version>10.X.X</version> <!-- 請?zhí)鎿Q為最新版本號 -->
    </dependency>
</dependencies>

repositories {
    maven { url 'https://repo.e-iceblue.cn/repository/maven-public/' }
}
dependencies {
    implementation 'e-iceblue:spire.pdf:10.X.X' // 請?zhí)鎿Q為最新版本號
}

import com.spire.pdf.PdfDocument;
import com.spire.pdf.utilities.PdfTable;
import com.spire.pdf.utilities.PdfTableExtractor;

import java.io.FileWriter;
import java.io.IOException;
import java.util.List;

public class PdfTableToCsvConverter {

    public static void main(String[] args) {
        // 1. 加載 PDF 文檔
        String pdfFilePath = "data/tableSample.pdf"; // 替換為你的 PDF 文件路徑
        String csvOutputFilePath = "output/extracted_table.csv"; // CSV 輸出路徑

        PdfDocument pdfDocument = new PdfDocument();
        try {
            pdfDocument.loadFromFile(pdfFilePath);
            System.out.println("PDF 文檔加載成功：" + pdfFilePath);

            // 2. 創(chuàng)建 PdfTableExtractor 實例
            PdfTableExtractor extractor = new PdfTableExtractor(pdfDocument);

            // 使用 FileWriter 寫入 CSV 文件
            try (FileWriter csvWriter = new FileWriter(csvOutputFilePath)) {
                // 遍歷 PDF 的每一頁
                for (int pageIndex = 0; pageIndex < pdfDocument.getPages().getCount(); pageIndex++) {
                    System.out.println("正在處理第 " + (pageIndex + 1) + " 頁...");

                    // 3. 識別并提取表格
                    // extractTable(int pageIndex) 方法返回當前頁的所有表格
                    PdfTable[] tableLists = extractor.extractTable(pageIndex);

                    if (tableLists != null && tableLists.length > 0) {
                        for (PdfTable table : tableLists) {
                            System.out.println("  發(fā)現(xiàn)表格，行數(shù)：" + table.getRowCount() + ", 列數(shù)：" + table.getColumnCount());

                            // 4. 將提取到的數(shù)據(jù)寫入 CSV
                            for (int row = 0; row < table.getRowCount(); row++) {
                                StringBuilder rowData = new StringBuilder();
                                for (int column = 0; column < table.getColumnCount(); column++) {
                                    // 獲取單元格文本
                                    String cellText = table.getText(row, column);
                                    // 對包含逗號或雙引號的文本進行處理，避免 CSV 格式錯誤
                                    if (cellText.contains(",") || cellText.contains(""")) {
                                        cellText = """ + cellText.replace(""", """") + """;
                                    }
                                    rowData.append(cellText);
                                    if (column < table.getColumnCount() - 1) {
                                        rowData.append(","); // 添加逗號分隔
                                    }
                                }
                                csvWriter.append(rowData.toString()).append("\n"); // 寫入一行數(shù)據(jù)并換行
                            }
                            // 每個表格之間可以添加一個空行，或者其他分隔符，以便區(qū)分
                            csvWriter.append("\n"); 
                        }
                    } else {
                        System.out.println("  第 " + (pageIndex + 1) + " 頁未檢測到表格。");
                    }
                }
                System.out.println("數(shù)據(jù)已成功提取并保存到：" + csvOutputFilePath);
            } catch (IOException e) {
                System.err.println("寫入 CSV 文件時發(fā)生錯誤：" + e.getMessage());
            }

        } catch (Exception e) {
            System.err.println("處理 PDF 文件時發(fā)生錯誤：" + e.getMessage());
            e.printStackTrace();
        } finally {
            if (pdfDocument != null) {
                pdfDocument.close(); // 關閉文檔，釋放資源
                pdfDocument.dispose();
            }
        }
    }
}

for (int i = 0; i < pdf.getPages().getCount(); i++) {
    PdfTableExtractor extractor = new PdfTableExtractor(pdf);
    PdfTable[] tableLists = extractor.extractTable(i);

    if (tableLists != null) {
        for (int t = 0; t < tableLists.length; t++) {
            PdfTable table = tableLists[t];
            StringBuilder tableContent = new StringBuilder();

            for (int row = 0; row < table.getRowCount(); row++) {
                for (int col = 0; col < table.getColumnCount(); col++) {
                    tableContent.append(escapeCsvField(table.getText(row, col)));
                    if (col < table.getColumnCount() - 1) {
                        tableContent.append(",");
                    }
                }
                tableContent.append("\n");
            }

            try (Writer writer = new OutputStreamWriter(
                    new FileOutputStream("output/Tables/Table_Page" + i + "_Index" + t + ".csv"), "UTF-8")) {
                writer.write(sb.toString());
            }
        }
    }
}

StringBuilder sb = new StringBuilder();

for (int i = 0; i < pdf.getPages().getCount(); i++) {
    PdfTableExtractor extractor = new PdfTableExtractor(pdf);
    PdfTable[] tables = extractor.extractTable(i);

    if (tables != null) {
        for (PdfTable table : tables) {
            for (int row = 0; row < table.getRowCount(); row++) {
                for (int col = 0; col < table.getColumnCount(); col++) {
                    sb.append(escapeCsvField(table.getText(row, col)));
                    if (col < table.getColumnCount() - 1) sb.append(",");
                }
                sb.append("\n");
            }
        }
    }
}

FileWriter writer = new FileWriter("MergedTables.csv");
writer.write(sb.toString());
writer.close();