Java实现HTML转PDF(deepSeekAi-＞html-＞pdf)

Java实现HTML转PDF,主要为了解决将ai返回的html文本数据转为PDF文件方便用户下载查看。

一、deepSeek-AI提问词

基于以上个人数据。总结个人身体信息，分析个人身体指标信息。再按一个月为维度，详细列举一个月内训练计划，维度详细至每周每天，要求：不可省略表格内容以精简示例，文本结构顺序为标题个人信息，第一步，第二步。最终回答结果以标准的html形式返回结果,不能带有meta标签，字体为STSong-Light,SimSun,html内容禁止使用单标签。

二、表设计

CREATE TABLE `p_deep_seek_task` (
  `id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '任务id',
  `user_id` bigint(20) NOT NULL COMMENT '用户id',
  `status` char(2) NOT NULL DEFAULT '0' COMMENT '任务状态(0:待处理,1:处理中,2:处理成功,3:异常处理失败,4:重试中,5:重试失败)',
  `try_time` int(11) NOT NULL DEFAULT '0' COMMENT '执行次数',
  `result_url` varchar(255) DEFAULT NULL COMMENT '结果文件url',
  `prompt` longtext NOT NULL COMMENT '提问内容',
  `content` longtext COMMENT '结果内容',
  `reasoning_content` longtext COMMENT '思考过程',
  `create_time` datetime DEFAULT NULL COMMENT '创建时间',
  `create_by` bigint(20) DEFAULT NULL COMMENT '创建人',
  `update_by` bigint(20) DEFAULT NULL COMMENT '更新人',
  `update_time` datetime DEFAULT NULL COMMENT '更新时间',
  `task_time` date DEFAULT NULL COMMENT '任务日期',
  `execute_time` datetime DEFAULT NULL COMMENT '执行时间',
  `exception_msg` longtext COMMENT '异常信息',
  `cost_time` bigint(20) NOT NULL DEFAULT '0' COMMENT '执行耗时(s)'
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=16 DEFAULT CHARSET=utf8mb4 COMMENT='deepseek任务';

三、导入Jar包

        <dependency>
            <groupId>io.github.pig-mesh.ai</groupId>
            <artifactId>deepseek-spring-boot-starter</artifactId>
        </dependency>

        <dependency>
            <groupId>org.xhtmlrenderer</groupId>
            <artifactId>flying-saucer-pdf</artifactId>
            <version>9.1.22</version> 
        </dependency>
        <dependency>
            <groupId>org.jsoup</groupId>
            <artifactId>jsoup</artifactId>
            <version>1.14.3</version> 
        </dependency>

四、工具类

ai请求工具

package com.company.project.service.client;

import io.github.pigmesh.ai.deepseek.config.DeepSeekProperties;
import io.github.pigmesh.ai.deepseek.core.DeepSeekClient;
import io.github.pigmesh.ai.deepseek.core.chat.ChatCompletionRequest;
import io.github.pigmesh.ai.deepseek.core.chat.ChatCompletionResponse;
import org.springframework.stereotype.Service;

import javax.annotation.Resource;

/**
 * @author: reshui
 * description:DeepSeek服务
 * DateTime:2025/3/31-14:48
 */
@Service
public class DeepSeekAiClient {

    @Resource
    private DeepSeekClient deepSeekClient;
    @Resource
    private DeepSeekProperties deepSeekProperties;

    /**
     * 提问接口
     * 获取deepseek的响应结果
     * @param prompt 提示词
     */
    public ChatCompletionResponse syncChat(String prompt) {
        ChatCompletionRequest request = ChatCompletionRequest.builder()
                // 根据渠道模型名称动态修改这个参数
                .model(deepSeekProperties.getModel()).addUserMessage(prompt).build();
        return deepSeekClient.chatCompletion(request).execute();
    }

}

特定html字符内容过滤工具

package com.company.project.service.tools;

import cn.hutool.core.collection.CollUtil;
import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;

import java.util.regex.Matcher;
import java.util.regex.Pattern;

/**
 * @author reshui
 * description
 * dateTime 2025/04/17
 */
public class HtmlFormatter {

    private String htmlContent;

    private HtmlFormatter(String htmlContent) {
        this.htmlContent = htmlContent;
    }

    // 入口方法，创建处理器实例
    public static HtmlFormatter process(String htmlContent) {
        return new HtmlFormatter(htmlContent);
    }

    // 链式方法：前置过滤干扰字符
    public HtmlFormatter beforeFilter() {
        this.htmlContent = beforeFilterInterferenceCharacters(this.htmlContent);
        return this;
    }

    // 链式方法：替换内容标签
    public HtmlFormatter replaceTags() {
        this.htmlContent = replaceContentTag(this.htmlContent);
        return this;
    }

    // 链式方法：后置过滤干扰字符
    public HtmlFormatter afterFilter() {
        this.htmlContent = afterFilterInterferenceCharacters(this.htmlContent);
        return this;
    }

    // 获取最终结果
    public String get() {
        return this.htmlContent;
    }
    public static String formatHtml(String htmlContent) {
        // 过滤掉html中的干扰字符
        String filteredHtml = beforeFilterInterferenceCharacters(htmlContent);
        // 去除内容中的大于小于号干扰
        String replaceContentTag = replaceContentTag(filteredHtml);
        //过滤html中的干扰标签
        return afterFilterInterferenceCharacters(replaceContentTag);
    }

    /**
     * 替换html中的干扰内容
     *
     * @param html 文本
     */
    public static String replaceContentTag(String html) {
        Document doc = Jsoup.parse(html);
        removeTag(doc);
        traverse(doc.body());
        doc.outputSettings().prettyPrint(false);
        return doc.html();
    }

    /**
     * 去除不支持的meta标签
     * @param doc jsoupdoc
     */
    public static void removeTag(Document doc) {
        Elements meta = doc.getElementsByTag("meta");
        for (Element metaElement : meta) {
            metaElement.remove();
        }
    }
    public static void traverse(Element element) {
        if (CollUtil.isEmpty(element.children())) {
            String text = element.text()
                    .replace("<", "小于")
                    .replace(">", "大于");
            element.text(text);
        }
        for (Element child : element.children()) { // 遍历子元素
            traverse(child); // 递归调用以处理子元素及其子元素
        }
    }


    /**
     * 后置过滤掉html中的干扰字符
     *
     * @param html 文本
     */
    public static String afterFilterInterferenceCharacters(String html) {
        return html
                .replace("<br></br>", "<br/>")
                .replace("<br>", "<br/>")
                .replace("</br>", "<br/>")
                ;
    }

    /**
     * 前置过滤掉html中的干扰字符
     *
     * @param html 文本
     */
    public static String beforeFilterInterferenceCharacters(String html) {
        return html
                .replace("```html", "")
                .replace("```", "")
                .replace("<!DOCTYPE html>", "")
                .replace("<!doctype html>", "")
                ;
    }

    /**
     * 将HTML字符串中的所有标签转为小写
     *
     * @param html 原始HTML字符串
     * @return 转换后的HTML字符串
     */
    public static String convertTagsToLowerCase(String html) {
        // 正则表达式匹配HTML标签
        Pattern pattern = Pattern.compile("</?\\w+((\\s+\\w+(\\s*=\\s*(?:\".*?\"|'.*?'|[^'\">\\s]+))?)+\\s*|\\s*)/?>");
        Matcher matcher = pattern.matcher(html);
        StringBuffer result = new StringBuffer();
        while (matcher.find()) {
            // 将匹配到的标签转为小写
            String lowerCaseTag = matcher.group().toLowerCase();
            matcher.appendReplacement(result, lowerCaseTag);
        }
        matcher.appendTail(result);
        return result.toString();
    }

}

html转pdf工具

package com.company.project.service.tools;

import cn.hutool.core.date.DateUtil;
import cn.hutool.core.io.FileUtil;
import com.company.project.common.utils.SpringUtils;
import com.company.project.service.properties.PdfFontProperties;
import lombok.extern.slf4j.Slf4j;
import org.xhtmlrenderer.pdf.ITextRenderer;

import java.io.File;
import java.io.OutputStream;
import java.nio.file.Files;
import java.nio.file.Paths;
import java.util.Date;

/**
 * @author: reshui
 * description: html转pdf工具类
 * DateTime:2025/3/31-15:55
 */
@Slf4j
public class HtmlToPdfTools {

    /**
     * 文件暂存地址
     */
    private static final String TEMP_FILE_PATH = System.getProperty("java.io.tmpdir");

    /**
     * pdf文件暂存地址
     */
    private static final String PDF_FILE_PATH = TEMP_FILE_PATH + File.separator + "ai_train_pdf";

    /**
     * 时间格式
     */
    private static final String TIMESTAMP_FORMAT = "yyyyMMddHHmmss";

    /**
     * pdf配置文件
     */
    public final static PdfFontProperties CONFIG = SpringUtils.getBean(PdfFontProperties.class);

    /**
     * html转pdf文件
     *
     * @param htmlContent html内容文本
     */
    public static File convertHtmlToPdfFile(String htmlContent) throws Exception {
        String formatDateTimeStamp = DateUtil.format(new Date(), TIMESTAMP_FORMAT);
        String pdfFilePath = PDF_FILE_PATH + File.separator + formatDateTimeStamp + ".pdf";
        FileUtil.touch(pdfFilePath);
        String resultHtmlContent = HtmlFormatter.process(htmlContent)
                .beforeFilter()
                .replaceTags()
                .afterFilter()
                .get();
        generatePdfReport(pdfFilePath, resultHtmlContent);
        log.info("pdf文件储存地址：{}", pdfFilePath);
        return new File(pdfFilePath);
    }

    /**
     * 生成pdf文件
     *
     * @param outputPath  输出文件地址
     * @param htmlContent html内容文本
     */
    public static void generatePdfReport(String outputPath, String htmlContent) throws Exception {
        try (OutputStream outputStream = Files.newOutputStream(Paths.get(outputPath))) {
            ITextRenderer renderer = new ITextRenderer();
            renderer.getFontResolver().addFont(
                    CONFIG.getPath(),
                    CONFIG.getEncoding(),
                    CONFIG.getEmbedded()
            );
//            renderer.getFontResolver().addFont(
//                    "c://Windows//Fonts//simsun.ttc",
//                    "Identity-H",
//                    true
//            );
            renderer.setDocumentFromString(htmlContent);
            renderer.layout();
            renderer.createPDF(outputStream);
        }
    }


    public static void main(String[] args) throws Exception {
        String html = "";
        convertHtmlToPdfFile(html);
    }
}

配置文件yml

# deepseek配置文件
deepseek:
  base-url: https://api.deepseek.com/v1
  api-key: xxxxxxxxxxxxx
  model: deepseek-reasoner
  connectTimeout: 60
  readTimeout: 240
  callTimeout: 360

# windows-pdf字体配置
pdf:
    font:
      path: c://Windows//Fonts//simsun.ttc
      encoding: Identity-H
      embedded: true

# linux-pdf字体配置
pdf:
    font:
      path: c://Windows//Fonts//simsun.ttc
      encoding: Identity-H
      embedded: true