项目需求

使用velocity模板生成html内容，将含有标题 / 段落 / 表格 / 图片 的html文件转化成PDF和word格式，并保留良好样式。

iText

起初只想实现word功能，但经过尝试和查询一些例如poi等工具，都无法较好的保留原有的复杂样式，后更换pdf需求进行尝试，itext生成pdf已经非常完美了。

生成word样式没有尝试，感觉比较麻烦。

后来采用spire的free包将pdf转word样式保持很好，但是项目环境中遇到问题，转战html转word。

注意：不要将spire的pdf和word包同时引用，如果需要可以直接使用office包，会产生一些例如注册许可方面的冲突。

spire.pdf / spire.doc / spire.office

很好的第三方工具，能完美实现html转pdf和word等格式文件，最关键的是有free版本，虽然有10页的限制，结合需求页可能够用和付费版功能上无差别。

付费版本也一样可以用，只是会在左上角有红色警告信息，其实word可以手动删除，pdf可以通过添加图片的方式遮盖掉。

注意：我们的项目环境是Linux / jdk1.7 / weblogic12.1.3.0.0（pdf转word–使用doc包就会报错UnkownSource【原因不明】，最好直接html转word最终解决问题），html转word表头的css样式可能会无法解析，使用基本样式就好。

spire.pdf / spire.doc 冰蓝科技 中文文档

html转pdf

引入jar包

引入itext仓库

<repository><id>itext</id><name>iText Repository - releases</name><url>/releases</url></repository>

在pom.xml引入maven（不包括spire.pdf）

<!-- itextpdf --><dependency><groupId>com.lowagie</groupId><artifactId>itext</artifactId><version>2.1.7</version></dependency><dependency><groupId>com.lowagie</groupId><artifactId>itext-rtf</artifactId><version>2.1.7</version></dependency><dependency><groupId>com.itextpdf</groupId><artifactId>itext-asian</artifactId><version>5.2.0</version></dependency><dependency><groupId>com.itextpdf</groupId><artifactId>itextpdf</artifactId><version>5.5.10</version></dependency><!-- pdfHTML --><dependency><groupId>com.itextpdf</groupId><artifactId>itext7-core</artifactId><version>7.1.0</version><type>pom</type><exclusions><exclusion><artifactId>slf4j-api</artifactId><groupId>org.slf4j</groupId></exclusion></exclusions></dependency><dependency><groupId>com.itextpdf</groupId><artifactId>html2pdf</artifactId><version>2.0.0</version></dependency><dependency><groupId>mons</groupId><artifactId>commons-io</artifactId><version>1.3.2</version></dependency><dependency><groupId>org.xhtmlrenderer</groupId><artifactId>flying-saucer-pdf</artifactId><version>9.0.8</version></dependency></dependencies>

可以下载spire.pdf.free以及各种实用jar包的maven仓库

maven仓库（内含spire.pdf.free和spire.doc.free）

生成pdf

package com.sy.util;import com.itextpdf.html2pdf.ConverterProperties;import com.itextpdf.html2pdf.HtmlConverter;import com.itextpdf.html2pdf.attach.impl.DefaultTagWorkerFactory;import com.itextpdf.html2pdf.css.media.MediaDeviceDescription;import com.itextpdf.html2pdf.css.media.MediaType;import com.itextpdf.kernel.pdf.*;import com.itextpdf.layout.font.FontProvider;import mons.io.FileUtils;import org.apache.poi.util.IOUtils;import java.io.*;import static com.itextpdf.kernel.geom.PageSize.A4;import static com.itextpdf.kernel.geom.PageSize.A3;public class PdfService {//private static final Logger LOGGER = LoggerFactory.getLogger(PdfFromHTML.class);public static final String FONT_NAME = "simsun.ttf";public void copyFont(String targetDir,String fontPath) {FileOutputStream fout = null;FileInputStream fileInputStream = null;try {//InputStream inputStream = this.getClass().getClassLoader().getResourceAsStream(FONT_CLASS_PATH);fileInputStream = new FileInputStream(fontPath+"/WEB-INF/fonts/simsun.ttf");File dir = new File(targetDir);FileUtils.forceMkdir(dir);fout = new FileOutputStream(targetDir + "simsun.ttf");if (fileInputStream != null && fout != null) {mons.io.IOUtils.copy(fileInputStream, fout);}} catch (Exception e) {//LOGGER.error("failed to copy font: ", e);e.printStackTrace();} finally {if(null!=fileInputStream){try {fileInputStream.close();} catch (IOException e) {e.printStackTrace();}}if(null!=fout){try {fout.close();} catch (IOException e) {e.printStackTrace();}}}}public void createPdfFromHtml(String pdfFileName, InputStream htmlInputStream, String resourcePrefix){PdfDocument pdfDoc = null;FileOutputStream outputStream = null;PdfWriter pdfWriter = null;try {outputStream = new FileOutputStream(resourcePrefix + pdfFileName);WriterProperties writerProperties = new WriterProperties();writerProperties.addXmpMetadata();pdfWriter = new PdfWriter(outputStream, writerProperties);pdfDoc = createPdfDoc(pdfWriter);PdfDocument pdfDocument = new PdfDocument(pdfWriter);pdfDocument.setDefaultPageSize(A3);//转换器ConverterProperties props = createConverterProperties(resourcePrefix);props.setBaseUri(resourcePrefix);//设置媒体的描述方式MediaDeviceDescription mediaDeviceDescription = new MediaDeviceDescription(MediaType.SCREEN);props.setMediaDeviceDescription(mediaDeviceDescription);HtmlConverter.convertToPdf(htmlInputStream,pdfDocument, props);pdfDocument.close();} catch (Exception e) {//LOGGER.error("failed to create pdf from html exception: ", e);e.printStackTrace();} finally {try {outputStream.close();pdfWriter.close();} catch (IOException e) {e.printStackTrace();}finally {outputStream = null;pdfWriter = null;}IOUtils.closeQuietly(pdfDoc);}}private PdfDocument createPdfDoc(PdfWriter pdfWriter) {PdfDocument pdfDoc;pdfDoc = new PdfDocument(pdfWriter);pdfDoc.getCatalog().setLang(new PdfString("zh-CN"));pdfDoc.setTagged();pdfDoc.getCatalog().setViewerPreferences(new PdfViewerPreferences().setDisplayDocTitle(true));return pdfDoc;}private ConverterProperties createConverterProperties(String resourcePrefix) {ConverterProperties props = new ConverterProperties();props.setFontProvider(createFontProvider(resourcePrefix));props.setBaseUri(resourcePrefix);props.setCharset("UTF-8");DefaultTagWorkerFactory tagWorkerFactory = new DefaultTagWorkerFactory();props.setTagWorkerFactory(tagWorkerFactory);return props;}private FontProvider createFontProvider(String resourcePrefix) {FontProvider fp = new FontProvider();fp.addStandardPdfFonts();fp.addDirectory(resourcePrefix);return fp;}}

下面为main方法

//pdf文件加载和生成统一路径String pdfPath = localFilePath + name + ".pdf";File temFile = new File(localFilePath);if (!temFile.exists()) {temFile.mkdirs();}// 添加字体,这一步很关键，否则中文会显示不出来PdfService pdfService = new PdfService();File file = new File(localFilePath + PdfService.FONT_NAME);if (!file.exists()) {//加载字体库pdfService.copyFont(localFilePath,realPath);}//---------------------FileInputStream inputStream = new FileInputStream(url);File pdfFile = new File(pdfPath);//TODO html To PDFpdfService.createPdfFromHtml(pdfFile.getName(), inputStream, localFilePath);

pdf布局设置（以下内容通过spire.doc/pdf包处理）

package com.sy.util;import com.spire.doc.FileFormat;import com.spire.doc.documents.*;import com.spire.doc.fields.DocPicture;import com.spire.doc.fields.TextRange;import com.spire.pdf.*;import com.spire.doc.*;import com.spire.pdf.annotations.PdfRubberStampAnnotation;import com.spire.pdf.annotations.appearance.PdfAppearance;import com.spire.pdf.automaticfields.PdfCompositeField;import com.spire.pdf.automaticfields.PdfPageCountField;import com.spire.pdf.automaticfields.PdfPageNumberField;import com.spire.pdf.graphics.*;import java.awt.*;import java.awt.geom.Dimension2D;import java.awt.geom.Point2D;import java.awt.geom.Rectangle2D;import com.spire.doc.FileFormat.*;public class PdfPageSetting {public static final int FLAG = 2;public void pdfPageSetting(String url,int flag){//创建PdfDocument对象PdfDocument originalDoc = new PdfDocument();//加载PDF文件originalDoc.loadFromFile(url);//第一页插入空白页（目的是删除spirefree的内置红色警告语，主要是为了生成word使用，pdf可以用添加图片方法进行遮盖）originalDoc.getPages().insert(0);//创建一个新的PdfDocument实例PdfDocument newDoc = new PdfDocument();//遍历所有PDF 页面Dimension2D dimension2D = new Dimension();for (int i = 0; i < originalDoc.getPages().getCount(); i++) {PdfPageBase page = originalDoc.getPages().get(i);if (flag==1){//设置新文档页面宽、高为原来的1倍float scale = 1.0f;float width = (float) page.getSize().getWidth() * scale;float height = (float) page.getSize().getHeight() * scale;dimension2D.setSize(width, height);//设置新文档第一页的页边距为左右PdfMargins margins = new PdfMargins(0, 20);PdfPageBase newPage = newDoc.getPages().add(dimension2D, margins);//复制原文档的内容到新文档newPage.getCanvas().drawTemplate(page.createTemplate(), new Point2D.Float());}if (flag==2){//设置新文档页边距为左右0、上下20(不能随意更改，会影响主标题位置)PdfMargins margins = new PdfMargins(0,20,0,0);//设置新文档页面大小为A3PdfPageBase newPage = newDoc.getPages().add(PdfPageSize.A3, margins);//调整画布，设置内容也根据页面的大小进行缩放double wScale = (PdfPageSize.A3.getWidth() - 10) / PdfPageSize.A3.getWidth();double hScale = (PdfPageSize.A3.getHeight() - 10) / PdfPageSize.A3.getHeight();newPage.getCanvas().translateTransform(wScale, hScale);//复制原文档的内容到新文档newPage.getCanvas().drawTemplate(page.createTemplate(), new Point2D.Float());}if (flag==3){//设置新文档页边距PdfMargins margins = new PdfMargins(0, 0);//设置新文档页面大小为A3, 页面旋转角度为0，纸张方向为水平PdfPageBase newPage = newDoc.getPages().add(PdfPageSize.A4, margins, PdfPageRotateAngle.Rotate_Angle_0, PdfPageOrientation.Landscape);//调整画布，设置内容也根据页面的大小进行缩放double wScale = PdfPageSize.A4.getHeight() / page.getSize().getWidth();double hScale = PdfPageSize.A4.getWidth() / page.getSize().getHeight();newPage.getCanvas().translateTransform(wScale, hScale);//复制原文档的内容到新文档newPage.getCanvas().drawTemplate(page.createTemplate(), new Point2D.Float());}}//保存PDFnewDoc.saveToFile(url);newDoc.close();originalDoc.close();}

pdf添加页码

/*** 添加页码* @param targetPath*/public void addPageNum(String targetPath) {//加载PDF文档PdfDocument doc = new PdfDocument();doc.loadFromFile(targetPath);//删除第一页空白页doc.getPages().removeAt(0);//创建字体PdfTrueTypeFont font = new PdfTrueTypeFont(new Font("Default", Font.PLAIN, 10),true);//获取页面尺寸Dimension2D pageSize = doc.getPages().get(0).getSize();//初始化y坐标float y = (float) pageSize.getHeight() - 20;//遍历文档中的页for (int i = 0; i < doc.getPages().getCount(); i++) {//初始化页码域PdfPageNumberField number = new PdfPageNumberField();//初始化总页数域//PdfPageCountField count = new PdfPageCountField();//创建复合域PdfCompositeField compositeField = new PdfCompositeField(font, PdfBrushes.getBlack(),"{0}", number);//设置复合域内文字对齐方式compositeField.setStringFormat(new PdfStringFormat(PdfTextAlignment.Right, PdfVerticalAlignment.Top));//测量文字大小Dimension2D textSize = font.measureString(compositeField.getText());//设置复合域的在PDF页面上的位置及大小compositeField.setBounds(new Rectangle2D.Float(((float) pageSize.getWidth() - (float) textSize.getWidth())/2, y, (float) textSize.getWidth(), (float) textSize.getHeight()));//将复合域添加到PDF页面compositeField.draw(doc.getPages().get(i).getCanvas());}//保存为另外一个文档doc.saveToFile(targetPath);doc.close();}

pdf添加图片

/*** 添加图片到PDF* @param url pdf文件* @param baseUrl 项目路径* @param savePath 文件保存路径*/public static void addImgToPdf(String url,String baseUrl) {//创建PdfDocument对象PdfDocument doc = new PdfDocument();//加载PDF文档doc.loadFromFile(url);//删除第一页空白页//doc.getPages().removeAt(0);for (int i = 0; i < doc.getPages().getCount(); i++) {//获取第i页PdfPageBase page = doc.getPages().get(i);//加载图片到PdfImage对象PdfImage image = PdfImage.fromFile(baseUrl + "img\\custom\\pdf.png");//获取图片高宽int width = image.getWidth();int height = image.getHeight();//创建PdfTemplate对象，大小跟图片一致PdfTemplate template = new PdfTemplate(width, height);//在模板上绘制图片template.getGraphics().drawImage(image, 0, 0, width, height);//创建PdfRubebrStampAnnotation对象，指定大小和位置Rectangle2D rect = new Rectangle2D.Float((float)(0), (float)(0), 835, 50);PdfRubberStampAnnotation stamp = new PdfRubberStampAnnotation(rect);//创建PdfAppearance对象PdfAppearance pdfAppearance = new PdfAppearance(stamp);//将模板应用为PdfAppearance的一般状态pdfAppearance.setNormal(template);//将PdfAppearance 应用为图章的样式stamp.setAppearance(pdfAppearance);//添加图章到PDFpage.getAnnotationsWidget().add(stamp);}//保存文档doc.saveToFile(url);doc.close();}

pdf转word（windows/linux和tomcat下测试可用）

//创建PdfDocument对象PdfDocument doc = new PdfDocument();//加载PDF文档doc.loadFromFile(url);//保存文档doc.saveToFile(url);

html转word

//TODO html To DOCXDocument word = new Document();//html文件地址word.loadFromFile(url);word.saveToFile(wordPath, FileFormat.Docx);word.close();//TODO 添加页眉和页脚，设置word页边距(realPath为页眉图片地址)setUpPageOfWord(wordPath,realPath)；

word添加页眉页脚和页码

/*** word添加页眉* @param url* @param baseUrl* @param*/public static void setUpPageOfWord(String url,String baseUrl) {//创建Document对象Document doc = new Document();doc.loadFromFile(url);//获取第一页Section section = doc.getSections().get(0);Dimension2D dimension2D = new Dimension();dimension2D.setSize(800,1200);section.getPageSetup().setPageSize(dimension2D);//设置页边距section.getPageSetup().getMargins().setTop(40f);section.getPageSetup().getMargins().setBottom(40f);section.getPageSetup().getMargins().setLeft(60f);section.getPageSetup().getMargins().setRight(60f);//调用insertHeaderAndFooter方法插入页眉页脚到第一个sectioninsertHeaderAndFooter(section,baseUrl);//保存doc.saveToFile(url);doc.close();}private static void insertHeaderAndFooter(Section section,String baseUrl) {//分别获取section的页眉页脚HeaderFooter header = section.getHeadersFooters().getHeader();HeaderFooter footer = section.getHeadersFooters().getFooter();//添加段落到页眉Paragraph headerParagraph = header.addParagraph();//插入图片到页眉的段落DocPicture headerPicture = headerParagraph.appendPicture(baseUrl + "img/custom/word.png");headerPicture.setHorizontalAlignment(ShapeHorizontalAlignment.Left);headerPicture.setVerticalOrigin(VerticalOrigin.Top_Margin_Area);headerPicture.setVerticalAlignment(ShapeVerticalAlignment.Bottom);//添加文字到页眉的段落TextRange text = headerParagraph.appendText("上海烟草集团");//text.getCharacterFormat().setFontName("default");text.getCharacterFormat().setFontSize(10);text.getCharacterFormat().setItalic(true);headerParagraph.getFormat().setHorizontalAlignment(HorizontalAlignment.Right);//设置文字环绕方式headerPicture.setTextWrappingStyle(TextWrappingStyle.Behind);//设置页眉段落的底部边线样式headerParagraph.getFormat().getBorders().getBottom().setBorderType(BorderStyle.Single);headerParagraph.getFormat().getBorders().getBottom().setLineWidth(1f);//添加段落到页脚Paragraph footerParagraph = footer.addParagraph();//添加Field_Page和Field_Num_Pages域到页脚段落，用于显示当前页码和总页数footerParagraph.appendField("page number", FieldType.Field_Page);footerParagraph.appendText(" of ");footerParagraph.appendField("number of pages", FieldType.Field_Num_Pages);footerParagraph.getFormat().setHorizontalAlignment(HorizontalAlignment.Right);//设置页脚段落的顶部边线样式footerParagraph.getFormat().getBorders().getTop().setBorderType(BorderStyle.Single);footerParagraph.getFormat().getBorders().getTop().setLineWidth(1f);}

highcharts转svg插入html模板

/**将highcharts生成<img>的src标签插入html中方便pdf转化**/function img() {$(".highchartImg").css("display","none");$.each($(".highchartImg"),function (idx,c) {var charData = $(this).highcharts().getSVG();var canvas = document.getElementById("canvas");canvg(canvas,charData);var image = new Image();image.src = canvas.toDataURL("image/png");$(this).next().attr("src",image.src);})}

设置highcharts原始和导出图片大小

chart: {zoomType: 'xy', //图片可拉伸type: 'column',width: 1200}，exporting: {sourceWidth: 650,sourceHeight: 360,enabled: false, //屏蔽下载图片选择框}

总结：html的元素标签一定要注意规范性，否则样式会有偏差，不过慢慢调试也能弄出来，比如一些表头的格式偏差，rowspan=“0” 这种就不要显示在标签里了，默认就好。

好吧，先到这里。这只是个人一个阶段性的总结和记录，感谢很多提供思路的朋友，最终解决了问题。如果能帮到有同样需求的伙伴就更好了。