使用Office JS Api从Word文档的文本框中提取文本

问题描述

我在Word文档中有一些文本框（“形状”>“文本框”）。该文档是一个CV模板，其中包括很多模板。我想选择文档的所有文本框，提取文本，删除文本框并注入提取的文本。我尝试过

  const test = myArray.reduce((accumulator,value) => {
    return []
      .concat(
        accumulator,value,value.sub && value.sub?.reduce((acc,val) => [].concat(acc,val))
      ).filter((val) => val !== null);
  }); // Output only the first and 2 level

，然后同步上下文，以便获取文本。

解决方法

我终于采取了以下解决方法。它既快速又可以在Windows和macOS上很好地工作

1. 获取文档正文的OOXML

2. 解析OOXL.value并生成xmlDocument（xmlDoc）

3. 检测包含文本的现有文本框和形状：getElementsByTagName（“ wps：wsp”）

4. 从（3）中提取文本

5. 生成一个提取了文本的简单xml TextElement

6. 用（5）代替（3）

7. 将更新后的xmlDoc序列化为xmlString并获取更新后的OOXML.value

8. 插入更新的OOXML.value，以文档替换现有的OOXML.value

   Word.run（函数（上下文）{

        //Select document body and extract OOXML 
        const body = context.document.body;
        const ooxml = body.getOoxml();
   
        return context.sync().then(function () {
   
            //Initialize DOM Parser
            const parser = new DOMParser();
            const xmlDoc = parser.parseFromString(ooxml.value,"text/xml");
   
            //Get all runs
            const rows = xmlDoc.getElementsByTagName("w:r");
            for (let j = 0; j < rows.length; j++) {
                const row = rows[j];
                const rowHasTextBox = row.getElementsByTagName("wps:txbx").length > 0;
                //If no textbox,shape,wordart exists skip current run
                if (!rowHasTextBox) continue;
   
                //Select textbox,wordart and get paragraphs
                const textboxContainer = row.getElementsByTagName("wps:txbx")[0];
                const paragraphs = textboxContainer.getElementsByTagName("w:p");
   
                // Create a new run which will replace the existing run
                const newRow = xmlDoc.createElement("w:r");
                const breakLine = xmlDoc.createElement("w:br");
                //Append breakline and "{{"
                newRow.appendChild(breakLine);
                newRow.appendChild(startRow);
   
                for (let p = 0; p < paragraphs.length; p++) {
                    //Check whether paragrapj has text
                    const paragraphHasText = paragraphs[p].getElementsByTagName("w:t").length > 0;
                    if (!paragraphHasText) continue;
                    //Extract text
                    let textExtracted = "";
                    const textBoxTexts = paragraphs[p].getElementsByTagName("w:t");
                    for (let k = 0; k < textBoxTexts.length; k++) {
                        const textBoxText = textBoxTexts[k].innerHTML;
                        textExtracted = textExtracted + textBoxText;
                        textExtracted = textExtracted + " ";
                    }
                     // Create a temp run which will hold the etxtracted text
                    const tempRow = xmlDoc.createElement("w:r");
                    const newText = xmlDoc.createElement('w:t');
                    newText.setAttribute("xml:space","preserve");
                    newText.innerHTML = textExtracted;
                    textExtracted = "";
                    tempRow.appendChild(newText);
                    newRow.appendChild(tempRow);
                    const breakLine = xmlDoc.createElement("w:br");
                    newRow.appendChild(breakLine);
                }
   
   
                //Replace existing run with the new one
                row.replaceWith(newRow);
            }
            //Serialize dom,clear body and replace OOXML
            const serializedXML = new XMLSerializer().serializeToString(xmlDoc.documentElement);
            body.clear();
            return context.sync().then(function () {
                body.insertOoxml(serializedXML,Word.InsertLocation.replace);
                console.log('done');
            });
        });
    })
    .catch(error => {
        console.log('Error: ',error);
        resolve(false);
    });