function getFilesRecursiveSync(dir, fileList, optionalFilterFunction) {
    if (!fileList) {
        grunt.log.error("Variable 'fileList' is undefined or NULL.");
        return;
    }
    var files = fs.readdirSync(dir);
    for (var i in files) {
        if (!files.hasOwnProperty(i)) continue;
        var name = dir + '/' + files[i];
        if (fs.statSync(name).isDirectory()) {
            getFilesRecursiveSync(name, fileList, optionalFilterFunction);
        } else {
            if (optionalFilterFunction && optionalFilterFunction(name) !== true)
                continue;
            fileList.push(name);
        }
    }
}

IMO完成此类任务最方便的方法是使用glob工具。这是node.js的glob包

npm install glob

然后使用通配符匹配文件名（示例取自软件包的网站）

var glob = require("glob")

// options is optional
glob("**/*.js", options, function (er, files) {
  // files is an array of filenames.
  // If the `nonull` option is set, and nothing
  // was found, then files is ["**/*.js"]
  // er is an error object or null.
})

如果您计划使用globby，这里有一个示例来查找当前文件夹下的任何xml文件

var globby = require('globby');

const paths = await globby("**/*.xml");  

加载fs:

const fs = require('fs');

异步读取文件：

fs.readdir('./dir', function (err, files) {
    // "files" is an Array with files names
});

读取文件同步：

var files = fs.readdirSync('./dir');

这是一个TypeScript，可选递归，可选错误日志和异步解决方案。可以为要查找的文件名指定正则表达式。

我使用了fs extra，因为这是对fs的一个简单的超集改进。

import * as FsExtra from 'fs-extra'

/**
 * Finds files in the folder that match filePattern, optionally passing back errors .
 * If folderDepth isn't specified, only the first level is searched. Otherwise anything up
 * to Infinity is supported.
 *
 * @static
 * @param {string} folder The folder to start in.
 * @param {string} [filePattern='.*'] A regular expression of the files you want to find.
 * @param {(Error[] | undefined)} [errors=undefined]
 * @param {number} [folderDepth=0]
 * @returns {Promise<string[]>}
 * @memberof FileHelper
 */
public static async findFiles(
    folder: string,
    filePattern: string = '.*',
    errors: Error[] | undefined = undefined,
    folderDepth: number = 0
): Promise<string[]> {
    const results: string[] = []

    // Get all files from the folder
    let items = await FsExtra.readdir(folder).catch(error => {
        if (errors) {
            errors.push(error) // Save errors if we wish (e.g. folder perms issues)
        }

        return results
    })

    // Go through to the required depth and no further
    folderDepth = folderDepth - 1

    // Loop through the results, possibly recurse
    for (const item of items) {
        try {
            const fullPath = Path.join(folder, item)

            if (
                FsExtra.statSync(fullPath).isDirectory() &&
                folderDepth > -1)
            ) {
                // Its a folder, recursively get the child folders' files
                results.push(
                    ...(await FileHelper.findFiles(fullPath, filePattern, errors, folderDepth))
                )
            } else {
                // Filter by the file name pattern, if there is one
                if (filePattern === '.*' || item.search(new RegExp(filePattern, 'i')) > -1) {
                    results.push(fullPath)
                }
            }
        } catch (error) {
            if (errors) {
                errors.push(error) // Save errors if we wish
            }
        }
    }

    return results
}

从Node v10.10.0开始，可以将fs.readdir和fs.readderSync的新withFileTypes选项与dirent.isDirectory（）函数结合使用，以过滤目录中的文件名。看起来像这样：

fs.readdirSync('./dirpath', {withFileTypes: true})
.filter(item => !item.isDirectory())
.map(item => item.name)

返回的数组的格式为：

['file1.txt', 'file2.txt', 'file3.txt']

如果有人还在搜索这个，我会这样做：

从“fs”导入fs；从“path”导入路径；const getAllFiles=目录=>fs.readdirSync（dir）.reduce（（files，file）=>{常量名称=路径.连接（目录，文件）；const isDirectory=fs.statSync（名称）.isDirectory（）；return isDirectory？[…file，…getAllFiles（名称）]：[…files，名称]；}, []);

它的工作对我很好