find files by extension, *.html under a folder in nodejs

Question

I\'d like to find all *.html files in src folder and all its sub folders using nodejs. What is the best way to do it?

var folder = \'/project1/src\';
var ex         


        

Answer 1

You can use OS help for this. Here is a cross-platform solution:

1. The bellow function uses ls and dir and does not search recursively but it has relative paths

var exec = require('child_process').exec;
function findFiles(folder,extension,cb){
    var command = "";
    if(/^win/.test(process.platform)){
        command = "dir /B "+folder+"\\*."+extension;
    }else{
        command = "ls -1 "+folder+"/*."+extension;
    }
    exec(command,function(err,stdout,stderr){
        if(err)
            return cb(err,null);
        //get rid of \r from windows
        stdout = stdout.replace(/\r/g,"");
        var files = stdout.split("\n");
        //remove last entry because it is empty
        files.splice(-1,1);
        cb(err,files);
    });
}

findFiles("folderName","html",function(err,files){
    console.log("files:",files);
})

2. The bellow function uses find and dir, searches recursively but on windows it has absolute paths

var exec = require('child_process').exec;
function findFiles(folder,extension,cb){
    var command = "";
    if(/^win/.test(process.platform)){
        command = "dir /B /s "+folder+"\\*."+extension;
    }else{
        command = 'find '+folder+' -name "*.'+extension+'"'
    }
    exec(command,function(err,stdout,stderr){
        if(err)
            return cb(err,null);
        //get rid of \r from windows
        stdout = stdout.replace(/\r/g,"");
        var files = stdout.split("\n");
        //remove last entry because it is empty
        files.splice(-1,1);
        cb(err,files);
    });
}

findFiles("folder","html",function(err,files){
    console.log("files:",files);
})

Answer 2

my two pence, using map in place of for-loop

var path = require('path'), fs = require('fs');

var findFiles = function(folder, pattern = /.*/, callback) {
  var flist = [];

  fs.readdirSync(folder).map(function(e){ 
    var fname = path.join(folder, e);
    var fstat = fs.lstatSync(fname);
    if (fstat.isDirectory()) {
      // don't want to produce a new array with concat
      Array.prototype.push.apply(flist, findFiles(fname, pattern, callback)); 
    } else {
      if (pattern.test(fname)) {
        flist.push(fname);
        if (callback) {
          callback(fname);
        }
      }
    }
  });
  return flist;
};

// HTML files   
var html_files = findFiles(myPath, /\.html$/, function(o) { console.log('look what we have found : ' + o} );

// All files
var all_files = findFiles(myPath);

Answer 3

Take a look into file-regex

let findFiles = require('file-regex')
let pattern = '\.js'

findFiles(__dirname, pattern, (err, files) => {  
   console.log(files);
})

This above snippet would print all the js files in the current directory.

Answer 4

The following code does a recursive search inside ./ (change it appropriately) and returns an array of absolute file names ending with .html

var fs = require('fs');
var path = require('path');

var searchRecursive = function(dir, pattern) {
  // This is where we store pattern matches of all files inside the directory
  var results = [];

  // Read contents of directory
  fs.readdirSync(dir).forEach(function (dirInner) {
    // Obtain absolute path
    dirInner = path.resolve(dir, dirInner);

    // Get stats to determine if path is a directory or a file
    var stat = fs.statSync(dirInner);

    // If path is a directory, scan it and combine results
    if (stat.isDirectory()) {
      results = results.concat(searchRecursive(dirInner, pattern));
    }

    // If path is a file and ends with pattern then push it onto results
    if (stat.isFile() && dirInner.endsWith(pattern)) {
      results.push(dirInner);
    }
  });

  return results;
};

var files = searchRecursive('./', '.html'); // replace dir and pattern
                                                // as you seem fit

console.log(files);

Answer 5

I just noticed, you are using sync fs methods, that might block you application, here is a promise-based async way using async and q, you can execute it with START=/myfolder FILTER=".jpg" node myfile.js, assuming you put the following code in a file called myfile.js:

Q = require("q")
async = require("async")
path = require("path")
fs = require("fs")

function findFiles(startPath, filter, files){
    var deferred;
    deferred = Q.defer(); //main deferred

    //read directory
    Q.nfcall(fs.readdir, startPath).then(function(list) {
        var ideferred = Q.defer(); //inner deferred for resolve of async each
        //async crawling through dir
        async.each(list, function(item, done) {

            //stat current item in dirlist
            return Q.nfcall(fs.stat, path.join(startPath, item))
                .then(function(stat) {
                    //check if item is a directory
                    if (stat.isDirectory()) {
                        //recursive!! find files in subdirectory
                        return findFiles(path.join(startPath, item), filter, files)
                            .catch(function(error){
                                console.log("could not read path: " + error.toString());
                            })
                            .finally(function() {
                                //resolve async job after promise of subprocess of finding files has been resolved
                                return done();
                             });
                    //check if item is a file, that matches the filter and add it to files array
                    } else if (item.indexOf(filter) >= 0) {
                        files.push(path.join(startPath, item));
                        return done();
                    //file is no directory and does not match the filefilter -> don't do anything
                    } else {
                        return done();
                    }
                })
                .catch(function(error){
                    ideferred.reject("Could not stat: " + error.toString());
                });
        }, function() {
            return ideferred.resolve(); //async each has finished, so resolve inner deferred
        });
        return ideferred.promise;
    }).then(function() {
        //here you could do anything with the files of this recursion step (otherwise you would only need ONE deferred)
        return deferred.resolve(files); //resolve main deferred
    }).catch(function(error) {
        deferred.reject("Could not read dir: " + error.toString());
        return
    });
    return deferred.promise;
}


findFiles(process.env.START, process.env.FILTER, [])
    .then(function(files){
        console.log(files);
    })
    .catch(function(error){
        console.log("Problem finding files: " + error);
})

Answer 6

Old post but ES6 now handles this out of the box with the includes method.

let files = ['file.json', 'other.js'];

let jsonFiles = files.filter(file => file.includes('.json'));

console.log("Files: ", jsonFiles) ==> //file.json