Unexpected empty writestream in collectionFS using graphicsmagick - javascript

I'm using CollectionFS for managing images. Furthermore I'm using graphicsmagick gm() for manipulating images.
Now I want to crop a already saved image. Therefore on a click event a server-method is called, which does the crop(). But after doing this, in the collection I find an empty image with size=0 updated on the correct date.
I don't see, what I am doing wrong.
shared.js
Images = new FS.Collection("images", {
stores: [
new FS.Store.FileSystem("thumbnail", {
transformWrite: function(fileObj, readStream, writeStream) {
gm(readStream, fileObj.name()).autoOrient().resize('96', '96' + '^').gravity('Center').extent('96', '96').stream().pipe(writeStream);
}
}),
new FS.Store.FileSystem("public"),
]
});
server.js
Meteor.methods({
'crop': function (fileId, selection) {
var file = Images.findOne({ _id: fileId }),
read = file.createReadStream('public'),
write = file.createWriteStream('public');
gm(read)
.crop(selection.width, selection.height, selection.left, selection.top)
.stream()
.pipe(write);
}
});
client.js
Template.editor.events({
'click #crop': function () {
var fileId = '123456789',
selection = { height: 100, width: 100, top: 10, left: 10 };
Meteor.call('crop', fileId, selection);
}
});
Update
As recommended by Christian I'm using a tmp-file for the writeStream, because the writeStream can't be the same like the readStream - which caused the empty result.
But after writing to the tmp-file, the content of it has to be copied back to the public store. How do I do that?
Meteor.methods({
'crop': function (fileId, selection) {
var fs = Meteor.npmRequire('fs'),
file = Images.findOne({ _id: fileId }),
read = file.createReadStream('public'),
filename = '/tmp/gm_' + Date.now(),
tmp = fs.createWriteStream(filename);
gm(read)
.crop(selection.width, selection.height, selection.left, selection.top)
.stream()
.pipe(tmp);
// After writing to tmp -> copy back to stream and delete tmp-file
}
});
Update 2
I tried this one:
// Add temp store
new FS.Store.FileSystem("temp")
// Method
Meteor.methods({
'crop': function (fileId, selection) {
var file = Images.findOne({ _id: fileId }),
read = file.createReadStream('public'),
temp = file.createWriteStream('temp');
gm(read)
.crop(selection.width, selection.height, selection.left, selection.top)
.stream()
.pipe(tmp)
.on('end', function () {
var tmpread = file.createReadStream('temp'),
write = file.createWriteStream('public');
gm(tmpread).stream().pipe(write);
});
}
});

You can't read and write into the same file. This is equivalent to things like
cat test | grep 1 > test
on the shell. You can try it and see that test will be empty afterwards.
You need to create an intermediate, temporary file in your crop method.
Assuming that is indeed the problem, then this is one way of doing this (not tested):
var fs = Meteor.npmRequire('fs');
var file = Images.findOne({ _id: fileId }),
var read = file.createReadStream('public'),
var filename = '/tmp/gm_' + Date.now();
var tmp = fs.createWriteStream(filename);
var gmread = gm(read)
.crop(selection.width, selection.height, selection.left, selection.top)
.stream();
gmread.on('end', function() {
// done streaming through GM, copy the result back:
var tmpread = fs.createReadStream(filename);
var write = file.createWriteStream('public');
tmpread.pipe(write);
});
gmread.pipe(tmp);

Related

how to save a PDF or image file into a collection with meteor?

how to save a PDF or image file into a collection with meteor
I tried but it just saves the pdf link or picture.
I tried the code below but it fits rather in the Files collection binary digits.
please I want to do is insert the file into the collection and not the link to the file.
'change input' : function(event,template){
var file = event.target.files;
if (!file) return;
var reader = new FileReader();
reader.onload = function(event){
var buffer = new Uint8Array(reader.result)
Meteor.call('saveFile', buffer);
}
reader.readAsArrayBuffer(file);
}
/*** server.js ***/
Files = new Mongo.Collection('files');
Meteor.methods({
'saveFile': function(buffer){
Files.insert({data:buffer})
}
});
You can use CollectionFS with dropbox or Amazon S3 store like this:
Collections file:
var dropboxStore = new FS.Store.Dropbox("files", {
key: //your key here,
secret: //Your secret here,
token: // Access tokenhere. Don’t share your access token with anyone.
folder: FolderName, //optional, which folder (key prefix) to use
// The rest are generic store options supported by all storage adapters
// transformWrite: myTransformWriteFunction, //optional
// transformRead: myTransformReadFunction, //optional
// maxTries: 1 //optional, default 5
});
Images = new FS.Collection("images", {
// stores: [new FS.Store.FileSystem("images", {path:"../../../../../.uploads"})]
stores: [dropboxStore]
});
Images.allow({
insert: function () {
return true;
},
update: function () {
return true;
},
download: function () {
return true;
}
});
Client side: On file change event
FS.Utility.eachFile(event, function (file) {
var imgfile = event.target.files[0];
var img = new Image();
img.src = window.URL.createObjectURL(imgfile);
img.onload = function () {
Images.insert(file, function (err, fileObj) {
free_spinz_symbol.set(fileObj);
});
};
});
It might not be the best solution for you if you're up-to-date with the meteor version, because the project is currently being deprecated but CollectionFS used to be a great solution for file storage. Nowdays, Meteor-Files seems to be a better choice to handle PDF/Images storage

MongoDB : Read a file stored using GridFS from MongoDB in Javascript(using Node.JS)

In a mongodb from javascript code I am able to insert a file(of any extension) using gridfs, It's stored as fs.chunks & fs.files. Now i want to read this file's stream, but on any of the read opration calls on gridStore its giving error : "this.data.buffer.slice is not a function". I have used mongojs library from Node.JS.
// Write a file into MongoDB - Working fine.
// Our file ID
var fileId = new ObjectID();
// Open a new file
var gridStore;
// Create a file and open it
gridStore = new GridStore(db, fileId, "D:/Files/Sofa 2 N050210.3DS", "w+");
var data = fs.readFileSync('D:/Files/Sofa 2 N050210.3DS');
gridStore.open(function (err, gridStore) {
gridStore.write(data, function (err, gridStore) {
// Flush the file to GridFS
gridStore.close(function (err, fileData) {
// Write successfull
});
});
});
Now problem is when reading, After Closing gridStore, I tried to open file for reading & it is giving error at read call i.e on line no-4 in below code.
gridStore.close(function (result) {
var gridStore = new GridStore(db, "test_gs_seek_with_buffer", "r");
gridStore.open(function (err, gridStore) {
gridStore.read(5, function (err, data) {
if(err){
console.log("Error"); return;
}
console.log(data.toString());
});
});
});
Please help me find solution or the way to read back the file stored in GridFS (From the javascript code not from the command prompt).
This code is working fine for me. Try with this. Hope it works.
var mongoose = require('mongoose');
var Grid = require('gridfs-stream');
mongoose.connection.once('open', function () {
gfs = new Grid(mongoose.connection.db, mongoose.mongo);//If you are using mongoose
});
var db = new mongo.Db('yourDatabaseName', new mongo.Server("127.0.0.1", 27017)); //if you are using mongoDb directily
var gfs = Grid(db, mongo);
var rstream = gfs.createReadStream(filename);
var bufs = [];
rstream.on('data', function (chunk) {
bufs.push(chunk);
}).on('error', function () {
res.send();
})
.on('end', function () { // done
var fbuf = Buffer.concat(bufs);
var File = (fbuf.toString('base64'));
res.send(File);
});

Nodejs non-blocking and Event Emitter. Refresh URL

fist of all im not shure if the following is a non-blocking problem?
im getting started with https://github.com/sahat/hackathon-starter
currently i try to read all files out of a folder and later process all files...
i used EventEmitter to kind of manage the workflow.
i want to clear all arrays if the URL is refeshed or loaded new, but somehow if i reaload the URL there seems to be something inside the arrays which cases multiple outputs of the same data?
at the moment i just would be happy to have a correct console.log output.
/**
* GET /
* Home page.
*/
var fs = require('fs');
//XML
var jsxml = require("node-jsxml");
var Namespace = jsxml.Namespace,
QName = jsxml.QName,
XML = jsxml.XML,
XMLList = jsxml.XMLList;
//EventEmitter
var EventEmitter=require('events').EventEmitter;
var dateinamenEE=new EventEmitter();
var dateiinhaltEE=new EventEmitter();
var dateinamen = [];
var dateiinhalt = [];
exports.index = function(req, res) {
fs.readdir('./data', function (err, files) {
if (!err) {
files.forEach(function(value) {
dateinamen.push(value);
});
dateinamenEE.emit('dateinamen_ready');
} else {
throw err;
}
});
dateinamenEE.on('dateinamen_ready',function(){
dateinamen.forEach(function(value) {
var buf = fs.readFileSync('./data/'+value, "utf8");
var xml = new XML(buf);
var list = xml.descendants("suggestion");
var ergebnis = "";
var basiswort = "";
var buchstabe = "";
var obj = null;
list.each(function(item, index){
ergebnis = item.attribute('data').toString()
//basiswort = value.replace("%2B", " ");
//basiswort = basiswort.replace(".xml", "");
//var pieces = buchstabe.split(" ");
obj = {k: basiswort, b: buchstabe, e: ergebnis};
dateiinhalt.push(obj);
});
});
dateiinhaltEE.emit('dateiinhalt_ready');
});
dateiinhaltEE.on('dateiinhalt_ready',function(){
//console.log(dateiinhalt);
console.log("dateinamen:" + dateinamen.length);
console.log("dateiinhalt:" + dateiinhalt.length);
});
res.render('home', {
title: 'Home'
});
};
If if log the length of the 2 arrays the output on the second reload shows. First time loading the url:
Express server listening on port 3000 in development mode
dateinamen:2
dateiinhalt:20
Second time / refreshing the url:
GET / 200 898.198 ms - -
GET /fonts/fontawesome-webfont.woff2?v=4.3.0 304 12.991 ms - -
GET /favicon.ico 200 4.516 ms - -
dateinamen:4
dateiinhalt:60
dateinamen:4
dateiinhalt:60
dateinamen:4
dateiinhalt:100
dateinamen:4
dateiinhalt:100
GET / 200 139.259 ms - -
What causes the code to extend the arrays while reloading the page?
The non-blocking problem is due do your for(...) loops.
Changing them by : array.forEach(function(elem, index){});
EDIT
The arrays should be initialized inside the index function :
exports.index = function(req, res) {
var dateinamen = [];
var dateiinhalt = [];
...
Also, I'm not sure you need the use of EventEmitter.
Something like
`
fs.readdir('./data', function (err, files) {
if (!err) {
files.forEach(function(file){
var buf = fs.readFileSync('./data/'+file, "utf8");
var xml = new XML(buf);
var list = xml.descendants("suggestion");
var ergebnis = null;
var obj = null;
list.each(function(item, index){
ergebnis = item.attribute('data').toString();
obj = {k: file, v: ergebnis};
dateiinhalt.push(obj);
});
});
console.log(dateiinhalt);
} else {
throw err;
}
});
`
could do the job no?
(I wanted to say this as a comment, but I'm still missing reputation)

Node.js node-csv module - working with a local CSV file

I'm trying to use the new version of the node-csv node module to do some CSV manipulation.
node-csv
I've used perl in the past, but would like to try JavaScript this time. I'm having trouble figuring out how to import a local CSV file instead of using the built in generator. The documentation for node-csv doesn't show how to do this as far as I can tell(although it does provide an example for the previous version).
Here is the example code, which works as expected.
var csv = require('csv');
var generator = csv.generate({seed: 1, columns: 2, length: 20});
var parser = csv.parse();
var transformer = csv.transform(function(data){
return data.map(function(value){return value.toUpperCase()});
});
var stringifier = csv.stringify();
generator.on('readable', function(){
while(data = generator.read()){
parser.write(data);
}
});
parser.on('readable', function(){
while(data = parser.read()){
transformer.write(data);
}
});
transformer.on('readable', function(){
while(data = transformer.read()){
stringifier.write(data);
}
});
stringifier.on('readable', function(){
while(data = stringifier.read()){
process.stdout.write(data);
}
});
I plan on using the FS module, but am not sure how to pass the local file into the node-csv functions.
var fs = require('fs');
Here is an example for the PREVIOUS version, which uses completely different syntax:
// node samples/sample.js
var csv = require('csv');
var fs = require('fs');
csv()
.from.stream(fs.createReadStream(__dirname+'/sample.in'))
.to.path(__dirname+'/sample.out')
.transform( function(row){
row.unshift(row.pop());
return row;
})
.on('record', function(row,index){
console.log('#'+index+' '+JSON.stringify(row));
})
.on('end', function(count){
console.log('Number of lines: '+count);
})
.on('error', function(error){
console.log(error.message);
});
Any suggestions?

How to read data From *.CSV file using JavaScript?

My CSV data looks like this:
heading1,heading2,heading3,heading4,heading5
value1_1,value2_1,value3_1,value4_1,value5_1
value1_2,value2_2,value3_2,value4_2,value5_2
...
How do you read this data and convert to an array like this using JavaScript?:
[
heading1: value1_1,
heading2: value2_1,
heading3: value3_1,
heading4: value4_1
heading5: value5_1
],[
heading1: value1_2,
heading2: value2_2,
heading3: value3_2,
heading4: value4_2,
heading5: value5_2
]
....
I've tried this code but no luck!:
<script type="text/javascript">
var allText =[];
var allTextLines = [];
var Lines = [];
var txtFile = new XMLHttpRequest();
txtFile.open("GET", "file://d:/data.txt", true);
txtFile.onreadystatechange = function()
{
allText = txtFile.responseText;
allTextLines = allText.split(/\r\n|\n/);
};
document.write(allTextLines);
document.write(allText);
document.write(txtFile);
</script>
No need to write your own...
The jQuery-CSV library has a function called $.csv.toObjects(csv) that does the mapping automatically.
Note: The library is designed to handle any CSV data that is RFC 4180 compliant, including all of the nasty edge cases that most 'simple' solutions overlook.
Like #Blazemonger already stated, first you need to add line breaks to make the data valid CSV.
Using the following dataset:
heading1,heading2,heading3,heading4,heading5
value1_1,value2_1,value3_1,value4_1,value5_1
value1_2,value2_2,value3_2,value4_2,value5_2
Use the code:
var data = $.csv.toObjects(csv):
The output saved in 'data' will be:
[
{ heading1:"value1_1",heading2:"value2_1",heading3:"value3_1",heading4:"value4_1",heading5:"value5_1" }
{ heading1:"value1_2",heading2:"value2_2",heading3:"value3_2",heading4:"value4_2",heading5:"value5_2" }
]
Note: Technically, the way you wrote the key-value mapping is invalid JavaScript. The objects containing the key-value pairs should be wrapped in brackets.
If you want to try it out for yourself, I suggest you take a look at the Basic Usage Demonstration under the 'toObjects()' tab.
Disclaimer: I'm the original author of jQuery-CSV.
Update:
Edited to use the dataset that the op provided and included a link to the demo where the data can be tested for validity.
Update2:
Due to the shuttering of Google Code. jquery-csv has moved to GitHub
NOTE: I concocted this solution before I was reminded about all the "special cases" that can occur in a valid CSV file, like escaped quotes. I'm leaving my answer for those who want something quick and dirty, but I recommend Evan's answer for accuracy.
This code will work when your data.txt file is one long string of comma-separated entries, with no newlines:
data.txt:
heading1,heading2,heading3,heading4,heading5,value1_1,...,value5_2
javascript:
$(document).ready(function() {
$.ajax({
type: "GET",
url: "data.txt",
dataType: "text",
success: function(data) {processData(data);}
});
});
function processData(allText) {
var record_num = 5; // or however many elements there are in each row
var allTextLines = allText.split(/\r\n|\n/);
var entries = allTextLines[0].split(',');
var lines = [];
var headings = entries.splice(0,record_num);
while (entries.length>0) {
var tarr = [];
for (var j=0; j<record_num; j++) {
tarr.push(headings[j]+":"+entries.shift());
}
lines.push(tarr);
}
// alert(lines);
}
The following code will work on a "true" CSV file with linebreaks between each set of records:
data.txt:
heading1,heading2,heading3,heading4,heading5
value1_1,value2_1,value3_1,value4_1,value5_1
value1_2,value2_2,value3_2,value4_2,value5_2
javascript:
$(document).ready(function() {
$.ajax({
type: "GET",
url: "data.txt",
dataType: "text",
success: function(data) {processData(data);}
});
});
function processData(allText) {
var allTextLines = allText.split(/\r\n|\n/);
var headers = allTextLines[0].split(',');
var lines = [];
for (var i=1; i<allTextLines.length; i++) {
var data = allTextLines[i].split(',');
if (data.length == headers.length) {
var tarr = [];
for (var j=0; j<headers.length; j++) {
tarr.push(headers[j]+":"+data[j]);
}
lines.push(tarr);
}
}
// alert(lines);
}
http://jsfiddle.net/mblase75/dcqxr/
Don't split on commas -- it won't work for most CSV files, and this question has wayyyy too many views for the asker's kind of input data to apply to everyone. Parsing CSV is kind of scary since there's no truly official standard, and lots of delimited text writers don't consider edge cases.
This question is old, but I believe there's a better solution now that Papa Parse is available. It's a library I wrote, with help from contributors, that parses CSV text or files. It's the only JS library I know of that supports files gigabytes in size. It also handles malformed input gracefully.
1 GB file parsed in 1 minute:
(Update: With Papa Parse 4, the same file took only about 30 seconds in Firefox. Papa Parse 4 is now the fastest known CSV parser for the browser.)
Parsing text is very easy:
var data = Papa.parse(csvString);
Parsing files is also easy:
Papa.parse(file, {
complete: function(results) {
console.log(results);
}
});
Streaming files is similar (here's an example that streams a remote file):
Papa.parse("http://example.com/bigfoo.csv", {
download: true,
step: function(row) {
console.log("Row:", row.data);
},
complete: function() {
console.log("All done!");
}
});
If your web page locks up during parsing, Papa can use web workers to keep your web site reactive.
Papa can auto-detect delimiters and match values up with header columns, if a header row is present. It can also turn numeric values into actual number types. It appropriately parses line breaks and quotes and other weird situations, and even handles malformed input as robustly as possible. I've drawn on inspiration from existing libraries to make Papa, so props to other JS implementations.
I am using d3.js for parsing csv file. Very easy to use.
Here is the docs.
Steps:
npm install d3-request
Using Es6;
import { csv } from 'd3-request';
import url from 'path/to/data.csv';
csv(url, function(err, data) {
console.log(data);
})
Please see docs for more.
Update -
d3-request is deprecated. you can use d3-fetch
Here's a JavaScript function that parses CSV data, accounting for commas found inside quotes.
// Parse a CSV row, accounting for commas inside quotes
function parse(row){
var insideQuote = false,
entries = [],
entry = [];
row.split('').forEach(function (character) {
if(character === '"') {
insideQuote = !insideQuote;
} else {
if(character == "," && !insideQuote) {
entries.push(entry.join(''));
entry = [];
} else {
entry.push(character);
}
}
});
entries.push(entry.join(''));
return entries;
}
Example use of the function to parse a CSV file that looks like this:
"foo, the column",bar
2,3
"4, the value",5
into arrays:
// csv could contain the content read from a csv file
var csv = '"foo, the column",bar\n2,3\n"4, the value",5',
// Split the input into lines
lines = csv.split('\n'),
// Extract column names from the first line
columnNamesLine = lines[0],
columnNames = parse(columnNamesLine),
// Extract data from subsequent lines
dataLines = lines.slice(1),
data = dataLines.map(parse);
// Prints ["foo, the column","bar"]
console.log(JSON.stringify(columnNames));
// Prints [["2","3"],["4, the value","5"]]
console.log(JSON.stringify(data));
Here's how you can transform the data into objects, like D3's csv parser (which is a solid third party solution):
var dataObjects = data.map(function (arr) {
var dataObject = {};
columnNames.forEach(function(columnName, i){
dataObject[columnName] = arr[i];
});
return dataObject;
});
// Prints [{"foo":"2","bar":"3"},{"foo":"4","bar":"5"}]
console.log(JSON.stringify(dataObjects));
Here's a working fiddle of this code.
Enjoy! --Curran
You can use PapaParse to help.
https://www.papaparse.com/
Here is a CodePen.
https://codepen.io/sandro-wiggers/pen/VxrxNJ
Papa.parse(e, {
header:true,
before: function(file, inputElem){ console.log('Attempting to Parse...')},
error: function(err, file, inputElem, reason){ console.log(err); },
complete: function(results, file){ $.PAYLOAD = results; }
});
If you want to solve this without using Ajax, use the FileReader() Web API.
Example implementation:
Select .csv file
See output
function readSingleFile(e) {
var file = e.target.files[0];
if (!file) {
return;
}
var reader = new FileReader();
reader.onload = function(e) {
var contents = e.target.result;
displayContents(contents);
displayParsed(contents);
};
reader.readAsText(file);
}
function displayContents(contents) {
var element = document.getElementById('file-content');
element.textContent = contents;
}
function displayParsed(contents) {
const element = document.getElementById('file-parsed');
const json = contents.split(',');
element.textContent = JSON.stringify(json);
}
document.getElementById('file-input').addEventListener('change', readSingleFile, false);
<input type="file" id="file-input" />
<h3>Raw contents of the file:</h3>
<pre id="file-content">No data yet.</pre>
<h3>Parsed file contents:</h3>
<pre id="file-parsed">No data yet.</pre>
function CSVParse(csvFile)
{
this.rows = [];
var fieldRegEx = new RegExp('(?:\s*"((?:""|[^"])*)"\s*|\s*((?:""|[^",\r\n])*(?:""|[^"\s,\r\n]))?\s*)(,|[\r\n]+|$)', "g");
var row = [];
var currMatch = null;
while (currMatch = fieldRegEx.exec(this.csvFile))
{
row.push([currMatch[1], currMatch[2]].join('')); // concatenate with potential nulls
if (currMatch[3] != ',')
{
this.rows.push(row);
row = [];
}
if (currMatch[3].length == 0)
break;
}
}
I like to have the regex do as much as possible. This regex treats all items as either quoted or unquoted, followed by either a column delimiter, or a row delimiter. Or the end of text.
Which is why that last condition -- without it it would be an infinite loop since the pattern can match a zero length field (totally valid in csv). But since $ is a zero length assertion, it won't progress to a non match and end the loop.
And FYI, I had to make the second alternative exclude quotes surrounding the value; seems like it was executing before the first alternative on my javascript engine and considering the quotes as part of the unquoted value. I won't ask -- just got it to work.
Per the accepted answer,
I got this to work by changing the 1 to a 0 here:
for (var i=1; i<allTextLines.length; i++) {
changed to
for (var i=0; i<allTextLines.length; i++) {
It will compute the a file with one continuous line as having an allTextLines.length of 1. So if the loop starts at 1 and runs as long as it's less than 1, it never runs. Hence the blank alert box.
$(function() {
$("#upload").bind("click", function() {
var regex = /^([a-zA-Z0-9\s_\\.\-:])+(.csv|.xlsx)$/;
if (regex.test($("#fileUpload").val().toLowerCase())) {
if (typeof(FileReader) != "undefined") {
var reader = new FileReader();
reader.onload = function(e) {
var customers = new Array();
var rows = e.target.result.split("\r\n");
for (var i = 0; i < rows.length - 1; i++) {
var cells = rows[i].split(",");
if (cells[0] == "" || cells[0] == undefined) {
var s = customers[customers.length - 1];
s.Ord.push(cells[2]);
} else {
var dt = customers.find(x => x.Number === cells[0]);
if (dt == undefined) {
if (cells.length > 1) {
var customer = {};
customer.Number = cells[0];
customer.Name = cells[1];
customer.Ord = new Array();
customer.Ord.push(cells[2]);
customer.Point_ID = cells[3];
customer.Point_Name = cells[4];
customer.Point_Type = cells[5];
customer.Set_ORD = cells[6];
customers.push(customer);
}
} else {
var dtt = dt;
dtt.Ord.push(cells[2]);
}
}
}
Actually you can use a light-weight library called any-text.
install dependencies
npm i -D any-text
use custom command to read files
var reader = require('any-text');
reader.getText(`path-to-file`).then(function (data) {
console.log(data);
});
or use async-await :
var reader = require('any-text');
const chai = require('chai');
const expect = chai.expect;
describe('file reader checks', () => {
it('check csv file content', async () => {
expect(
await reader.getText(`${process.cwd()}/test/files/dummy.csv`)
).to.contains('Lorem ipsum');
});
});
This is an old question and in 2022 there are many ways to achieve this. First, I think D3 is one of the best alternatives for data manipulation. It's open sourced and free to use, but also it's modular so we can import just the fetch module.
Here is a basic example. We will use the legacy mode so I will import the entire D3 library. Now, let's call d3.csv function and it's done. This function internally calls the fetch method therefore, it can open dataURL, url, files, blob, and so on.
const fileInput = document.getElementById('csv')
const outElement = document.getElementById('out')
const previewCSVData = async dataurl => {
const d = await d3.csv(dataurl)
console.log({
d
})
outElement.textContent = d.columns
}
const readFile = e => {
const file = fileInput.files[0]
const reader = new FileReader()
reader.onload = () => {
const dataUrl = reader.result;
previewCSVData(dataUrl)
}
reader.readAsDataURL(file)
}
fileInput.onchange = readFile
<script type="text/javascript" src="https://unpkg.com/d3#7.6.1/dist/d3.min.js"></script>
<div>
<p>Select local CSV File:</p>
<input id="csv" type="file" accept=".csv">
</div>
<pre id="out"><p>File headers will appear here</p></pre>
If we don't want to use any library and we just want to use pain JavaScrip (Vanilla JS) and we managed to get the text content of a file as data and we don't want to use d3 we can implement a simple function that will split the data into a text array then we will extract the first line and split into a headers array and the rest of the text will be the lines we will process. After, we map each line and extract its values and create a row object from an array created from mapping each header to its correspondent value from values[index].
NOTE:
We also we going to use a little trick array objects in JavaScript can also have attributes. Yes so we will define an attribute rows.headers and assign the headers to it.
const data = `heading_1,heading_2,heading_3,heading_4,heading_5
value_1_1,value_2_1,value_3_1,value_4_1,value_5_1
value_1_2,value_2_2,value_3_2,value_4_2,value_5_2
value_1_3,value_2_3,value_3_3,value_4_3,value_5_3`
const csvParser = data => {
const text = data.split(/\r\n|\n/)
const [first, ...lines] = text
const headers = first.split(',')
const rows = []
rows.headers = headers
lines.map(line => {
const values = line.split(',')
const row = Object.fromEntries(headers.map((header, i) => [header, values[i]]))
rows.push(row)
})
return rows
}
const d = csvParser(data)
// Accessing to the theaders attribute
const headers = d.headers
console.log({headers})
console.log({d})
Finally, let's implement a vanilla JS file loader using fetch and parsing the csv file.
const fetchFile = async dataURL => {
return await fetch(dataURL).then(response => response.text())
}
const csvParser = data => {
const text = data.split(/\r\n|\n/)
const [first, ...lines] = text
const headers = first.split(',')
const rows = []
rows.headers = headers
lines.map(line => {
const values = line.split(',')
const row = Object.fromEntries(headers.map((header, i) => [header, values[i]]))
rows.push(row)
})
return rows
}
const fileInput = document.getElementById('csv')
const outElement = document.getElementById('out')
const previewCSVData = async dataURL => {
const data = await fetchFile(dataURL)
const d = csvParser(data)
console.log({ d })
outElement.textContent = d.headers
}
const readFile = e => {
const file = fileInput.files[0]
const reader = new FileReader()
reader.onload = () => {
const dataURL = reader.result;
previewCSVData(dataURL)
}
reader.readAsDataURL(file)
}
fileInput.onchange = readFile
<script type="text/javascript" src="https://unpkg.com/d3#7.6.1/dist/d3.min.js"></script>
<div>
<p>Select local CSV File:</p>
<input id="csv" type="file" accept=".csv">
</div>
<pre id="out"><p>File contents will appear here</p></pre>
I used this file to test it
Here is another way to read an external CSV into Javascript (using jQuery).
It's a little bit more long winded, but I feel by reading the data into arrays you can exactly follow the process and makes for easy troubleshooting.
Might help someone else.
The data file example:
Time,data1,data2,data2
08/11/2015 07:30:16,602,0.009,321
And here is the code:
$(document).ready(function() {
// AJAX in the data file
$.ajax({
type: "GET",
url: "data.csv",
dataType: "text",
success: function(data) {processData(data);}
});
// Let's process the data from the data file
function processData(data) {
var lines = data.split(/\r\n|\n/);
//Set up the data arrays
var time = [];
var data1 = [];
var data2 = [];
var data3 = [];
var headings = lines[0].split(','); // Splice up the first row to get the headings
for (var j=1; j<lines.length; j++) {
var values = lines[j].split(','); // Split up the comma seperated values
// We read the key,1st, 2nd and 3rd rows
time.push(values[0]); // Read in as string
// Recommended to read in as float, since we'll be doing some operations on this later.
data1.push(parseFloat(values[1]));
data2.push(parseFloat(values[2]));
data3.push(parseFloat(values[3]));
}
// For display
var x= 0;
console.log(headings[0]+" : "+time[x]+headings[1]+" : "+data1[x]+headings[2]+" : "+data2[x]+headings[4]+" : "+data2[x]);
}
})
Hope this helps someone in the future!
A bit late but I hope it helps someone.
Some time ago even I faced a problem where the string data contained \n in between and while reading the file it used to read as different lines.
Eg.
"Harry\nPotter","21","Gryffindor"
While-Reading:
Harry
Potter,21,Gryffindor
I had used a library csvtojson in my angular project to solve this problem.
You can read the CSV file as a string using the following code and then pass that string to the csvtojson library and it will give you a list of JSON.
Sample Code:
const csv = require('csvtojson');
if (files && files.length > 0) {
const file: File = files.item(0);
const reader: FileReader = new FileReader();
reader.readAsText(file);
reader.onload = (e) => {
const csvs: string = reader.result as string;
csv({
output: "json",
noheader: false
}).fromString(csvs)
.preFileLine((fileLine, idx) => {
//Convert csv header row to lowercase before parse csv file to json
if (idx === 0) { return fileLine.toLowerCase() }
return fileLine;
})
.then((result) => {
// list of json in result
});
}
}
I use the jquery-csv to do this.
and I provide two examples as below
async function ReadFile(file) {
return await file.text()
}
function removeExtraSpace(stringData) {
stringData = stringData.replace(/,( *)/gm, ",") // remove extra space
stringData = stringData.replace(/^ *| *$/gm, "") // remove space on the beginning and end.
return stringData
}
function simpleTest() {
let data = `Name, Age, msg
foo, 25, hello world
bar, 18, "!! 🐬 !!"
`
data = removeExtraSpace(data)
console.log(data)
const options = {
separator: ",", // default "," . (You may want to Tab "\t" or somethings.
delimiter: '"', // default "
headers: true // default true
}
// const myObj = $.csv.toObjects(data, options)
const myObj = $.csv.toObjects(data) // If you want to use default options, then you can omit them.
console.log(myObj)
}
window.onload = () => {
const inputFile = document.getElementById("uploadFile")
inputFile.onchange = () => {
const inputValue = inputFile.value
if (inputValue === "") {
return
}
const selectedFile = document.getElementById('uploadFile').files[0]
const promise = new Promise(resolve => {
const fileContent = ReadFile(selectedFile)
resolve(fileContent)
})
promise.then(fileContent => {
// Use promise to wait for the file reading to finish.
console.log(fileContent)
fileContent = removeExtraSpace(fileContent)
const myObj = $.csv.toObjects(fileContent)
console.log(myObj)
})
}
}
<script src="https://cdnjs.cloudflare.com/ajax/libs/jquery/3.6.0/jquery.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/jquery-csv/1.0.11/jquery.csv.min.js"></script>
<label for="uploadFile">Demo 1</label>
<input type="file" id="uploadFile" accept=".csv"/>
<button onclick="simpleTest()">Demo 2</button>
With this function csvToObjs you can transform data-entries from format CSV to an array of objects.
function csvToObjs(string) {
const lines = data.split(/\r\n|\n/);
let [headings, ...entries] = lines;
headings = headings.split(',');
const objs = [];
entries.map(entry=>{
obj = entry.split(',');
objs.push(Object.fromEntries(headings.map((head, i)=>[head, obj[i]])));
})
return objs;
}
data = `heading1,heading2,heading3,heading4,heading5
value1_1,value2_1,value3_1,value4_1,value5_1
value1_2,value2_2,value3_2,value4_2,value5_2`
console.log(csvToObjs(data));

Categories

Resources