Chunk/Stream API data using Node.js - node.js

Question

We have requirement where we need to write a node application which can read URL of image from database (approx more than million). Use image-size npm package to retrieve image meta data like height, width. Here should be an API which can list out result.
I am able to console log data but when i convert it to API, i need to chunk data so it can start appearing on browser and i'm unable to do that and need help. Here is my code
var express = require('express');
var url = require('url');
var http = require('http');
var sizeOf = require('image-size');
const sql = require('mssql');
var app = express();
var port = process.env.PORT || 3000;
const hostname = 'localhost';
var config1 = {
user: '*********',
password: '*********',
server: '*********',
database: '*******',
port: 1433,
debug: true,
options: {
encrypt: false // Use this if you're on Windows Azure
}
};
app.get('/', function(req, res){
//res.writeHead(200, { 'Content-Type': 'application/json' });
var finalResult = [];
sql.close();
sql.connect(config1, function (err) {
if (err) console.log(err);
const request = new sql.Request()
var myQuery = `select imagename from media`;
request.stream = true;
request.query(myQuery);
request.on('row', row => {
//console.log('Image : ' + row.ImageUrl);
if (row.ImageUrl != ''){
if (row.ImageUrl.indexOf('http') < 0)
row.ImageUrl = "http:" + row.ImageUrl;
var options = url.parse(row.ImageUrl);
http.get(options, function (response) {
if (response.statusCode == 200)
{
var chunks = [];
response.on('data', function (chunk) {
chunks.push(chunk);
}).on('end', function() {
var buffer = Buffer.concat(chunks);
//console.log(options.href);
//console.log(sizeOf(buffer).height);
var result = {};
result.MaskUrl = row.MaskUrl;
result.ImageUrl = options.href;
result.Height = sizeOf(buffer).height;
result.Width = sizeOf(buffer).width;
result.statusCode = 200;
finalResult.push(result);
//console.log(result);
console.log(finalResult);
res.write(result, function(){
res.end();
});
});
}
else
{
var result = {};
result.MaskUrl = row.MaskUrl;
result.ImageUrl = options.href;
result.Height = 0;
result.Width = 0;
result.statusCode = response.statusCode;
finalResult.push(result);
console.log(result);
res.write(result, function(){
res.end();
});
}
});
}
})
request.on('error', err => {
console.log ('Error for ' + row.ImageUrl );
})
request.on('done', err => {
console.log('Last Time' + finalResult.length);
})
// request.query(myQuery,(err,result) =>{
// console.log(result);
// });
});
console.log('Last Time' + finalResult.length);
res.send(finalResult);
});
app.listen(port, hostname, function(){
console.log('ImageSize running on PORT: ' + port);
});
I tried res.write, res.end without any success.

Answer 1

The probable reason for your problem is that here:
res.write(result, function(){
res.end();
});
You end and close the request just after the first image is read.
I would rewrite the code a little and use some functional framework, like scramjet, to stream the data straight from the DB. As Nicholas pointed out it's not super easy to run your code so I'm writing blindly - but if you fix any of my obvious error this should just work:
First:
npm install scramjet JSONStream node-fetch
Next, try this code:
var express = require('express');
var sizeOf = require('image-size');
const sql = require('mssql');
var app = express();
var port = process.env.PORT || 3000;
const hostname = 'localhost';
const {DataStream} = require('scramjet');
const fetch = require('node-fetch');
const JSONStream = require('JSONStream');
var config1 = {
user: '*********',
password: '*********',
server: '*********',
database: '*******',
port: 1433,
debug: true,
options: {
encrypt: false // Use this if you're on Windows Azure
}
};
app.get('/', function(req, res, next){
// you should consider not doing these two lines on each request,
// but I don't want to mess you code...
sql.close();
sql.connect(config1, function (err) {
if (err) next(err);
res.writeHead(200, { 'Content-Type': 'application/json' });
const request = new sql.Request();
var myQuery = `select imagename from media`;
request.stream = true;
request.query(myQuery);
const stream = new DataStream();
request.on('row', row => stream.write(row));
stream.filter(
row => row.ImageUrl !== ''
)
.map(
async row => {
if (row.ImageUrl.indexOf('http') !== 0) // url must start with http.
row.ImageUrl = "http:" + row.ImageUrl;
const response = await fetch(row.ImageUrl);
let size = {width:0, height:0};
if (response.status === 200) {
const buffer = await response.buffer();
size = sizeOf(buffer);
}
return {
MaskUrl: row.MaskUrl,
ImageUrl: row.ImageUrl,
Height: size.height,
Width: size.width,
statusCode: response.status
};
}
)
.pipe(
JSONStream.stringify()
).pipe(
res
);
request.on('error', () => {
res.writeHead(500, { 'Content-Type': 'application/json' });
stream.end("{error:true}");
});
request.on('done', () => stream.end());
});
});
app.listen(port, hostname, function(){
console.log('ImageSize running on PORT: ' + port);
});