Closed focussing closed 5 years ago
@focussing The libraries only work asynchronously as they rely on interacting with a headless Chromium instance, which can only be done asynchronously. With async
/await
syntax you can write asynchronous code in a synchronous fashion. If you absolutely must do it synchronously, you may need to get inventive/hacky, like spawning a synchronous child process, however, this could have an impact on performance.
Yep, you are right. Sorry for bothering.
How about using existence of the generated SVG file? Would that be an option?
Can you please clarify what you mean? Possibly with an example.
Basically I want to convert a directory structure consisting of 1100 SVG files to the same structure but then consisting of JPG files.
What I do right now is to walk through the structure generated by readdirp. The readdirp callback is fired when the structure is complete. Next I loop through the files list and convert SVG to JPEG. Now I do that with a timer.
When the conversion is complete I want to delete the SVN file.
Another option would be to walk through the file structure again, and delete the SVN files. Of course that can be done in a lot of different ways.
Code snippet :)
// https://github.com/thlorenz/readdirp
var readdirp = require('readdirp');
var settings = {
root: outPath,
entryType: 'files',
fileFilter: '*.svg'
};
// In this example, this variable will store all the paths of the files and directories inside the providen path
var allFilePaths = [];
// Iterate recursively through a folder
readdirp(settings,
// This callback is executed everytime a file or directory is found inside the providen path
function(fileInfo) {
// Store the fullPath of the file/directory in our custom array
allFilePaths.push(
fileInfo.fullPath
);
},
// This callback is executed once
function(err, res) {
if (err) {
throw err;
}
// An array with all the fileEntry objects of the folder
// console.log(res);
//console.log(allFilePaths);
// ["c:/file.txt",""]
let i = 0;
// const stopAt = 10;
const stopAt = allFilePaths.length;
let timer = setInterval(function() {
let progress = i / stopAt * 100;
console.log(progress + '% ' + allFilePaths[i]);
const outputFilePath = convertFile(allFilePaths[i], options);
// console.log(outputFilePath);
if (++i == stopAt) {
clearInterval(timer);
console.log('conversion done')
}
}, 250)
// allFilePaths.forEach(function(f) {
// convertFile(f, options);
// });
}
);
This appears to be out of scope for this library and involves more of an understanding on managing asynchronous flows within Node.js. I'm not sure that I understand your timer logic, but the following should do what you're asking for:
const { convertFile } = require('convert-svg-to-jpeg');
const fs = require('fs');
const glob = require('glob');
const util = require('util');
const findFiles = util.promisify(glob);
const removeFile = util.promisify(fs.unlink);
(async() => {
const allFilePaths = await findFiles('*.svg', { absolute: true, cwd: outPath, nodir: true });
for (const filePath of allFilePaths) {
await convertFile(filePath, options);
await removeFile(filePath);
}
})();
All written like synchronous code with the benefit that it's fully asynchronous and non-blocking.
I'm not familiar with readdirp so I used glob for demonstration purposes.
This could even allow for parallel conversion and/or deletion if you wanted to, but that's even further out of the scope of this library. Lots of resources out there to help there though.
Yep you are right about out-of-scope. I just mentioned it related to checking whether the converted jpg is present and use that as a trigger.
But; thanks for your example code! I will test your solution and come back to that :) Have a great day!
I need to convert a big directory tree and delete the SVG file after creating the JPG file