bundling precompiled binary into electron app
Is there a good solution on how to include third party pre compiled binaries like imagemagick into an electron app? there are node.js modules but they are all wrappers or native binding to the system wide installed libraries. I wonder if it's possible to bundle precompiled binaries within the distribution.
See UPDATE below (this method isn't ideal now).
I did find a solution to this, but I have no idea if this is considered best practice. I couldn't find any good documentation for including 3rd party precompiled binaries, so I just fiddled with it until it finally worked with my ffmpeg binary. Here's what I did (starting with the electron quick start, node.js v6):
Mac OS X method
From the app directory I ran the following commands in Terminal to include the ffmpeg binary as a module:
mkdir node_modules/ffmpeg
cp /usr/local/bin/ffmpeg node_modules/ffmpeg/
cd node_modules/.bin
ln -s ../ffmpeg/ffmpeg ffmpeg
(replace /usr/local/bin/ffmpeg
with your current binary path, download it from here) Placing the link allowed electron-packager to include the binary I saved to node_modules/ffmpeg/
.
Then to get the bundled app path (so that I could use an absolute path for my binary... relative paths didn't seem to work no matter what I did) I installed the npm package app-root-dir by running the following command:
npm i -S app-root-dir
Now that I had the root app directory, I just append the subfolder for my binary and spawned from there. This is the code that I placed in renderer.js:.
var appRootDir = require('app-root-dir').get();
var ffmpegpath=appRootDir+'/node_modules/ffmpeg/ffmpeg';
console.log(ffmpegpath);
const
spawn = require( 'child_process' ).spawn,
ffmpeg = spawn( ffmpegpath, ['-i',clips_input[0]]); //add whatever switches you need here
ffmpeg.stdout.on( 'data', data => {
console.log( `stdout: ${data}` );
});
ffmpeg.stderr.on( 'data', data => {
console.log( `stderr: ${data}` );
});
Windows Method
-
Open your electron base folder (electron-quick-start is the default name), then go into the node_modules folder. Create a folder there called ffmpeg, and copy your static binary into this directory. Note: it must be the static version of your binary, for ffmpeg I grabbed the latest Windows build here.
-
To get the bundled app path (so that I could use an absolute path for my binary... relative paths didn't seem to work no matter what I did) I installed the npm package app-root-dir by running the following command from a command prompt in my app directory:
npm i -S app-root-dir
-
Within your node_modules folder, navigate to the .bin subfolder. You need to create a couple of text files here to tell node to include the binary exe file you just copied. Use your favorite text editor and create two files, one named
ffmpeg
with the following contents:#!/bin/sh basedir=$(dirname "$(echo "$0" | sed -e 's,\\,/,g')") case `uname` in *CYGWIN*) basedir=`cygpath -w "$basedir"`;; esac if [ -x "$basedir/node" ]; then "$basedir/node" "$basedir/../ffmpeg/ffmpeg" "$@" ret=$? else node "$basedir/../ffmpeg/ffmpeg" "$@" ret=$? fi exit $ret
And the the second text file, named ffmpeg.cmd
:
@IF EXIST "%~dp0\node.exe" (
"%~dp0\node.exe" "%~dp0\..\ffmpeg\ffmpeg" %*
) ELSE (
@SETLOCAL
@SET PATHEXT=%PATHEXT:;.JS;=;%
node "%~dp0\..\ffmpeg\ffmpeg" %*
)
Next you can run ffmpeg in your Windows electron distribution (in renderer.js) as follows (I'm using the app-root-dir node module as well). Note the quotes added to the binary path, if your app is installed to a directory with spaces (eg C:\Program Files\YourApp
) it won't work without these.
var appRootDir = require('app-root-dir').get();
var ffmpegpath = appRootDir + '\\node_modules\\ffmpeg\\ffmpeg';
const
spawn = require( 'child_process' ).spawn;
var ffmpeg = spawn( 'cmd.exe', ['/c', '"'+ffmpegpath+ '"', '-i', clips_input[0]]); //add whatever switches you need here, test on command line first
ffmpeg.stdout.on( 'data', data => {
console.log( `stdout: ${data}` );
});
ffmpeg.stderr.on( 'data', data => {
console.log( `stderr: ${data}` );
});
UPDATE: Unified Simple Method
Well, as time as rolled on and Node has updated, this method is no longer the easiest way to include precompiled binaries. It still works, but when npm install
is run the binary folders under node_modules will be deleted and have to be replaced again. The below method works for Node v12.
This new method obviates the need to symlink, and works similarly for Mac and Windows. Relative paths seem to work now.
-
You will still need appRootDir:
npm i -S app-root-dir
-
Create a folder under your app's root directory named
bin
and place your precompiled static binaries here, I'm usingffmpeg
as an example. -
Use the following code in your renderer script:
const appRootDir = require('app-root-dir').get();
const ffmpegpath = appRootDir + '/bin/ffmpeg';
const spawn = require( 'child_process' ).spawn;
const child = spawn( ffmpegpath, ['-i', inputfile, 'out.mp4']); //add whatever switches you need here, test on command line first
child.stdout.on( 'data', data => {
console.log( `stdout: ${data}` );
});
child.stderr.on( 'data', data => {
console.log( `stderr: ${data}` );
});
Here's another method, tested with Mac and Windows so far. Requires 'app-root-dir' package, doesn't require adding anything manually to node_modules dir.
Put your files under resources/$os/, where $os is either "mac", "linux", or "win". The build process will copy files from those directories as per build target OS.
Put
extraFiles
option in your build configs as follows:
package.json
"build": {
"extraFiles": [
{
"from": "resources/${os}",
"to": "Resources/bin",
"filter": ["**/*"]
}
],
- Use something like this to determine the current platform.
get-platform.js
import { platform } from 'os';
export default () => {
switch (platform()) {
case 'aix':
case 'freebsd':
case 'linux':
case 'openbsd':
case 'android':
return 'linux';
case 'darwin':
case 'sunos':
return 'mac';
case 'win32':
return 'win';
}
};
- Call the executable from your app depending on env and OS. Here I am assuming built versions are in production mode and source versions in other modes, but you can create your own calling logic.
import { join as joinPath, dirname } from 'path';
import { exec } from 'child_process';
import appRootDir from 'app-root-dir';
import env from './env';
import getPlatform from './get-platform';
const execPath = (env.name === 'production') ?
joinPath(dirname(appRootDir.get()), 'bin'):
joinPath(appRootDir.get(), 'resources', getPlatform());
const cmd = `${joinPath(execPath, 'my-executable')}`;
exec(cmd, (err, stdout, stderr) => {
// do things
});
I think I was using electron-builder as base, the env file generation comes with it. Basically it's just a JSON config file.
The above answers helped me figure out how it could done but there is a much efficient way to distribute binary files.
Taking cues from tsuriga's answer, here is my code:
Note: replace or add OS path as required.
Update - 4 Dec 2020
This answer has been updated. Find the previous code to the bottom of this answer.
- Download the needed packages
yarn add electron-root-path electron-is-packaged
# or
npm i electron-root-path electron-is-packaged
- Create a directory ./resources/mac/bin
- Place you binaries inside this folder
- Create a file ./app/binaries.js and paste the following code:
import path from 'path';
import { rootPath as root } from 'electron-root-path';
import { isPackaged } from 'electron-is-packaged';
import { getPlatform } from './getPlatform';
const IS_PROD = process.env.NODE_ENV === 'production';
const binariesPath =
IS_PROD && isPackaged // the path to a bundled electron app.
? path.join(root, './Contents', './Resources', './bin')
: path.join(root, './build', getPlatform(), './bin');
export const execPath = path.resolve(
path.join(binariesPath, './exec-file-name')
);
- Create a file ./app/get-platform.js and paste the following code:
'use strict';
import { platform } from 'os';
export default () => {
switch (platform()) {
case 'aix':
case 'freebsd':
case 'linux':
case 'openbsd':
case 'android':
return 'linux';
case 'darwin':
case 'sunos':
return 'mac';
case 'win32':
return 'win';
}
};
- Add these lines inside the ./package.json file:
"build": {
....
"extraFiles": [
{
"from": "resources/mac/bin",
"to": "Resources/bin",
"filter": [
"**/*"
]
}
],
....
},
- import binary as:
import { execPath } from './binaries';
#your program code:
var command = spawn(execPath, arg, {});
Why this is better?
-
The above answers require an additional package called app-root-dir
-
tsuriga's answer doesn't handle the (env=production) build or the pre-packed versions properly. He/she has only taken care of development and post-packaged versions.
Previous answer
- Avoid using
electron.remote
as it is getting depreciated -
app.getAppPath
might throw errors in the main process.
./app/binaries.js
'use strict';
import path from 'path';
import { remote } from 'electron';
import getPlatform from './get-platform';
const IS_PROD = process.env.NODE_ENV === 'production';
const root = process.cwd();
const { isPackaged, getAppPath } = remote.app;
const binariesPath =
IS_PROD && isPackaged
? path.join(path.dirname(getAppPath()), '..', './Resources', './bin')
: path.join(root, './resources', getPlatform(), './bin');
export const execPath = path.resolve(path.join(binariesPath, './exec-file-name'));
tl;dr:
yes you can! but it requires you to write your own self-contained addon which does not make any assumptions on system libraries. Moreover in some cases you have to make sure that your addon is compiled for the desired OS.
Lets break this question in several parts:
- Addons (Native modules):
Addons are dynamically linked shared objects.
In other words you can just write your own addon with no dependency on system wide libraries (e.g. by statically linking required modules) containing all the code you need.
You have to consider that such approach is OS-specific, meaning that you need to compile your addon for each OS that you want to support! (depending on what other libraries you may use)
- Native modules for electron:
The native Node modules are supported by Electron, but since Electron is using a different V8 version from official Node, you have to manually specify the location of Electron's headers when building native modules
This means that a native module which has been built against node headers must be rebuilt to be used inside electron. You can find how in electron docs.
- Bundle modules with electron app:
I suppose you want to have your app as a stand-alone executable without requiring users to install electron on their machines. If so, I can suggest using electron-packager.