Related
I had my tests designed to be run in an environment which assumed to be in UTC.
I used cy.clock() to set a date-time in UTC format. The Web App to test will display current browser date-time in UTC format.
I had TZ=UTC in my environment variables and I run my test in Cypress Runner in Linux machine using TZ=UTC npx cypress open. The date and the time display on the web app is same as the date time I set using cy.clock()
However, the same method seems doesn't work in Windows. I couldn't use TZ=UTC npx cypress open in Windows. Instead I used npx cypress open. The time is correct but the date differs by one day. If I change my windows pc timezone to UTC, the both date and time is as expected.
Is there any other way other than change the computer timezone in Windows ? I do not need to do so in Linux
My tests runs fine in CI without the need for TZ before the script. Setting it in environment variable in cypress.json is sufficient.
Can someone please enlightened me on this please ?
My cypress.json
{
"projectId": "blahblahbah",
"baseUrl": "http://localhost:4200",
"numTestsKeptInMemory": 10,
"chromeWebSecurity": false,
"TZ": "UTC",
"env": {
"BASE_URL": "http://localhost:4200",
"TZ": "UTC",
},
"integrationFolder": "cypress/tests/",
"retries": {
"runMode": 2,
"openMode": 0
},
"reporter": "cypress-mochawesome-reporter",
"reporterOptions": {
"charts": true,
"reportPageTitle": "XXXX Test",
"embeddedScreenshots": true,
"inlineAssets": true,
"reportDir": "cypress/reports/",
"timestamp": "mmddyyyy_HHMMss"
},
"video": false
}
I think the Windows equiv to TZ=UTC npx cypress open is set TZ=UTC && npx cypress open.
There's also cross-env, allows a single script to work across different OS.
// package.json
{
"scripts": {
"cy:open": "cross-env TZ=UTC npx cypress open"
}
}
But it seems to be fussy about the shell
Windows Issues
Please note that npm uses cmd by default and that doesn't support command substitution, so if you want to leverage that, then you need to update your .npmrc to set the script-shell to powershell.
Another option is run-script-os
See Run different NPM scripts depending on the platform
// package.json
{
"scripts": {
"test": "run-script-os",
"test:win32": "set TZ=UTC && npx cypress open",
"test:darwin:linux": "TZ=UTC npx cypress open",
}
}
I found a workaround in Windows other than setting your computer clock.
You can use WSL2 to run Linux command in Windows 11. If you do not have Windows 11, update it. If your computer cannot support Windows 11, buy a new one.
Make sure you are in Windows 11 Build 22000
Install Docker. Followed the instruction to get Docker running successfully. That also includes setting up WSL2. In this step you will be directed to Microsoft guide to install WSL 2 and install a Linux distro from Microsoft Store. Make sure you did RESTART
Install driver for vGPU
Intel GPU driver for WSL
AMD GPU driver for WSL
NVIDIA GPU driver for WSL
Links can be obtained here: https://learn.microsoft.com/en-us/windows/wsl/tutorials/gui-apps
After that Open Powershell as Admin
wsl --update followed by wsl --shutdown
RESTART COMPUTER
Follow the below sub-steps if you wish to try to luck or increase chances of success.
Can do this for non Windows 11 PC
a. Install VcXsrv
(Multiple windows and Start no client is recommended)
On the page that lets you enable extra settings, disable access control
b. In VSCode, open project folder and a new terminal using dropdown. (There is a + button) Look for Ubuntu (WSL). Here you have access to the bash shell.
c. Type the following command in sequence
set DISPLAY variable to the IP automatically assigned to WSL2
export DISPLAY=$(cat /etc/resolv.conf | grep nameserver | awk '{print $2; exit;}'):0.0
echo $DISPLAY
# expect something like 192.168.64.1:0.0
sudo /etc/init.d/dbus start &> /dev/null
sudo visudo -f /etc/sudoers.d/dbus
After this command, an editor will launch. Add the following line with your distro username.
<your_username> ALL = (root) NOPASSWD: /etc/init.d/dbus
Remember to set permission in firewall to allow VcXsrv on domain, private and public network.
RESTART COMPUTER
In VSCode, open project folder and a new terminal using dropdown. (There is a + button) Look for Ubuntu (WSL). Here you have access to the bash shell.
Install cypress dependencies by using
sudo apt-get install libgtk2.0-0 libgtk-3-0 libgbm-dev libnotify-dev libgconf-2-4 libnss3 libxss1 libasound2 libxtst6 xauth xvfb
Install npm and nodejs by using
'''
sudo apt-get install npm nodejs
'''
Note: Prior to 3 and . It is assumed that you have already done apt-get update
npm i
Open another New Terminal ( Ubuntu WSL) to run npm serve
You can attempt to
npx cypress open
to see if it works well.
Remember npm serve, npm i and npx cypress open need to be in bash ( Ubuntu WSL)
If you are using Windows 11 Build 22000 and above, you can try to exit VcXsrv, and do npx cypress open again. I tried and it still work.
Reference:
https://learn.microsoft.com/en-us/windows/wsl/tutorials/gui-apps
https://shouv.medium.com/how-to-run-cypress-on-wsl2-989b83795fb6
ANOTHER KISS SOLUTION: USE GIT BASH
Just launch Git Bash from VSCODE followed by TZ=UTC npx cypress open
Cypress manual for configuring Reporters contains rm command which doesn't work in Windows, namely
rm cypress/results/* || true
I believe that this is the manual bug regarding using Linux command in Windows environment, so the question is how it can be done in Windows? Moreover, I suppose to use the config file containing this command in Docker container which uses Linux core I believe. So the another problem is if I convert the command for Windows will I be able to use this file in my Docker container? Could somebody point me out the right way of using this command for the both OS? here is the link to Cypress manual containing this command https://docs.cypress.io/guides/tooling/reporters#Examples
rmdir can be used in windows
"delete:reports": "rmdir /S /Q cypress\\results && mkdir cypress\\results",
I do not think this will work for docker.
Cross platform,
yarn add -D rimraf mkdirp
"delete:reports": "rimraf cypress/results && mkdirp cypress/results",
How do I use a local version of a module in node.js. For example, in my app, I installed coffee-script:
npm install coffee-script
This installs it in ./node_modules and the coffee command is in ./node_modules/.bin/coffee. Is there a way to run this command when I'm in my project's main folder? I guess I'm looking for something similar to bundle exec in bundler. Basically, I'd like to specify a version of coffee-script that everyone involved with the project should use.
I know I can add the -g flag to install it globally so coffee works fine anywhere, but what if I wanted to have different versions of coffee per project?
UPDATE: As Seyeong Jeong points out in their answer below, since npm 5.2.0 you can use npx [command], which is more convenient.
OLD ANSWER for versions before 5.2.0:
The problem with putting
./node_modules/.bin
into your PATH is that it only works when your current working directory is the root of your project directory structure (i.e. the location of node_modules)
Independent of what your working directory is, you can get the path of locally installed binaries with
npm bin
To execute a locally installed coffee binary independent of where you are in the project directory hierarchy you can use this bash construct
PATH=$(npm bin):$PATH coffee
I aliased this to npm-exec
alias npm-exec='PATH=$(npm bin):$PATH'
So, now I can
npm-exec coffee
to run the correct copy of coffee no matter of where I am
$ pwd
/Users/regular/project1
$ npm-exec which coffee
/Users/regular/project1/node_modules/.bin/coffee
$ cd lib/
$ npm-exec which coffee
/Users/regular/project1/node_modules/.bin/coffee
$ cd ~/project2
$ npm-exec which coffee
/Users/regular/project2/node_modules/.bin/coffee
You don't have to manipulate $PATH anymore!
From npm#5.2.0, npm ships with npx package which lets you run commands from a local node_modules/.bin or from a central cache.
Simply run:
$ npx [options] <command>[#version] [command-arg]...
By default, npx will check whether <command> exists in $PATH, or in the local project binaries, and execute that.
Calling npx <command> when <command> isn't already in your $PATH will automatically install a package with that name from the NPM registry for you, and invoke it. When it's done, the installed package won’t be anywhere in your globals, so you won’t have to worry about pollution in the long-term. You can prevent this behaviour by providing --no-install option.
For npm < 5.2.0, you can install npx package manually by running the following command:
$ npm install -g npx
Use the npm bin command to get the node modules /bin directory of your project
$ $(npm bin)/<binary-name> [args]
e.g.
$ $(npm bin)/bower install
Use npm run[-script] <script name>
After using npm to install the bin package to your local ./node_modules directory, modify package.json to add <script name> like this:
$ npm install --save learnyounode
$ edit packages.json
>>> in packages.json
...
"scripts": {
"test": "echo \"Error: no test specified\" && exit 1",
"learnyounode": "learnyounode"
},
...
$ npm run learnyounode
It would be nice if npm install had a --add-script option or something or if npm run would work without adding to the scripts block.
update: If you're on the recent npm (version >5.2)
You can use:
npx <command>
npx looks for command in .bin directory of your node_modules
old answer:
For Windows
Store the following in a file called npm-exec.bat and add it to your %PATH%
#echo off
set cmd="npm bin"
FOR /F "tokens=*" %%i IN (' %cmd% ') DO SET modules=%%i
"%modules%"\%*
Usage
Then you can use it like
npm-exec <command> <arg0> <arg1> ...
For example
To execute wdio installed in local node_modules directory, do:
npm-exec wdio wdio.conf.js
i.e. it will run .\node_modules\.bin\wdio wdio.conf.js
Update: I no longer recommend this method, both for the mentioned security reasons and not the least the newer npm bin command. Original answer below:
As you have found out, any locally installed binaries are in ./node_modules/.bin. In order to always run binaries in this directory rather than globally available binaries, if present, I suggest you put ./node_modules/.bin first in your path:
export PATH="./node_modules/.bin:$PATH"
If you put this in your ~/.profile, coffee will always be ./node_modules/.bin/coffee if available, otherwise /usr/local/bin/coffee (or whatever prefix you are installing node modules under).
Use npm-run.
From the readme:
npm-run
Find & run local executables from node_modules
Any executable available to an npm lifecycle script is available to npm-run.
Usage
$ npm install mocha # mocha installed in ./node_modules
$ npm-run mocha test/* # uses locally installed mocha executable
Installation
$ npm install -g npm-run
TL;DR: Use npm exec with npm#>=7.
The npx command which was mentioned in other answers has been completely rewritten in npm#7 which ships by default with node#15 and can be installed on node#>=10. The implementation is now equal to the newly introduced npm exec command, which is similar but not equal to the previous npx command implementation.
One difference is e.g. that it always interactively asks if a dependency should be downloaded when it is not already installed (can also be overwritten with the params --yes or --no).
Here is an example for npm exec. The double dashes (--) separates the npm exec params from the actual command params:
npm exec --no -- jest --coverage
See also the updated, official documentation to npm exec.
If you want to keep npm, then npx should do what you need.
If switching to yarn (a npm replacement by facebook) is an option for you, then you can call:
yarn yourCmd
scripts inside the package.json will take precedence, if none is found it will look inside the ./node_modules/.bin/ folder.
It also outputs what it ran:
$ yarn tsc
yarn tsc v0.27.5
$ "/home/philipp/rate-pipeline/node_modules/.bin/tsc"
So you don't have to setup scripts for each command in your package.json.
If you had a script defined at .scripts inside your package.json:
"tsc": "tsc" // each command defined in the scripts will be executed from `./node_modules/.bin/` first
yarn tsc would be equivalent to yarn run tsc or npm run tsc:
yarn tsc
yarn tsc v0.27.5
$ tsc
The PATH solution has the issue that if $(npm bin) is placed in your .profile/.bashrc/etc it is evaluated once and is forever set to whichever directory the path was first evaluated in. If instead you modify the current path then every time you run the script your path will grow.
To get around these issues, I create a function and used that. It doesn't modify your environment and is simple to use:
function npm-exec {
$(npm bin)/$#
}
This can then be used like this without making any changes to your environment:
npm-exec r.js <args>
I prefer to not rely on shell aliases or another package.
Adding a simple line to scripts section of your package.json, you can run local npm commands like
npm run webpack
package.json
{
"scripts": {
"test": "echo \"Error: no test specified\" && exit 1",
"webpack": "webpack"
},
"devDependencies": {
"webpack": "^4.1.1",
"webpack-cli": "^2.0.11"
}
}
If you want your PATH variable to correctly update based on your current working directory, add this to the end of your .bashrc-equivalent (or after anything that defines PATH):
__OLD_PATH=$PATH
function updatePATHForNPM() {
export PATH=$(npm bin):$__OLD_PATH
}
function node-mode() {
PROMPT_COMMAND=updatePATHForNPM
}
function node-mode-off() {
unset PROMPT_COMMAND
PATH=$__OLD_PATH
}
# Uncomment to enable node-mode by default:
# node-mode
This may add a short delay every time the bash prompt gets rendered (depending on the size of your project, most likely), so it's disabled by default.
You can enable and disable it within your terminal by running node-mode and node-mode-off, respectively.
I've always used the same approach as #guneysus to solve this problem, which is creating a script in the package.json file and use it running npm run script-name.
However, in the recent months I've been using npx and I love it.
For example, I downloaded an Angular project and I didn't want to install the Angular CLI globally. So, with npx installed, instead of using the global angular cli command (if I had installed it) like this:
ng serve
I can do this from the console:
npx ng serve
Here's an article I wrote about NPX and that goes deeper into it.
Same #regular 's accepted solution, but Fish shell flavour
if not contains (npm bin) $PATH
set PATH (npm bin) $PATH
end
zxc is like "bundle exec" for nodejs. It is similar to using PATH=$(npm bin):$PATH:
$ npm install -g zxc
$ npm install gulp
$ zxc which gulp
/home/nathan/code/project1/node_modules/.bin/gulp
You can also use direnv and change the $PATH variable only in your working folder.
$ cat .envrc
> export PATH=$(npm bin):$PATH
Add this script to your .bashrc. Then you can call coffee or anyhting locally. This is handy for your laptop, but don't use it on your server.
DEFAULT_PATH=$PATH;
add_local_node_modules_to_path(){
NODE_MODULES='./node_modules/.bin';
if [ -d $NODE_MODULES ]; then
PATH=$DEFAULT_PATH:$NODE_MODULES;
else
PATH=$DEFAULT_PATH;
fi
}
cd () {
builtin cd "$#";
add_local_node_modules_to_path;
}
add_local_node_modules_to_path;
note: this script makes aliase of cd command, and after each call of cd it checks node_modules/.bin and add it to your $PATH.
note2: you can change the third line to NODE_MODULES=$(npm bin);. But that would make cd command too slow.
For Windows use this:
/* cmd into "node_modules" folder */
"%CD%\.bin\grunt" --version
I encountered the same problem and I don't particularly like using aliases (as regular's suggested), and if you don't like them too then here's another workaround that I use, you first have to create a tiny executable bash script, say setenv.sh:
#!/bin/sh
# Add your local node_modules bin to the path
export PATH="$(npm bin):$PATH"
# execute the rest of the command
exec "$#"
and then you can then use any executables in your local /bin using this command:
./setenv.sh <command>
./setenv.sh 6to5-node server.js
./setenv.sh grunt
If you're using scripts in package.json then:
...,
scripts: {
'start': './setenv.sh <command>'
}
I'd love to know if this is an insecure/bad idea, but after thinking about it a bit I don't see an issue here:
Modifying Linus's insecure solution to add it to the end, using npm bin to find the directory, and making the script only call npm bin when a package.json is present in a parent (for speed), this is what I came up with for zsh:
find-up () {
path=$(pwd)
while [[ "$path" != "" && ! -e "$path/$1" ]]; do
path=${path%/*}
done
echo "$path"
}
precmd() {
if [ "$(find-up package.json)" != "" ]; then
new_bin=$(npm bin)
if [ "$NODE_MODULES_PATH" != "$new_bin" ]; then
export PATH=${PATH%:$NODE_MODULES_PATH}:$new_bin
export NODE_MODULES_PATH=$new_bin
fi
else
if [ "$NODE_MODULES_PATH" != "" ]; then
export PATH=${PATH%:$NODE_MODULES_PATH}
export NODE_MODULES_PATH=""
fi
fi
}
For bash, instead of using the precmd hook, you can use the $PROMPT_COMMAND variable (I haven't tested this but you get the idea):
__add-node-to-path() {
if [ "$(find-up package.json)" != "" ]; then
new_bin=$(npm bin)
if [ "$NODE_MODULES_PATH" != "$new_bin" ]; then
export PATH=${PATH%:$NODE_MODULES_PATH}:$new_bin
export NODE_MODULES_PATH=$new_bin
fi
else
if [ "$NODE_MODULES_PATH" != "" ]; then
export PATH=${PATH%:$NODE_MODULES_PATH}
export NODE_MODULES_PATH=""
fi
fi
}
export PROMPT_COMMAND="__add-node-to-path"
I am a Windows user and this is what worked for me:
// First set some variable - i.e. replace is with "xo"
D:\project\root> set xo="./node_modules/.bin/"
// Next, work with it
D:\project\root> %xo%/bower install
Good Luck.
In case you are using fish shell and do not want to add to $path for security reason. We can add the below function to run local node executables.
### run executables in node_module/.bin directory
function n
set -l npmbin (npm bin)
set -l argvCount (count $argv)
switch $argvCount
case 0
echo please specify the local node executable as 1st argument
case 1
# for one argument, we can eval directly
eval $npmbin/$argv
case '*'
set --local executable $argv[1]
# for 2 or more arguments we cannot append directly after the $npmbin/ since the fish will apply each array element after the the start string: $npmbin/arg1 $npmbin/arg2...
# This is just how fish interoperate array.
set --erase argv[1]
eval $npmbin/$executable $argv
end
end
Now you can run thing like:
n coffee
or more arguments like:
n browser-sync --version
Note, if you are bash user, then #Bob9630 answers is the way to go by leveraging bash's $#, which is not available in fishshell.
I propose a new solution I have developed (05/2021)
You can use lpx https://www.npmjs.com/package/lpx to
run a binary found in the local node_modules/.bin folder
run a binary found in the node_modules/.bin of a workspace root from anywhere in the workspace
lpx does not download any package if the binary is not found locally (ie not like npx)
Example :
lpx tsc -b -w will run tsc -b -w with the local typescript package
Include coffee-script in package.json with the specific version required in each project, typically like this:
"dependencies":{
"coffee-script": ">= 1.2.0"
Then run npm install to install dependencies in each project. This will install the specified version of coffee-script which will be accessible locally to each project.
My OS is win7, and I using MSYS2(version:MSYS_NT-6.1), Please give advice how to install nodejs and npm on this terminal, Thanks!
I found a solution for resolving the problem,
64bit env.
pacman -S mingw-w64-x86_64-nodejs
32bit env.
pacman -S mingw-w64-i686-nodejs
after installed, Open terminal
$ node -v
v6.11.0
As of 2020, the package mingw-w64-x86_64-nodejs is not available any more. The simplest way to have Node.js, npm and git installed on a Windows machine is using the official Windows installers:
Git: https://git-scm.com/download/win
Node.js (npm is shipped with it): https://nodejs.org/en/download/
After installation, open a command prompt (click on start, and then type cmd and [ENTER]) and verify that all three tools are there:
git --version
node --version
npm --version
Later on, to update Node.js, simply reinstall it from the same source.
I wasted a lot of time on this. My solution is:
Download the Windows Binary (.zip) from nodejs site (https://nodejs.org/en/download/current/)
Extract it to some folder
Add that folder to the PATH env variable
It does work to use the Windows installer, and Node.js helpfully provides bash-script versions of npm and npx in C:\Program Files\nodejs\ to help streamline the process.
However, contrary to Cerclanism's comment # jmgonet's answer, you should not use --full-path with MinGW, no matter what terminal you're using, since that will by default bring the entire Windows path into your MinGW environment.
(Assuming you're a typical Windows developer with things like MSVC, Windows Python, and etc. install dirs on your path, containing plenty of names that clash with MinGW path members, you can see how that might bite you at some point down the road. My full Windows CMD.exe %PATH% is 1236 characters! I don't want all that sucked into MinGW.)
Instead, you should add the nodejs install dir to your MinGW shell $PATH, say by using everyone's favorite ~/.profile/~/.zprofile $PATH-munging trick:
# Append node.js to path
case ${PATH} in
*"/c/program files/nodejs"*)
;;
*)
export PATH="$PATH:/c/program files/nodejs:"
;;
esac
You'll probably also want to set some configuration, since by default Windows npm will use ${APPDATA}/npm for prefix, ${LOCALAPPDATA}/npm-cache for cache, C:\Windows\system32\cmd.exe for shell, etc.
# To view the full config including all defaults and overrides
npm config ls -l
# To view the active config for the specified environment
npm config list -L {global,user,project}
Maybe I was just confused, but to me it seemed, from what the configs show/say, that setting prefix= in my user config would override even local installs. (The project-specific ones where you npm install without --global, directly into a node_modules subdir of the current dir.) But after testing, happily I can report that's not the case, so it's safe to override the builtin prefix= from your $HOME/.npmrc.
Whether or not you move the cache= or let it stay at C:\Users\<you>\AppData\Local\npm-cache\ is your call. I'm sure it'll work that way. (Well, maybe not from an MSYS shell, but from MinGW it should be fine.)
There are minor differences I haven't overcome, but the only one that comes to mind right now is:
npm help <command> opens a browser window to HTML documentation, instead of displaying man page content directly in the terminal like it does on Linux. (Makes sense, as I don't think the manpages are even installed on Windows. Still disconcerting, though.)
You can simply install nvm, then install nodejs through there. In your MSYS2 shell just run the following to download and install nvm. Its better to go directly here and copy the download commands as the version numbers will change in the url.
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.2/install.sh | bash
or
wget -qO- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.2/install.sh | bash
then run the following to setup nvm on your bash path:
export NVM_DIR="$([ -z "${XDG_CONFIG_HOME-}" ] && printf %s "${HOME}/.nvm" || printf %s "${XDG_CONFIG_HOME}/nvm")"
[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh" # This loads nvm
After running those commands you can use nvm install 16 or whatever node major version number you want. Type just nvm to get a list of available commands.
How do I use a local version of a module in node.js. For example, in my app, I installed coffee-script:
npm install coffee-script
This installs it in ./node_modules and the coffee command is in ./node_modules/.bin/coffee. Is there a way to run this command when I'm in my project's main folder? I guess I'm looking for something similar to bundle exec in bundler. Basically, I'd like to specify a version of coffee-script that everyone involved with the project should use.
I know I can add the -g flag to install it globally so coffee works fine anywhere, but what if I wanted to have different versions of coffee per project?
UPDATE: As Seyeong Jeong points out in their answer below, since npm 5.2.0 you can use npx [command], which is more convenient.
OLD ANSWER for versions before 5.2.0:
The problem with putting
./node_modules/.bin
into your PATH is that it only works when your current working directory is the root of your project directory structure (i.e. the location of node_modules)
Independent of what your working directory is, you can get the path of locally installed binaries with
npm bin
To execute a locally installed coffee binary independent of where you are in the project directory hierarchy you can use this bash construct
PATH=$(npm bin):$PATH coffee
I aliased this to npm-exec
alias npm-exec='PATH=$(npm bin):$PATH'
So, now I can
npm-exec coffee
to run the correct copy of coffee no matter of where I am
$ pwd
/Users/regular/project1
$ npm-exec which coffee
/Users/regular/project1/node_modules/.bin/coffee
$ cd lib/
$ npm-exec which coffee
/Users/regular/project1/node_modules/.bin/coffee
$ cd ~/project2
$ npm-exec which coffee
/Users/regular/project2/node_modules/.bin/coffee
You don't have to manipulate $PATH anymore!
From npm#5.2.0, npm ships with npx package which lets you run commands from a local node_modules/.bin or from a central cache.
Simply run:
$ npx [options] <command>[#version] [command-arg]...
By default, npx will check whether <command> exists in $PATH, or in the local project binaries, and execute that.
Calling npx <command> when <command> isn't already in your $PATH will automatically install a package with that name from the NPM registry for you, and invoke it. When it's done, the installed package won’t be anywhere in your globals, so you won’t have to worry about pollution in the long-term. You can prevent this behaviour by providing --no-install option.
For npm < 5.2.0, you can install npx package manually by running the following command:
$ npm install -g npx
Use the npm bin command to get the node modules /bin directory of your project
$ $(npm bin)/<binary-name> [args]
e.g.
$ $(npm bin)/bower install
Use npm run[-script] <script name>
After using npm to install the bin package to your local ./node_modules directory, modify package.json to add <script name> like this:
$ npm install --save learnyounode
$ edit packages.json
>>> in packages.json
...
"scripts": {
"test": "echo \"Error: no test specified\" && exit 1",
"learnyounode": "learnyounode"
},
...
$ npm run learnyounode
It would be nice if npm install had a --add-script option or something or if npm run would work without adding to the scripts block.
update: If you're on the recent npm (version >5.2)
You can use:
npx <command>
npx looks for command in .bin directory of your node_modules
old answer:
For Windows
Store the following in a file called npm-exec.bat and add it to your %PATH%
#echo off
set cmd="npm bin"
FOR /F "tokens=*" %%i IN (' %cmd% ') DO SET modules=%%i
"%modules%"\%*
Usage
Then you can use it like
npm-exec <command> <arg0> <arg1> ...
For example
To execute wdio installed in local node_modules directory, do:
npm-exec wdio wdio.conf.js
i.e. it will run .\node_modules\.bin\wdio wdio.conf.js
Update: I no longer recommend this method, both for the mentioned security reasons and not the least the newer npm bin command. Original answer below:
As you have found out, any locally installed binaries are in ./node_modules/.bin. In order to always run binaries in this directory rather than globally available binaries, if present, I suggest you put ./node_modules/.bin first in your path:
export PATH="./node_modules/.bin:$PATH"
If you put this in your ~/.profile, coffee will always be ./node_modules/.bin/coffee if available, otherwise /usr/local/bin/coffee (or whatever prefix you are installing node modules under).
Use npm-run.
From the readme:
npm-run
Find & run local executables from node_modules
Any executable available to an npm lifecycle script is available to npm-run.
Usage
$ npm install mocha # mocha installed in ./node_modules
$ npm-run mocha test/* # uses locally installed mocha executable
Installation
$ npm install -g npm-run
TL;DR: Use npm exec with npm#>=7.
The npx command which was mentioned in other answers has been completely rewritten in npm#7 which ships by default with node#15 and can be installed on node#>=10. The implementation is now equal to the newly introduced npm exec command, which is similar but not equal to the previous npx command implementation.
One difference is e.g. that it always interactively asks if a dependency should be downloaded when it is not already installed (can also be overwritten with the params --yes or --no).
Here is an example for npm exec. The double dashes (--) separates the npm exec params from the actual command params:
npm exec --no -- jest --coverage
See also the updated, official documentation to npm exec.
If you want to keep npm, then npx should do what you need.
If switching to yarn (a npm replacement by facebook) is an option for you, then you can call:
yarn yourCmd
scripts inside the package.json will take precedence, if none is found it will look inside the ./node_modules/.bin/ folder.
It also outputs what it ran:
$ yarn tsc
yarn tsc v0.27.5
$ "/home/philipp/rate-pipeline/node_modules/.bin/tsc"
So you don't have to setup scripts for each command in your package.json.
If you had a script defined at .scripts inside your package.json:
"tsc": "tsc" // each command defined in the scripts will be executed from `./node_modules/.bin/` first
yarn tsc would be equivalent to yarn run tsc or npm run tsc:
yarn tsc
yarn tsc v0.27.5
$ tsc
The PATH solution has the issue that if $(npm bin) is placed in your .profile/.bashrc/etc it is evaluated once and is forever set to whichever directory the path was first evaluated in. If instead you modify the current path then every time you run the script your path will grow.
To get around these issues, I create a function and used that. It doesn't modify your environment and is simple to use:
function npm-exec {
$(npm bin)/$#
}
This can then be used like this without making any changes to your environment:
npm-exec r.js <args>
I prefer to not rely on shell aliases or another package.
Adding a simple line to scripts section of your package.json, you can run local npm commands like
npm run webpack
package.json
{
"scripts": {
"test": "echo \"Error: no test specified\" && exit 1",
"webpack": "webpack"
},
"devDependencies": {
"webpack": "^4.1.1",
"webpack-cli": "^2.0.11"
}
}
If you want your PATH variable to correctly update based on your current working directory, add this to the end of your .bashrc-equivalent (or after anything that defines PATH):
__OLD_PATH=$PATH
function updatePATHForNPM() {
export PATH=$(npm bin):$__OLD_PATH
}
function node-mode() {
PROMPT_COMMAND=updatePATHForNPM
}
function node-mode-off() {
unset PROMPT_COMMAND
PATH=$__OLD_PATH
}
# Uncomment to enable node-mode by default:
# node-mode
This may add a short delay every time the bash prompt gets rendered (depending on the size of your project, most likely), so it's disabled by default.
You can enable and disable it within your terminal by running node-mode and node-mode-off, respectively.
I've always used the same approach as #guneysus to solve this problem, which is creating a script in the package.json file and use it running npm run script-name.
However, in the recent months I've been using npx and I love it.
For example, I downloaded an Angular project and I didn't want to install the Angular CLI globally. So, with npx installed, instead of using the global angular cli command (if I had installed it) like this:
ng serve
I can do this from the console:
npx ng serve
Here's an article I wrote about NPX and that goes deeper into it.
Same #regular 's accepted solution, but Fish shell flavour
if not contains (npm bin) $PATH
set PATH (npm bin) $PATH
end
zxc is like "bundle exec" for nodejs. It is similar to using PATH=$(npm bin):$PATH:
$ npm install -g zxc
$ npm install gulp
$ zxc which gulp
/home/nathan/code/project1/node_modules/.bin/gulp
You can also use direnv and change the $PATH variable only in your working folder.
$ cat .envrc
> export PATH=$(npm bin):$PATH
Add this script to your .bashrc. Then you can call coffee or anyhting locally. This is handy for your laptop, but don't use it on your server.
DEFAULT_PATH=$PATH;
add_local_node_modules_to_path(){
NODE_MODULES='./node_modules/.bin';
if [ -d $NODE_MODULES ]; then
PATH=$DEFAULT_PATH:$NODE_MODULES;
else
PATH=$DEFAULT_PATH;
fi
}
cd () {
builtin cd "$#";
add_local_node_modules_to_path;
}
add_local_node_modules_to_path;
note: this script makes aliase of cd command, and after each call of cd it checks node_modules/.bin and add it to your $PATH.
note2: you can change the third line to NODE_MODULES=$(npm bin);. But that would make cd command too slow.
For Windows use this:
/* cmd into "node_modules" folder */
"%CD%\.bin\grunt" --version
I encountered the same problem and I don't particularly like using aliases (as regular's suggested), and if you don't like them too then here's another workaround that I use, you first have to create a tiny executable bash script, say setenv.sh:
#!/bin/sh
# Add your local node_modules bin to the path
export PATH="$(npm bin):$PATH"
# execute the rest of the command
exec "$#"
and then you can then use any executables in your local /bin using this command:
./setenv.sh <command>
./setenv.sh 6to5-node server.js
./setenv.sh grunt
If you're using scripts in package.json then:
...,
scripts: {
'start': './setenv.sh <command>'
}
I'd love to know if this is an insecure/bad idea, but after thinking about it a bit I don't see an issue here:
Modifying Linus's insecure solution to add it to the end, using npm bin to find the directory, and making the script only call npm bin when a package.json is present in a parent (for speed), this is what I came up with for zsh:
find-up () {
path=$(pwd)
while [[ "$path" != "" && ! -e "$path/$1" ]]; do
path=${path%/*}
done
echo "$path"
}
precmd() {
if [ "$(find-up package.json)" != "" ]; then
new_bin=$(npm bin)
if [ "$NODE_MODULES_PATH" != "$new_bin" ]; then
export PATH=${PATH%:$NODE_MODULES_PATH}:$new_bin
export NODE_MODULES_PATH=$new_bin
fi
else
if [ "$NODE_MODULES_PATH" != "" ]; then
export PATH=${PATH%:$NODE_MODULES_PATH}
export NODE_MODULES_PATH=""
fi
fi
}
For bash, instead of using the precmd hook, you can use the $PROMPT_COMMAND variable (I haven't tested this but you get the idea):
__add-node-to-path() {
if [ "$(find-up package.json)" != "" ]; then
new_bin=$(npm bin)
if [ "$NODE_MODULES_PATH" != "$new_bin" ]; then
export PATH=${PATH%:$NODE_MODULES_PATH}:$new_bin
export NODE_MODULES_PATH=$new_bin
fi
else
if [ "$NODE_MODULES_PATH" != "" ]; then
export PATH=${PATH%:$NODE_MODULES_PATH}
export NODE_MODULES_PATH=""
fi
fi
}
export PROMPT_COMMAND="__add-node-to-path"
I am a Windows user and this is what worked for me:
// First set some variable - i.e. replace is with "xo"
D:\project\root> set xo="./node_modules/.bin/"
// Next, work with it
D:\project\root> %xo%/bower install
Good Luck.
In case you are using fish shell and do not want to add to $path for security reason. We can add the below function to run local node executables.
### run executables in node_module/.bin directory
function n
set -l npmbin (npm bin)
set -l argvCount (count $argv)
switch $argvCount
case 0
echo please specify the local node executable as 1st argument
case 1
# for one argument, we can eval directly
eval $npmbin/$argv
case '*'
set --local executable $argv[1]
# for 2 or more arguments we cannot append directly after the $npmbin/ since the fish will apply each array element after the the start string: $npmbin/arg1 $npmbin/arg2...
# This is just how fish interoperate array.
set --erase argv[1]
eval $npmbin/$executable $argv
end
end
Now you can run thing like:
n coffee
or more arguments like:
n browser-sync --version
Note, if you are bash user, then #Bob9630 answers is the way to go by leveraging bash's $#, which is not available in fishshell.
I propose a new solution I have developed (05/2021)
You can use lpx https://www.npmjs.com/package/lpx to
run a binary found in the local node_modules/.bin folder
run a binary found in the node_modules/.bin of a workspace root from anywhere in the workspace
lpx does not download any package if the binary is not found locally (ie not like npx)
Example :
lpx tsc -b -w will run tsc -b -w with the local typescript package
Include coffee-script in package.json with the specific version required in each project, typically like this:
"dependencies":{
"coffee-script": ">= 1.2.0"
Then run npm install to install dependencies in each project. This will install the specified version of coffee-script which will be accessible locally to each project.