Git commit error after using Git Husky and esLint - eslint

It can be submitted normally without configuration using Husky. After using Hushy, an error will be reported directly when committing.enter image description here
[1]: https://i.stack.imgur.com/PB1O3.png

There is something wrong with executing the pre commit hook. Which is located in .git/hooks/pre-commit*
Since the second line of your screenshot says 'node' is not recognized as an internal or external command you should first try to add node to your PATH variable.

Related

npm: command not found in jenkins for MAC OS

I tried to implement CI/CD pipeline by first cloning a react project from git and then build the project in my local machine.
I have reinstalled node, configured nodejs in jenkins and tried adding PATH but nothing worked for me..
Method one that i tried: I create a freestyle item and added below script after providing the git repo link under pipeline script from SCM.. this is throwing me npm command not found error.
export PATH=/opt/homebrew
npm install
Edit:
I restarted jenkins and lost my old build..now, after restarting jenkins and running the build(above method 1 steps) i see groovy.lang.MissingPropertyException: No such property: install for class: groovy.lang.Binding
Second method i tried is by creating a new pipeline item and passing pipeline script as below.
pipeline{
agent any
tools {nodejs "default"}
stages{
stage('Build'){
steps{
git 'git#github.com:patebija/simple-node-js-react-npm-app.git'
sh 'npm install'
}
}
}
}
Edit:
After restarting jenkins, when i run the above steps(method 2), i still see "npm: command not found"(Not sure if **sh 'npm install'** is valid in MACOS.. And if i just give **npm install** removing sh, i get exception -- groovy.lang.MissingPropertyException: No such property: install for class: groovy.lang.Binding .
I am new to jenkins and stuck with this for a long time...Appreciate any help in this regard.
Thank you
export PATH=/opt/homebrew would override all the paths stored in $PATH, to replace it with that unique one.
That would not work very well, since all system commands would no longer be referenced.
Usually, you would add the npm path to the existing PATH
export PATH="/usr/local/bin/npm:/usr/local/bin/node:/usr/local/bin:$PATH"
(Plus possibly your existing npm packages path)
Make sure your Jenkins main controller and agents are running with your account (not as root), in order for them to inherit your environment variables.
So check first npm or git are working properly from your shell session, then launch Jenkins, and try executing pipelines.

Node,exe not found despite correct environment variables CI build windows

I'm dealing with a strange issue. When yarn install is ran during my automated CI build, I get the following error:
error C:\GitLab-Runner\builds\6sT6aNYk\0\[redacted]\node_modules\fibers: Command failed.
Exit code: 1
Command: node build.js || nodejs build.js
Arguments:
Directory: C:\GitLab-Runner\builds\6sT6aNYk\0\[redacted]\node_modules\fibers
Output:
'node' is not recognized as an internal or external command,
operable program or batch file.
'nodejs' is not recognized as an internal or external command,
operable program or batch file.
My environment variables are correct. I've manually browsed to the directory, and executed the same command using powershell. Then the command does not fail, and node is found in its installation directory. Powershell is the runner configured for this system/project.
During the CI build I can See node.exe spawning under gitlabrunner.exe. So the path for the process is correct. It's just yarn install that fails to find node.exe.
I'm not a javascript developer, and I do not know how to dive deeper into what is happening. How can I see what causes this problem? Has anyone experienced something like this before?
Other things I've looked at:
path length disabled
Building from other directories in the system (manually, it all works)
Removing node & npm and re-installing
Edit:
The issue also occurs if I use npm install instead of yarn install.
I found the problem. Another item in my build chain was adding items to the path in a loop. previously this was not a problem, but maybe an update added more items.
The solution was to clear the environment and add the original path back.

cucumber-js parse error when run on Jenkins

I am trying to setup a jenkins pipeline step to runs some test scenarios using cucumber-js but I am getting an error back from the build as follows:
Error: Parse error in 'e2e/definitions/login.js': (1:1): expected:
#EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got 'const { Given, When, Then } = require('cucumber');'
The command being run in the pipeline step is as follows:
cucumber-js e2e/features/**/*.feature --require e2e/**/*.js
The opening lines of the login.js file the error is referencing are:
const { Given, When, Then } = require('cucumber');
const { Selector } = require('testcafe');
I'm wondering if this has something to do with nodejs version differences, as I am running 8.11.2 on my machine and dont see these errors, on Jenkins we are running 10.5.0
Does anyone know what the problem could be and point me in the right direction please?
Thanks
Likely you have this problem because the glob pattern specified after the --require pattern isn't resolved to real file names, but on your Jenkins it does. Try to wrap e2e/**/*.js in double quotes:
cucumber-js e2e/features/**/*.feature --require "e2e/**/*.js"
The error you're getting is a Gherkin parsing error, so I think cucumber is treating your step definition file as a Gherkin file (feature file). I would check which version of cucumber-js you're using locally versus the version that your using in CI. If the versions are different, your CI might be missing a bugfix or it might be using an different version of the CLI.
I also highly recommend setting up your local environment the same way as your CI (same version of node, pinned versions for your npm dependencies), it has saved me a lot of pain.

Error while running the command npm run dev

I am starting to do a new application for the first time and i installed git and node.js and set the path.. With the link from bitbucket i cloned the project then i gave npm install.. Upto this everything was fine but when i gave npm run dev, it was throwing this error:
I think the issue was related to path and i don't know what to change in it. My path looks as below:
The other questions in the forums doesn't solve my issue.
Try to remove the first ';' in your PATH system environment variable. Check that your PATH does not end with ';'
Check also that you don't have multiple value for ComSpec. (and no semicolon)

Heroku App is Not Using the Right Node Module

I'm trying to test a fork of a node module on my Heroku app. Here's what I did:
Forked a node module repo and made some changes to the code. Pushed to my own remote repo.
Ran the following on my Heroku bash terminal:
npm install git+https://git#github.com/Nsrose/node_model_updated.git
Ran heroku restart.
This actually updated the file I edited under node_modules/ folder. However, the error that was fixed with this file change is not changing. Before the npm install, the app said this error:
ERROR TypeError: Cannot read property 'channel' of undefined (line 97)
After the series of commands above, even though the file is updated on the heroku server under node_modules/, the error persists. In fact, the file I changed now doesn't even have anything related to 'channel' on line 97.
Why is my heroku app still using the old node_module/ and how do I force it to update?
To run node.js application in heroku you need a "Procfile" with no extension, there you write the command heroku has to run to get your app running, this file has to be your main git folder for example
web: node app.js
So after cloning the git repository, saving your changes and testing locally, all you have to do is:
git add .
git commit -m "message"
git push heroku master
I'll leave a link to Getting started with Node on Heroku
EDIT: Also your undefined is probably a code problem not heroku's

Resources