Came for a coin, and found a goldmine. I was just searching how to write a SIMPLE message while committing a code. Ended up learning about Conventional Commits 1.0.0 and SemVer and made my repo more professional. My project evaluator was IMPRESSED. A big thanks to you, Bro. Keep up the good work 👍🏻
I swear i was looking at some repos in github and I really loved and was curious to know what kind of structure they were following, this video is a gem explaining everything in simple terms and examples, very well done friend!
Best video I have seen on this!! Would love to get a follow up with a more complex example, and maybe some more guidance on documenting things in general and using issues in github.
Not the creator of the video, but I'll reply just in case someone sees this and wants to know as well. It is usually thought to be a good idea to write in imperative form because you are implicitly telling the machine what to do when the commit is added to your repo. So let's pretend that you are building a website, and the commits are a string of messages going directly to the machine that is hosting that website. "commit 1: added a button that is blue" "commit 2: translated header text from English to Latin" The computer, in this very contrived example, wouldn't be able to make sense of that list as opposed to: "commit 1: add a blue button in card component" "commit 2: translate header text to English" Now, this is the primary reason most blogs etc use for their examples, that the commit should be a _command_ so that you can easily read what exactly each commit does to the codebase once it is applied - and this is a legit reason to use it for many. However, two other good reasons that people have stated are; 1: It forces the developer to write a better, more specific commit message because it needs to be distilled into its purest form in order to be a command, and 2: It gives you a simple way to notice if you might have done _too much_ work for it to be one commit, and maybe it should be split into several, smaller? If the commands become something like "Add blue button, remove border on header logo, reduce complexity of state by colocating with related component, make the world a better place" you might realise how it's time to split that one into several, especially the last commit :-) Splitting a large pull request into many smaller ones, or at least into many smaller commits is a really good idea if you want to have an easy time with code reviews in your project. It makes it _way_ easier to read and understand each other's code that way. At the end of the day, there is no Council of Coders™ that will banish you from the land of binary trees and search algorithms just because you use past tense rather than imperative when writing commit messages. So long as the project you are in have an agreed upon way of writing commit messages so that it makes sense for the project and is easy to track the history if something goes awry, that's all that matters. It _is_ worth pointing out though, that Git themselves prefer to write commit messages using imperative form. Do with that what you will :-)
This is awesome, love the way you explain things. Now I'm going to be using these conventions. You are the best. What's your thought on using git client?
Liked this video until you started typing yarn commands and then I got lost. I thought this is a git commit message video. What is yarn? I guess I have to go away and figure that one out and them come back.
TOP was here hehe
Came for a coin, and found a goldmine.
I was just searching how to write a SIMPLE message while committing a code.
Ended up learning about Conventional Commits 1.0.0 and SemVer and made my repo more professional. My project evaluator was IMPRESSED.
A big thanks to you, Bro. Keep up the good work 👍🏻
Hi Bhai,
my commit messages are always trash, this was very informative.
I swear i was looking at some repos in github and I really loved and was curious to know what kind of structure they were following, this video is a gem explaining everything in simple terms and examples, very well done friend!
I have been avoiding to write good commit messages, because we are just 2 persons. This video motivates me to do it, thank you!
I'm so glad! It really helps my workflow
Thank you so much ! I have an internship now and thisis very helpfull !!!
holy shit, congrats dude!
Best video I have seen on this!!
Would love to get a follow up with a more complex example, and maybe some more guidance on documenting things in general and using issues in github.
My commit message was like "sadfjhkasfk" lol
Oh Lord ahahahahah
If one day your boss fires you, the reason on the official document will be "sadfjhkasfk" 🤣
@@DraxTheDestroyer 😂😂😂
git commit -m “idfk”
"made some changes"
Very helpful 😃 thanks
This channel content is precise and conscience 👍 keep it up bro.
Thank you for the video.
What's the commit name for fixing styling on frontend?
Thank you!
Amazing video
video was good, last section lacked context, watched it twice, still didn't understand it.
2:04 why should we use imperative for the description?
Not the creator of the video, but I'll reply just in case someone sees this and wants to know as well.
It is usually thought to be a good idea to write in imperative form because you are implicitly telling the machine what to do when the commit is added to your repo.
So let's pretend that you are building a website, and the commits are a string of messages going directly to the machine that is hosting that website.
"commit 1: added a button that is blue"
"commit 2: translated header text from English to Latin"
The computer, in this very contrived example, wouldn't be able to make sense of that list as opposed to:
"commit 1: add a blue button in card component"
"commit 2: translate header text to English"
Now, this is the primary reason most blogs etc use for their examples, that the commit should be a _command_ so that you can easily read what exactly each commit does to the codebase once it is applied - and this is a legit reason to use it for many.
However, two other good reasons that people have stated are;
1: It forces the developer to write a better, more specific commit message because it needs to be distilled into its purest form in order to be a command, and
2: It gives you a simple way to notice if you might have done _too much_ work for it to be one commit, and maybe it should be split into several, smaller? If the commands become something like "Add blue button, remove border on header logo, reduce complexity of state by colocating with related component, make the world a better place" you might realise how it's time to split that one into several, especially the last commit :-)
Splitting a large pull request into many smaller ones, or at least into many smaller commits is a really good idea if you want to have an easy time with code reviews in your project. It makes it _way_ easier to read and understand each other's code that way.
At the end of the day, there is no Council of Coders™ that will banish you from the land of binary trees and search algorithms just because you use past tense rather than imperative when writing commit messages. So long as the project you are in have an agreed upon way of writing commit messages so that it makes sense for the project and is easy to track the history if something goes awry, that's all that matters.
It _is_ worth pointing out though, that Git themselves prefer to write commit messages using imperative form. Do with that what you will :-)
How would you create a changelog for python or java?
Me: "Work in progress"
"Work in progress"
"Work in progress"
"Work in progress"
"Work in progress"
...
Najs👍👍
This is awesome, love the way you explain things. Now I'm going to be using these conventions.
You are the best. What's your thought on using git client?
I love them, I use GitKraken
Thanks man
sooooooooo usefull
Please Make simple tut series for Vue3 with Typescript and composite API
great
Liked this video until you started typing yarn commands and then I got lost. I thought this is a git commit message video. What is yarn? I guess I have to go away and figure that one out and them come back.
would've been nice to see you actually do this in the terminal instead of the slid show
The video was useful until min 3:00.
I don't get the need to go at warp speed on a "how-to" video.
my commit message was "changes" lmao
top.... yarn ?
A little too fast but good :)
yes, especially the last part is a little quick :d
Judging from the thumbnail I thought it was @fireship for a second😅