For List filenames without the diffs, I find that --stat is better than --name-only because it at least gives you a quick count of changes/deletions per file while still keeping the one-line-per-file format.

It's not correct to say that time.Sleep takes Nanoseconds. It takes a time.Duration as you can see from the signature. Go is strongly typed, so you can't pass an int in there at all.

The confusion is created by the way go handles untyped constants like integer-literals: They don't have a type and arbitrary precision and they get a type inferred at the place they are used. If you use it as an integer-type, like int, uint32, rune… or time.Duration, then that's the inferred type. That's how time.Sleep(42) can work - 42 is then inferred to be a time.Duration. However, that's more an unlucky consequence of the design, than intention. If no type can be inferred - for example when assigning it to a new variable, like x := 42 - a default type (int for integer literals) is used, which is why above link creates a compiler-error.

I acknowledge that the exact details are hard to get right for "outsiders" not familiar with the language, but I just thought I should report that that statement isn't correct, technically :) It's not that time.Sleep takes nanoseconds, it just that a time.Duration is stored as an integer type with nanosecond granularity, so that's how an integer literal in place of a Duration is interpreted.

For Move The Latest Commit To A New Branch, you can do it even more simply:

git branch my-new-branch
git reset --hard HEAD~

because git branch, unlike git checkout -b, doesn't actually take you to the newly created branch.

In sed/grab-the-first-line-of-a-file.md missing -n in the first example.

Hi,

It's a small issue in:
til / vim / Wrap With Some Room

This will not work with a vanilla vim.
It requires this plugin:
https://github.com/tpope/vim-surround

You should mention it.
Thanks!

Work around: https://github.com/davidbau/seedrandom/blob/released/seedrandom.js

Thanks for sharing this repo, fun stuff!

In css/dry-up-scss-with-mixins.md misspelled "navgation"...

I tried compiling all markdown files of every category in one file in every directory(category) locally with pandocs
I tried this in powershell

 foreach ($d in $dir){
$folderName = $d.Name;
   pandoc (get-item D:/til/til/$folderName/*.md) -o D:/til/til/$folderName/$folderName-compilation.md
}

this created a compilation file like ack-compilation.md in ack folder containing combined markdown file content,
Maybe someone can automate the process for this github repo as reading TILS together for any particular category is much better than clicking links for every TIL.
This is the result

The link in the README points to /turning-timing-on.md which is 404. The correct path is /turn-timing-on.md.

This file is empty on master. Looks like it may have been emptied in 38d2833.

The above is listed in the README.md, but the actual file is at vim/scrolling-relative-to-the-cursor.md.

Excellent tip by the way, I'll be using this all the time.

In
https://github.com/jbranchaud/til/blob/master/postgres/count-records-by-type.md

I prefer select type, count(*) from table group by 1 as is if the elements counted are something like select substring(unnaccent(upper('firstname',1,2))) .... there are less character to type at the end of the command :)

Simpler solution:
Array(9).fill().map((_, y) => y+1)

Or (Array(10)+"").replace(/,/g,(_,y)=>y+1)

@jbranchaud The line below is referencing vim/turning-off-search-highlighting.md

but the file in that folder is missing an f and is called vim/turning-of-search-highlighting.md

Suggested fix is to rename vim/turning-of-search-highlighting.md -> vim/turning-off-search-highlighting.md

I appreciate your guidance on integrating the wisdom

Small changes make great things

into my life

Hey! The issue you are seeing there with [0] is not an issue at all. The behaviour is completley normal if you embrace JS coercion (implicit casting) in == operator.

the issue is that in the expression [0] == false, false is actually coerced to a Number by the built-in object Number, and of course: Number(false) is 0!
Then, [0] is first coerced to the String "0" and finally coerced to the Number 0.
So, the last comparision is actually 0 === 0. Which is obviously true.

Hope the explanation shed some light on the subject!

For security reasons you probably shouldn't have a boilerplate to seed a random number generator with predictable/structured input like the current time (which can be guessed by an attacker). Use crypto/rand to seed your RNG.

It's also possible to recover the state of most PRNGs using the output, so you should usually just use crypto/rand directly for your random data until the RNG becomes your program's bottleneck.

Hey there,

Straight from signing up, one will have to look in the trash for the confirmation email. I'm guessing your newsletter will wind up there as well.

I don't think ConverKit is doing anything wrong...they're just sending a message from your email from some wonky-random-generated webpage.

I'm sure if you set these settings up the problem would go away (or not, because it's still gmail!).

This is how ProtonMail shows it, and the sneaker tracker.

Believe it or not, this is better than Microsoft 365 in Office, which just auto-trashes it.

The newsletter on your website from tinyletter works just fine :)

Maybe you're not even using the one in the README.md and problem solved with a small edit

The guide "List All The Say Voices" is filed under Unix; however, say is an OS X-only command, so it should probably be moved to the Mac folder.

Hi!
I love the idea and am adopting it myself. I've created a script which will auto-generate the README.md file using the information already in the directory structure and the titles of each of the TIL markdown files. I've got it working and producing output that is close: the diffs are:

• alphabetization is different when python does it than some of your choices
• you missed a # in front of Webpack
• Someplaces you have custom capitalization (CSS instead of Css)
• You also missed this line:

• SSH With A Specific Key

As I said, I've got the generation script, but I'm still thinking about how to get it into the work-flow. The options I can think of tonight are:

1. a git hook which runs on pre-commit which modifies the README.md
2. just have it as a command in the project that can be run manually

Neither of these are making me happy this evening, so I'm asking you if you have any ideas, that is, if you're interested at all.

Either way, Thanks for the great idea! I really like what you did. I'll admit I created my own before I saw your instructions (or thought about) forking your starter repo.

Thanks again!