Programming

:blush: - react newb and yesterday was the first time I really looked at React Router.

I deduce there was a request to remove location info for a user. Not that I was seriously accused, but just putting it out there for my rep that I didn’t put the sn together with the general location.

1 Like

I asked for it to be removed, it’s on me for commenting about it, all is well, microbet is cool and his project is cool too.

React question

Say you need data from a db and you’re going to present different parts of this data in different ways as people click on your page. You could grab this piece or that as you need it or you could just grab a bunch and put it in a slightly complicated object and keep that on the React side (in state) and use the parts you need as you need them.

Like, how much data would you want to do this with?

Seems like you’d want to do this with even what seems like a lot of text data. Is that pretty standard?

I see this: reactjs - What is considered "Too much data" in react state - Stack Overflow and the recommendation seems to be “don’t store 100000 data rows in the user’s RAM”, but what about 1000 or 10000?

Other similar SO questions are answered with “test it before optimizing”, but this seems like maybe people who do this a lot have some kind of general rule of thumb.

1000 is fine. 10k probably fine unless it’s an android.

1 Like

5000 absolute max for react, 1000-2000 max for best dom performance.

1 Like

Or what Grue said. I wasn’t thinking react-specific.

So, I have a react application I’m working on that allows the user to search for companies that file certain kinds of SEC reports. As it’s configured now, the application has a big (~1MB) data object with about 13k entries that it downloads upfront and uses to do a search preview as you type. I think that’s not state because it’s constant, so is that OK? But it would be bad if my app stored a state object for “companies starting with the letter _”, that then would be unwise?

All that matters is row count in the browser. Loading 13k rows via ajax is a bad idea.

What’s the appropriate way to do a responsive search preview? Does it help to structure the data in some way (a tree with 13k leaves)? Or is it just a bad idea if you have that much data?

We have a pretty complicated react table that loads 1500 rows for our biggest donor. It sorts, paginates (all client side - the entire object is loaded up front), filters, etc. no problem - even on Android phones. I was surprised.

hmm a “search preview” is just a search right? Again its really just a row count thing, doesn’t matter what they look like, 13k dom elements of any kind is ~5x too much. Limit it on the server side.

Long shot but does anyone have significant experience with requestAnimationFrame? I’m trying to make a game but am not sure its going to work at its core as I think it may get desynced between clients (websockets). But will probably just do it and see what happens when I get it in front of users.

So I know this is a weird place to ask this… but in a hypothetical world where I was thinking about becoming a freelance penetration tester where would I start? (I’m totally happy with being a freight broker, but it never hurts to diversify and I feel like having this option at least explored is good from both the ‘what if my industry blows up for some reason’ and a ‘world goes completely to shit’ angles.)

I have access to a really strong community college system here in Austin that has a 2 year associates in cyber security + I’m god tier on the phone… which means I can probably do the social engineering part at an expert level.

What do I need to know and where should I start?

Kinda makes sense because the compiler can check to make sure the string literal exists as a property - but not the variable at run time.

Maybe typescript has some kind of enum-like thing (a list of literals) that you can use so the compiler knows it’s getting a valid property name.

Edit: ponied

@jmakin

I’m going to the OC Python meetup tonight.

Speaking of python I’ve been trying to install my work’s back end locally for about a week and it fails on SSL on pip install wtf who thought this through?

It’s culty and you need to trust pypi.org

Um what do I do to make this work :expressionless:

Meh? But I googled and got told you should set "–trusted-host pypi.org " or something like that, but I assume you googled it and saw that stuff too.