Web Development

Skip to slide 20.

That last part - optimistic move application with what games people sometimes call “rollback” - is about 1,600 lines of code that took me a ~7 days of fulltime work to write. I don’t remember the last time I wrestled with a problem that hard!

🚀 Built my own toolbox of 10 free, browser-based dev tools – no sign-up, no install, no distractions.

🧰 Tools included: • JSON Formatter & Cleaner
• Password Generator
• Text Case Converter
• Regex Tester
• HTML Entities Encoder/Decoder
• URL & Base64 Encoder/Decoder
• Lorem Ipsum Generator
• Text Cleaner
• Hex to RGB Converter

🌐 Try them here: https://www.jscripted.com/devbox
Let me know if there's a tool you want added!

Developer Information / Release Notes

Hey all, I've been working in web agencies for years now. Not using site builders (obviously I've had to use them at times but I don't love it), building custom themes from scratch with PHP/HTML/JS etc., custom plugins, interesting client needs, some Shopify development, design work, all the standard digital agency drudgery.

I'm starting to get pretty disillusioned and would like a new challenge (and a salary increase). I'm currently in a pretty well-paid agency, earning around $70,000 a year.

I've worked briefly with React, Angular, GraphQL, Vue, but not in a full-time capacity. Is React/Node the path I should be looking towards if I want to make more cash and have a bit more security? I'm safe in my current role, but I'm bored and frustrated.

Also.. is there any way to be in this industry and not be constantly pestered about AI? I'm honestly sick to death of hearing about ChatGPT or Claude, lol

Thanks!

https://somework.atlassian.net/images/errors.png

Why save bandwidth if you can download a skull 5 times, the number 5 twice, the 4 even three times??

https://www.w3schools.com/css/css_image_sprites.asp

I made something to try out for "funtional webcomponents" with vanillaJs. I'm working towards a UI framework for my personal projects. It's far from finished but i thought it might be an interesting concept to share.

Specifically: Because the .mu registry (TLD operator) is a ccTLD, it can re-price any domain at any time.

This is a follow up to: https://programming.dev/post/27587527.

Conclusion:
Carefully read the terms and conditions if you choose to register a ccTLD. They are not bound by the ICANN Registry Agreement.

Gandi contacted the .mu registry about the price increase, and the registry has assigned premium pricing to all 2-character and 3-character domains regardless of whether they're premium or not. This would not be allowed if .mu was a gTLD. The .mu registry will not be grandfathering any domains registered and continually renewed since before premium domains were invented. I have reached out to the .mu TLD registry directly, and I am waiting to see if my personal plea will sway them. If they don't change their minds, I'll lose this domain on Tuesday.

Timeline:
I went to renew my domain about two weeks before it expiring, and discovered the renewal price went from $170 to $5,600 (or 32 years price for just one year). Usually, I do this much sooner, but my life has been crazy. Even three weeks ago I was barely functioning because of a sinus infection.

I contacted Gandi, my registrar, and at first they were less than helpful. Eventually, they did escalate my ticket to their legal department who contacted the .mu registry.

Gandi informed me that the .mu registry reclassified all 2-character and 3-character domains as premium. Gandi was not notified of the change, so that explains why I was not notified. Gandi said they do not agree with the reclassification of these domains—including mine. But, unfortunately, it's not up to Gandi. They requested that the .mu registry reconsider but have had no reply.

I then emailed ICANN about this because I was under the impression that the Registry Agreement governed all registries, and the reclassification is clearly a breach of the agreement. Unfortunately, ICANN let me know that this agreement doesn't apply to ccTLDs.

I have now sent an email directly to the .mu registry, hoping that a personal plea may make a difference.

I am not optimistic. In the end, I will probably lose the domain. I will make sure to complain about .mu domains as much as possible if I lose it. I will also be weary of registering any ccTLDs in the future. I only have a couple.

It's been a wild ride. Many points where I was almost hopeful, but it's probably just another case of the little guy getting screwed, and the big guy too rich to care.

This may not be a appropriate post for this community, and is not strictly web development. It is browser related, specifically qutebrowser, but i don't think the behaviour is qutebrowser specific, it may be chromium specific though, but i don't know how to test on any other browser because i hardly know js. If this is not valid for this community, please do tell.

I was writing a way to redirect other lemmy instance links to my home instance. I am basically trying to write lemmyverse.link (it's working is explained in the following issue, but essentially we just ask the instance where we want to be redirected, if they have the post, if so, they return json, which will have post_id in that instance, and we can use that.

https://github.com/RikudouSage/lemmyverse.link/issues/14

I think I have implemented that somewhat successfully, and it works for all the instances that i have tried, except lemmy.world (why could it not be some tiny instance which i could ignore). For lemmy.world, browser ui shows as if it is loading. I checked dev tools, and we do get the redirection (see the post_id change below in the attched video, and when i click that, i get redirected to correct page). It shows a red errors, but i can not find how to find errors (none of tabs show any errors).

https://envs.sh/s/NhX_dHEZh52CD1MwcXXLwg/cSI.mkv

(the video show succesful redirection for some instance, and unsucceful redirection for other)

I hope someone who knows how to do this help, I could not find anything wrong, and hence am not able to even search for errors online.

code for anyone wanting to see it

import qutebrowser.api.interceptor
import re
import requests
import sys
def match(patterns, url):
    for pattern in patterns:
        if re.match(pattern, url):
            return True
    return False
with open('/home/sg/data/dev/program-data/lemmings_coookie_jwt', 'r') as f:
    cookie = f.read().strip()
def resolve_lemmings_object(query):
    url = "https://lemmings.world/api/v3/resolve_object"
    params = {"q": query}  # Pass the query as a parameter
    headers = { "Cookie": cookie}
    try:
        response = requests.get(url, params=params, headers=headers)
        response.raise_for_status()  # Raise HTTPError for bad responses (4xx or 5xx)
        return response.json()['post']['post']['id']
    except requests.exceptions.RequestException as e:
        print(f"Error: {e}")
        return None

def per_website_rules(request: qutebrowser.api.interceptor.Request):
	lemmy_patterns = [ r"crazypeople\.online", r"discuss\.online", r"discuss\.tchncs\.de", r"feddit\.nl", r"feddit\.org", r"feddit\.uk", r"jlai\.lu", r"lemm\.ee", r"lemmy\.blahaj\.zone", r"lemmy\.ca", r"lemmy\.cafe", r"lemmy\.dbzer0\.com", r"lemmy\.ml", r"lemmy\.one", r"lemmy\.run", r"lemmy\.selfhostcat\.com", r"lemmy\.wtf", r"lemmy\.zip", r"mander\.xyz", r"ponder\.cat", r"programming\.dev", r"sh\.itjust\.works", r"slrpnk\.net", r"sopuli\.xyz", r"toast\.ooo", r"lemmy\.world", r"spaffel\.social" ]
	post_pattern = [ r"^/post/.*$" ]
	if match(lemmy_patterns, request.request_url.host()) and match(post_pattern, request.request_url.path()):
		url = request.request_url.scheme() + "://" + request.request_url.host() + request.request_url.path()
		# post_id_in_my_prefered_instance = os.popen("lemmy-redirect " + url).read()
		post_id_in_my_prefered_instance = resolve_lemmings_object(url)
		request.request_url.setHost('lemmings.world')
		request.request_url.setPath('/post/' + str(post_id_in_my_prefered_instance))
		try:
			request.redirect(request.request_url)
		except:
			pass

What a gorgeous demo page!

Biome is a formatter and linter for JavaScript, TypeScript and other web languages.

With this partnership, we aim to develop TypeScript-compatible type inference that works out of the box for use in our lint rules.