39 comments

[ 2.7 ms ] story [ 86.2 ms ] thread
Is there a reason this url points exclusively at a second website (cleancss.com) for all the tools?
This made me think of Matt’s script archive for some reason.
Too bad these tools run on the server side, which means you can't study how they are implemented.

Also, the pages are filled with big banner advertisements.

I flipped shields on/off with Brave. What was pleasant soon became unpleasant.
(comment deleted)
The notion of pasting vast swaths of data into an online format converter gives me the most powerful kind of creeps.
Came here to say this. If this was a bunch of front end stuff that you could grab from Github then I would use it. Although for non-sensitive things it may be handy.
(comment deleted)
SnD Reverser Tool if you're looking for an oldschool, really tiny / lightweight, native (and also cool looking) application. Such tools are meant to be native.
Why are such tools meant to be native? Browser JS is pretty fast for certain tasks and it runs everywhere
I used this a lot to work with timestamps until I switched to Raycast and made an extension to do the same.
Slightly tangential - does anyone know if there is a running time parser?

Like:

Convert 60mi in 23hr15m23s to:

miles per minute

km per minute

mph

kph

etc., etc.

qalculate (qalc on the command line):

    $ qalc
    > 60mi / 23h15min23s to mph
      (60 × mile) / ((23 × hour) + (15 × minute) + (23 × second)) ≈ 2.579936218 mph
    > 60mi / 23h15min23s to km/min
      (60 × mile) / ((23 × hour) + (15 × minute) + (23 × second)) ≈ 0.06920008122 km/min
Cool! I should have clarified. km/min is actually how many min per 1 km, aka pace

https://www.depicus.com/swim-bike-run/pace-conversion-chart

Sure, you want

  23h15min23s / 60mi to min/km
You can give it a try here: https://flaviutamas.com/qalculate-wasm/

Although it's very experimental and this particular query doesn't work right now. I'll take a look soon

actually, it turns out it works fine on my desktop but not mobile. Not sure why, but wasm is a bit janky sometimes.
also, one more clarification. the result needs to be formatted into HH:MM:SS

trying to see if qalculate can do that as well.

This was kinda tricky to figure out, but here you go:

    > 60mi/23h15min23s × (min/km) to time
    0:04:09.1202924
There is (perhaps unsuprisingly) a lot of overlap between what these scripts do and what Easy Data Transform does. But Easy Data Transform gives you the option to chain multiple transforms together and it runs locally (rather than uploading your potentially sensitive data to someone else's server). However it is commercial software, rather than free.
I can do all of it and more in Excel and even run chain transformations. And I can do most of it with Notepad++ replace feature.
In theory, you can do it all in assembler. But it probably wouldn't be very convenient or a good use of your time.
In other words, small shell scripts put on server side. Stuff like echo $TEXT | wc -w.
totally off-topic: my god the ads. they were so triggering for me, maybe I'm just not in a good mood.
Speaking of which, what are people using to procedurally convert JSON to XML locally? Let's say for JSON API use by a language that provides an XML library.
If you’re converting to XML, you presumably wish to transform the data, in which case you should be using XPath through XSLT or XQuery; JSON conversion is now a built-in feature.
It's good to hear JSON conversion is built in. I've got a bit of experience with XPath. Thanks!
You might be interested in hearing that Saxon is available on four technology platforms:

* Java, with language bindings to other JVM-based languages such as Scala

* .NET, with language bindings to .NET-based languages such as C# and Visual Basic

* Native, with language bindings to C, C++, PHP, and Python

* JavaScript, running in all modern browsers and on Node.js

XPath is a powerful language in itself; you may not need to use XSLT or XQuery at all.

One curious thing about the What's My IP Address? page (the only one I looked at) -- it also detected 4 other devices on my network (192.168.xx.xx, 10.0.xx.xx, etc). But how?

Well, there's a list of 92 ip addresses (iplist -- it's a slightly weird list but I suppose it's based on common router and cable modem addresses). For each one, it runs an XmlHTTPRequest(). But none of the detected IP addresses are running web servers, you say. No matter. Most TCP stacks will send a TCP reset if there's nothing listening on the port (this can often be disabled with the proper incantation) which will result in an onerror callback and therefore there is a device If there is no device (or TCP reset is disabled) it will eventually result in the ontimeout callback. And of course, if there is a web server running, the success callback will trigger. So success or an failure indicates a device on your network.

I'm mildly surprised this is allowed by browsers, but I can't think of an easy way to prevent it, either.

(That said, were I the tool author, once I found any hit, I would then scan the whole /24 that the hit was in. But then I'd be here being more than mildly surprised.)

It is indeed a weird list[1].

[1] https://gist.github.com/wlonkly/b39f8fe37c8455bcc0bce6a7549a...