Nothing new about lexicographic order versus natural sort order. However, for people like me who love "creeping featurism", the UI and UX could be improved. First, both lexicographic order and natural sort order are not absolute, they are relative to an underlying character order, itself relying on character grouping algorithm (it is not the same thing to say a byte is a character and to group bytes to have an utf8 character or an utf32LE character). So far, the UI is just an "arrow" upward or downward next to Name column or things like that: easy as pie, one or two clicks next to a column name and you have the chosen order. But if you copy what is available in ERP with web apps like an ERP done with Django or another framework, you can:
- sort on many columns, each column sort being one element of a bigger lexicographic order on the chosen columns,
- propose many distinct sorts for the same column: most UIs stop at "increasing/decreasing" order, but with a modal, or one or more selects, you can propose to chose your wanted sort with more than two possible values. For example, if dealing with strings, most UIs only propose increasing/decreasing natural sort order grouping with UTF8 characters (GNU/Linux) or UTF16 characters (Windows) and default system collation. But instead you could treat any string as a sequence of bytes and select your character grouping algorithm (including a cascade : try to group as UTF8, if fails, try as UTF32LE, etc., if all fails use each byte as a character (not an ASCII one, UTF8 would have worked, an ISO-8859 for example, see collation after) ... you can start laughing ;) XD I'm still serious but I also know it is quite funny :) ), then select your collation, then select your sort between lexicographic and natural.
This is usually not needed, but would give full control to the user. And I love "creeping featurism" and giving full control to the user :). Typically, if one day an AI can code a variant of Gnome and Nautilus in a matter of hours, then with this kind of knowledge, if you know how to ask, you could have such a complex and juicy UI that is functional in a matter of hours :). Maybe one day we will all have custom-made OSs instead of ready-made OSs, and people with knowledge and a taste for complexity will have options mere mortals never thought of :). To return on the current ground, distinct sorts is common in ERPs with "type" or "status" columns: imagine you have an ERP with a webpage for today deliveries, and in some office someone needs the list of today deliveries with those that are already delivered at the top, in another office someone else needs those that are still currently in delivery at the top, and again someone else needs those that have an anomaly at the top, 3 distinct orders at least. I made up this example, and it is simple enough to think that a filter could replace an order, since I only talked about the top. But I can tell you such things do exist in real world applications.
Hello,
I hope other people will find that it is a good idea.
Moreover, you can already use it from PyPI: https://pypi.org/project/python-none-objects/
Best regards,
Laurent Lyaudet
Just spotted a typo "mulitple" in case the author lurks in the comments here :).
("You'll pay it mulitple times over when you need to debug why your code doesn't work (either locally, in the test suite, or in production).")
You could add a puzzle mode with a fixed set of instructions introducing gameplay elements like rotate after a few puzzles to make it easier and more interesting to learn the game. My first play, I got a lot of sub instructions and my registers had all zero value. It was somehow partly pointless: Should I rotate and do sub R1, R2, R3 or do sub R2, R3, R4 to get a nop somewhere ? ;) XD
Yep not even a random display in JS. I looked at the source code of the webpage.
There is nothing curious. Custom JS is here: https://verdagon.dev/components/annotated.js and does only trivial things. Not a real easter egg for the client part. Maybe the easter egg is in server part that sends the HTML with or without it depending on hour or something. No need to waste more time on it, there is no funny code behind :)
SPOILER ALERT Am I the only one who thought
"What's the problem?
- The basic loop with xrange in Python tops at 68,000,000 (loop.py),
- but write_to_disk.py tops at 342,000,000,
- and write_to_memory.py tops at 2,000,000,000
" ?
And then thought "Ahah but it is while loops instead of iterators! Iterators are that bad for performance! XD"
Or 1 - (a/b) ;)
The rational behind my choice of a/b - 1 is that it corresponds exactly to what I was asked to display in an accounting report and use it to filter some lines.
I would be happy if you can elaborate why it makes more sense to you :)
