I just noticed today that fluid.dataset~
orders things in the same way that polybuffer~
does, with this kind of thing:
rows: 402 cols: 21
1 16.045 4.2655 3.7807 … -0.45747 92.665 -20.335
10 13.913 3.6963 2.764 … 0.49614 96.902 -30.718
100 3.9571 7.844 2.9822 … 0.43909 105.56 -83.162
…
97 7.2313 7.5109 4.7202 … 0.4376 96.902 -73.173
98 15.57 2.9553 2.1929 … 0.39056 73.063 -22.447
99 21.078 9.3573 4.5202 … -0.15598 96.902 -14.506
This is a dataset that was created with indices going from 1 to 402, but as you can see they are ordered in this way.
If you’re doing stuff for matching/querying it doesn’t so much matter the order of things, but if you are dumping out a json file, this gets really tedious to deal with.
So this is a bit of a feature request as well as kind of asking why it’s this sort of ordering and not how most OS-level things handle ordering.