If a human can rewrite code from one language to do the same thing in another language, it's definitely possible to automate. But I acknowledge that translation will be very difficult to achieve. Saying something is impossible because it's very hard is ultimately self-limiting.
-
-
As
@o_guest once told me, there's no formal spec for Matlab: https://uk.mathworks.com/matlabcentral/answers/48745-where-is-the-syntax-for-matlab-rigourously-defined … Yacc and Bison aren't going to help anyone here. Plenty of things compile to JS, so one can avoid JS.pic.twitter.com/V2I9UTGbjM
-
Meanwhile, who wants to determine the epsilon above or below which Matlab refuses to use (wretch, heave) a *float* as an array index empirically? How many other such "features" lurk like mines beneath the surface? I wondered if even Octave had a grammar. -Nah.pic.twitter.com/xVCamkjbNX
-
And obviously the fact that it rounds the index after sufficient decimal points is idiotic too... I've had some similarly intractable nonsense happening too occasionally.
-
Yeah, I'd guess there is no explicit epsilon for that indexing check -- it's just using the limit of double accuracy to work out whether it's an integer. I assume this can produce some wacky behaviour for Large arrays.
-
In my case it has been the opposite where it claimed a variable was a certain integer, say 3, but x==3 would be false. Has produced some very annoying bugs...
- 14 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

