Fix: Strip integer part in "decimals" #22251
Conversation
|
Nice! One thing I'd like to mention is that the precision goes up to maxn-1, as I'm sure you know. To increase the decimal points it checks against, both maxn as well as the doubles within the sd-array have to be increased (i.e. 1.000000005 currently returns 0, but by increasing them by one, returns 9). |
aaronfranke
force-pushed
the
decimal-int-strip
branch
from
29ded92 to
a3323c8
Compare
|
@DeuxAlpha Note that when grabbing the value from a single-precision float around 1, it cuts off at 7 (or less, if the number is large). But yes, I've changed that to go up to 9 like it did before. |
aaronfranke
force-pushed
the
decimal-int-strip
branch
from
a3323c8 to
3188eb2
Compare
Note: Core only.
aaronfranke
force-pushed
the
decimal-int-strip
branch
from
3188eb2 to
5f4f9ca
Compare
|
Why not include the same fix in |
|
@akien-mga The current C# "Decimals" behaves as a decimal counter ("1.2345" returns 4) so I think whoever implemented that must've been confused of what the C++ implementation was. The GDscript and C++ behavior is not present in C# yet. I can't fix code which doesn't exist. This is why I added it to my other PR instead of making a new one, but I made one here since you requested it. If I changed the C# code in this PR, then it would be basically the same as the other PR except for lacking the adding of the "step_decimals" GDscript alias. May as well merge the other one. |
|
Thanks, looks good. |
CC @akien-mga Fixes #22216 See also #21425 Suggested milestone: 3.1.