that's not much better than `BigDecimal(0.5)`

# language-proposals

[jw]

that's not much better than BigDecimal(0.5)

[jlleitschuh]

It removes the requirement to explicitly import it. ¯\_(ツ)_/¯

[voddan]

In what way does it affect imports?

[raulraja]

@voddan you would be able to get the conversion to BigDecimal without explicitly importing it to use its constructor.

[voddan]

Ether way you have to import a function, be it BigDecimal() or toBigDecimal(), no?

[jlleitschuh]

Not if it's in the stdlib

[raulraja]

I think in Scala there were some concerns with BigDecimal at some point regarding rounding precision from Double but I can't recall what the reasons were

[voddan]

you are right, they are in the auto-importing package

[jlleitschuh]

If you have 0.8 in big decimal it becomes 0.8000000000000000000523... ect

[raulraja]

https://github.com/scala/scala/blob/v2.12.1/src/library/scala/math/BigDecimal.scala#L359

[karelpeeters]

@jlleitschuh Technically it already was .800..00523, BigDecimal just has a different toString.

[karelpeeters]

The documentation specifically says it creates a new bigint with the exact double value.

[jlleitschuh]

?! Can you explain that? Also, does big decimal have some sort of "from string" method? I couldn't find one.

[benleggiero]

The one advantage is you can do nullable?.toBigDecimal() but not BigDecimal(nullable)

[karelpeeters]

@jlleitschuh It has a constructor for it: https://docs.oracle.com/javase/7/docs/api/java/math/BigDecimal.html#BigDecimal(java.lang.String,%20java.math.MathContext)

[jlleitschuh]

Oh, thanks for finding that.

[karelpeeters]

As for the 0.3 thing, you're in for a ride ☺️. This page seems to explain it pretty well:

[karelpeeters]

http://floating-point-gui.de/