I am creating my study guide, leading up to the java programmer exam, and I ran into this: floatmin_float = -1.40129846432481707e-45f , max_float = 3.40282346638528860e38f; compiles just fine. but this (note - no 'f' at end): floatmin_float = -1.40129846432481707e-45 , max_float = 3.40282346638528860e38; provokes this error: ValidPrimitives.java:22: Incompatible type for double. Explicit cast needed to convert double to float. float min_float = -1.40129846432481707e-45 ^ ValidPrimitives.java:23: Incompatible type for double. Explicit cast needed to convert double to float. , max_float = 3.40282346638528860e38; ^ I can't find anything about this in my reference books. Can someone enlighten me? regards, Bret
posted 15 years ago
Hi, Part I The Java Language Specification 3.10.2 Floating-Point Literals specifies that A floating-point literal is of type float if it is suffixed with an ASCII letter F or f; otherwise its type is double and it can optionally be suffixed with an ASCII letter D or d. thus the literal 2.3 is by default double and not a float . to specificy it as float one has to say 2.3f or 2.3F Part II A conversion from a double to byte, short, char, int, long, or float is a narrowing conversion where in data is definetly lost. As such one has to tell the compiler that it is "OK " by casting it into a respecive type. for example float f=1.2f; if one tries int i=f; //compiler complains one thus has to cast it like this int i=(int)f; check out the Java Language Specification for Conversions and Promotions 5.1.3 here Regds. Rahul.