Use of %lf and %le

[WebDrake]

Hello all,

A bit of a dumb question here but I'll ask anyway.

What is the difference between %f and %e, and their counterparts %lf and %le?

I note that if used in printf it makes no difference, at least with the compiler I am using---the same on-screen precision is given.

However, I seem to recall code not compiling when anything but %le was used in a scanf function.

Can someone explain?

Best wishes,

-- Joe

 

[Diancecht]

They should be almost the same, but have slightly differences:

http://beej.us/guide/bgc/output/html/scanf.html

http://www.cs.utah.edu/~zachary/isp/tutorials/io/io.html

Cheers,
Dian

 

[ArkM]

You MUST make a differences between long (with l prefix) and ordinal floating point format specifiers in scanf(). Long forms must correspond with double and ordinal - with float arguments. You may have serious troubles if (most of) the target computer has different double and float internal (binary) representation. The scanf (and printf) family functions do not know its arguments real types (remember prototype int scanf(const char*,...);).

 

[WebDrake]

Aaahhh, that makes sense. Thanks very much.

I've always used %le for scanning doubles, but couldn't bring to mind why...