• Philip Trettner <philip trettner.de> Jul 27 2010
• Alix Pexton <alix.DOT.pexton gmail.DOT.com> Jul 28 2010
• Jerry Quinn <jlquinn optonline.net> Jul 28 2010

Philip Trettner <philip trettner.de> writes:

Hello.

When I came across the lexical definition of a FloatLiteral I wondered
how '1f' gets recognized.

The definition says:
FloatLiteral:
	Float
	Float Suffix
	Integer ImaginarySuffix
	Integer FloatSuffix ImaginarySuffix
	Integer RealSuffix ImaginarySuffix

Float contains a dot or an exponent, so this cannot be.
The suffixes for integers are only imaginary, so '1f' won't be
recognized either.

But the D Compiler compiles fine (as it should be).

Am I missing something or is this a flaw in the FloatLiteral
definition?

kind regards, Philip

Alix Pexton <alix.DOT.pexton gmail.DOT.com> writes:

On 27/07/2010 21:46, Philip Trettner wrote:
 Hello.

 When I came across the lexical definition of a FloatLiteral I wondered
 how '1f' gets recognized.

 The definition says:
 FloatLiteral:
 	Float
 	Float Suffix
 	Integer ImaginarySuffix
 	Integer FloatSuffix ImaginarySuffix
 	Integer RealSuffix ImaginarySuffix

 Float contains a dot or an exponent, so this cannot be.
 The suffixes for integers are only imaginary, so '1f' won't be
 recognized either.

 But the D Compiler compiles fine (as it should be).

 Am I missing something or is this a flaw in the FloatLiteral
 definition?

 kind regards, Philip


This is one of many discrepancies between the docs and the 
implementation. I'm still working on writing productions that match the 
code for my visual grammar project. The code isn't that hard to read if 
you ever get tempted to take a look, but be wary, it's a little spread 
out, not all cases are handled in the same place.

A...

Jerry Quinn <jlquinn optonline.net> writes:

 On 27/07/2010 21:46, Philip Trettner wrote:
 Hello.

 When I came across the lexical definition of a FloatLiteral I wondered
 how '1f' gets recognized.

 The definition says:
 FloatLiteral:
 	Float
 	Float Suffix
 	Integer ImaginarySuffix
 	Integer FloatSuffix ImaginarySuffix
 	Integer RealSuffix ImaginarySuffix

 Float contains a dot or an exponent, so this cannot be.
 The suffixes for integers are only imaginary, so '1f' won't be
 recognized either.

 But the D Compiler compiles fine (as it should be).

 Am I missing something or is this a flaw in the FloatLiteral
 definition?

 kind regards, Philip




This was reported in bug 949:

http://d.puremagic.com/issues/show_bug.cgi?id=949