String gap failure not due to cpp

[morabbin]

Have you tried StringGapBug2.hs that I sent in?  With 
my setup (vanilla 4.08.2) it fails even without -cpp.
The command line I use is:

    ghc-4.08.2 -c StringGapBug2.hs

and I get the following error:

    StringGapBug2.hs:27: error in character literal

I've had a quick look at Lex.lexstringgap and there 
doesn't seem to be any problem there (I had thought 
that the second \ may have been kept in the lexeme 
when starting to lex the string again, but that 
doesn't seem to be the case).

I've attached StringGapBug2.hs again.

[simonmar]

Logged In: YES 
user_id=48280

Yup, I tried it, and got the same error, on the line that 
begins '\\h', which really *is* an error, because '\h' is 
an invalid escape sequence.  If you delete that line and 
the previous one, the test compiles fine.  

Can I close this bug?

[simonmar]

Logged In: YES 
user_id=48280

Yup, I tried it, and got the same error, on the line that 
begins '\\h', which really *is* an error, because '\h' is 
an invalid escape sequence.  If you delete that line and 
the previous one, the test compiles fine.  

Can I close this bug?

[simonmar]

Logged In: YES 
user_id=48280

Yup, I tried it, and got the same error, on the line that 
begins '\\h', which really *is* an error, because '\h' is 
an invalid escape sequence.  If you delete that line and 
the previous one, the test compiles fine.  

Can I close this bug?