Move InitLineEnds and GetLineNumber to handles.cc to avoid...

Move InitLineEnds and GetLineNumber to handles.cc to avoid
issues with a raw this pointer and GC.

Fix issue with FlattenString where we may keep a StringShape
alive across a GC.

Committed: http://code.google.com/p/v8/source/detail?r=1481

[Mikhail Naganov, 2009-03-10 15:00:13]

LGTM

[Mads Ager (chromium), 2009-03-10 19:26:58]

LGTM

http://codereview.chromium.org/43020/diff/1/3
File src/handles.cc (right):

http://codereview.chromium.org/43020/diff/1/3#newcode189
Line 189:
CALL_HEAP_FUNCTION_VOID(string->TryFlattenIfNotFlat(StringShape(*string)));
I think this change makes the code much clearer (and I think we should get rid
of StringShape all together - we have talked about doing that for a while now).

I think that this code was safe the way it was though.  The only strings that
change their shape during GC are ConsStrings that have an emtpy right-hand side.
 For those strings, string->IsFlat(shape) will return true and we will bail out
before the part that can cause a GC.  For other strings, the shape is still
valid after a GC.

Anyway, this is a good change!