Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[imp-dev] Re: IParseController.parse() parameter 'scanOnly'

The motivation is in fact all of the above (ok, all of the below, for you nit-pickers).

On Mar 12, 2009, at 11:06 AM, Stan Sutton wrote:


Hi Bob,

Your reasons all sound reasonable, but I'm not quite sure why you want to do it.  To simplify the interface?  To remove elements that haven't been justified by use?  To eliminate a simplistic solution to a complex problem?

I guess I wouldn't care, except that in the abstract it does seem like a reasonable option to have.  (And not all scanners/parsers will operate incrementally.)  Could or should we remove the scanOnly parameter from parse() but add a scan() or scanOnly() method instead?

Otherwise, I'm fine with your suggestion.  :-)

Stan

"Robert M. Fuhrer" <rfuhrer@xxxxxxxxxxxxxx>

03/12/2009 10:52 AM

To
IMP Developers List <imp-dev@xxxxxxxxxxx>
cc
Jurgen Vinju <Jurgen.Vinju@xxxxxx>, Stan Sutton/Watson/IBM@IBMUS, "Philippe G. Charles" <pgcharles@xxxxxxxxxxx>
Subject
IParseController.parse() parameter 'scanOnly'





Hi All,


I'd like to remove this formal parameter from the interface. Its  
original intent
was to permit a quick re-scan of the source without a full parse, but  
the
problem of scheduling analyses for responsiveness is much deeper than
this one flag addresses (and arguably incremental scanning/parsing is a
superior solution). Oh, and no one is actually using it (that I'm  
aware of). :-)

Any objections?

--
Cheers,
  - Bob
-------------------------------------------------
Robert M. Fuhrer
Research Staff Member
Programming Technologies Dept.
IBM T.J. Watson Research Center

IMP Project Lead (http://www.eclipse.org/imp)
X10: Productivity for High-Performance Parallel Programming (http://x10.sf.net)


Back to the top