From: ahoward <ahoward@...>
Date: 2003-04-09T04:35:57+09:00
Subject: Re: 1.8 parse error?

On Wed, 9 Apr 2003, Yukihiro Matsumoto wrote:

> Hi,
>
> In message "1.8 parse error?"
>     on 03/04/09, ahoward <ahoward@fsl.noaa.gov> writes:
>
> |i'm getting :
> |
> |~/eg/ruby/btpgsql/btpgsql-0.2.0 > ruby test/btpgsql.rb
> |test/btpgsql.rb:1191: ERROR:  parser: parse error at or near "#!/" (PGError)
> |
> |        from test/btpgsql.rb:1186:in `each'
> |        from test/btpgsql.rb:1186
>
> Since it's PGError, not SyntaxError, the error must be in your SQL
> statement.

it's not the SQL, but your suggestion pointed me towards the error :

  /tmp > cat bug.rb
  path = (File.dirname (__FILE__) << File::SEPARATOR << 'btschema.sql')


  /tmp > ruby -v
  ruby 1.8.0 (2003-03-03) [i686-linux]

  /tmp > ruby bug.rb
  bug.rb


  /tmp > /usr/local/bin/ruby -v
  ruby 1.6.8 (2002-12-24) [i686-linux]

  /tmp > /usr/local/bin/ruby bug.rb
  ./btschema.sql

so

  (File.dirname (__FILE__) << File::SEPARATOR << 'btschema.sql')


is not the same under 1.6 and 1.8.  i can fix this by changing it to

  (File.dirname(__FILE__) << File::SEPARATOR << 'btschema.sql')

(note lack of space!)


has the precedence of '<<' changed or something?  i don't understand why this
should confuse the parser?

why the insistence the parens immeadiately follow method names?  i like the C
style of

  method (args);

much more than

  method(args);

especially when doing something like


  method0 (method1 (method2 (arg)))

which is less straining than

  method0(method1(method2(arg)))

IMHO.

-a





--
  ====================================
  | Ara Howard
  | NOAA Forecast Systems Laboratory
  | Information and Technology Services
  | Data Systems Group
  | R/FST 325 Broadway
  | Boulder, CO 80305-3328
  | Email: ahoward@fsl.noaa.gov
  | Phone:  303-497-7238
  | Fax:    303-497-7259
  ====================================