Reply-To: erik.wachtmeester@bighole.iaf.nl
Ron Tuynman wrote in a message to Erik Wachtmeester:
 RT>  //---Detecting:
 RT>  aRay := DBSTRUCT ( "NAME.DBF" )
 RT>  log := .f.
 RT>  FOR d = 1 to LEN aRay
 RT>      IF  "ADRESS" $ UPPER( aRay[ d ][ 1 ] )
 RT>          //---Log becomes true if a field named ADRESS exists
 RT>          log := .t.
 RT>      ENDIF
 RT>  NEXT
 EW> Instead of the for/next loop I would use something like:
 EW> log := ( ascan( aRay, { |x| x[1] = 'ADDRESS' } ) # 0 )
 RT>  ...Yep, but it's doesn't show as much what I'm actually doing as
 RT>  FOR...NEXT. I like to keep it as simple as possible. Especially when
 RT>  I'm explaining thing to someone else...
I agree on that one, but I simply love the speed and flexibility of code 
blocks
in Clipper, and I tend to use them whenever possible. If you're not using 
ode
blocks, you're not using Clipper to the full... ;-)
 RT>  And furthermore: I think your code wouldn't work..;-) Oh, sorry it
 RT>  would! Me myself I always use the '==' for comparison, not the soft
 RT>  '='. With == it would never have found ADRESS1 as a fieldname...
With strings '=' is a good replacement for e.g. '/^' in SED, AWK, Perl, etc. 
It
forces your program to check only at the beginning of the other string, while
'$' searches the whole string.
If you know the string to compare with must start with 'ADDRESS', it's just
natural to me to use '=' instead of '$'.
Regards,
Erik 
---
---------------