Date:         Wed, 21 Jun 2006 06:37:36 -0400
Reply-To:     Gerhard Hellriegel <ghellrieg@T-ONLINE.DE>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Gerhard Hellriegel <ghellrieg@T-ONLINE.DE>
Subject:      Re: Selective Unquoting

If you don't want the quotes, why don't you use

short=compress(short,'"');

??



On Tue, 20 Jun 2006 22:02:28 -0500, Rob Rohrbough
<Rob@ROHRBOUGH-SYSTEMS.COM> wrote:

>I am trying to import a comma-delimited dataset over which I have limited
>control.  The data contain some embedded quotes that have not been "doubled"
>for proper reading with the DSD option on the INFILE statement.  I am trying
>to read them anyway - of course with some problems.  A simple test case is
>as follows:
>
>data test;
>  infile datalines dsd truncover;
>  length Code $5;
>  length Short $40;
>  format AddDateQ mmddyy10.;
>  input Code : Short : AddDateQ : mmddyy.;
>  if substr(Short, 1, 1) = '"' and substr(Short, length(Short)-1, 1) = '"'
>then
>       Short = dequote(Short);
>datalines;
>"DUMM1",""LEADING AND" EMBEDDED QUOTES",04/01/2002
>"DUMM2","trail quote"",04/01/2002
>"DUMM3","NO QUOTE",04/01/2002
>"DUMM4","EMBEDDED" QUOTE ONLY",04/01/2002
>;
>
>This example shows that my little after-the-fact unquoting code handles the
>trailing quote, but not the record containing both in initial and middle
>"undoubled" embedded quote nor just the embedded quote.
>
>I can't understand why the unquoting code doesn't catch the first and last
>datalines.  If I examine the resulting "TEST" dataset, the quotes all are
>there, and my desk tracing seems to indicate that the unquoting should work.
>
>Ideas?  TIA,
>
>Rob
>
>
>Rob Rohrbough
>Omaha, NE, USA