Date: Mon, 11 May 2009 22:23:12 -0700
Reply-To: Savian <savian.net@GMAIL.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Savian <savian.net@GMAIL.COM>
Subject: Re: Report mining techniques question.
Content-Type: text/plain; charset=ISO-8859-1
On May 8, 12:55 pm, sk...@HEALTH.STATE.NY.US (Stephen Dybas) wrote:
> Hello SAS-Ls,
> I have been asked to create a datafile from a ASCII report file. The
> report has a lot of report titles that I want to skip over and extract the
> data from the lines that actually contain the information. The data within
> the report are in predictable colums but the data will vary greatly from
> row to row.
> Does anyone have any leads on papers obout report mining or common parsing
> techniques using SAS. Any information will be appreciated.
> Stephen Dybas
Download RegexBuddy and start to learn regular expressions. A small
amount of pain up front but it is meant specifically for this issue.
Translate that code into SAS (trivial) and you are ready to go.
You can do this stuff with SAS string functions but they are no
replacement for regex. There are certain SAS string functions that
cannot be duplicated in regex (easily) but you don't seem to be facing