Date: Sun, 2 May 2010 23:12:10 -0400
Reply-To: "Keintz, H. Mark" <mkeintz@WHARTON.UPENN.EDU>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "Keintz, H. Mark" <mkeintz@WHARTON.UPENN.EDU>
Subject: Re: Enhancements to SAS editors: was Notes from SAS Global Forum
In-Reply-To: <00af01caea65$9a4d3cb0$cee7b610$@savian.net>
Content-Type: text/plain; charset="us-ascii"
Another feature of KEDIT, it's mainframe predecessor XEDIT, and I presume ISPF, is the ability to hide columns, thereby showing columns 1-20 contiguous with, say, columns 60-80. Very handy for examining raw data, when you want to confirm the presence of certain data values in widely separated parts of the data record.
MK
> -----Original Message-----
> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of
> Alan Churchill
> Sent: Sunday, May 02, 2010 10:09 PM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: Re: Enhancements to SAS editors: was Notes from SAS Global
> Forum
>
> Nat,
>
> Personally, I would use a regex in ultraedit to do what you indicated.
> Regex, though, may be difficult for some people who aren't used to that
> low-level detail.
>
> This is a good case to have a) SAS R&D add it to their list, b) have
> someone
> write an EG Add-In that contains this functionality.
>
> Personally, I would like to see b because I think the user community
> should
> start participating and sharing code.
>
> Alan
>
> Alan Churchill
> Savian
> Work: 719-687-5954
> Cell: 719-310-4870
>
>
> -----Original Message-----
> From: Nat Wooding [mailto:nathani@VERIZON.NET]
> Sent: Sunday, May 02, 2010 8:18 AM
> Subject: Enhancements to SAS editors: was Notes from SAS Global Forum
>
> I just looked at the EG discussion group on SAS Community but decided
> to
> place the following thoughts on the L. It seemed time to change the
> subject
> line to be more meaningful.
>
> There is a feature in the ISPF editor on the mainframe that I sorely
> miss in
> the Windows environment and value to the point that I will move a pc
> file to
> the mainframe in order to have it available. In the ISPF editor, one
> has the
> ability to "exclude" or hide all of the lines of text (or just a few if
> that
> is useful) in a file. Then, one may issue find commands to locate the
> occurances of strings and the entire line is displayed wherever the
> string(s) is/are found. This is very nice if you need to change a
> variable
> name and need to ensure that what you are changing is actually what you
> want
> to change. For example, if you want to change "DATE", you do not want
> to
> change the format "DATE.". Yes, in the windows editor, you can do a
> find/replace and step through them one at a time but in the ISPF
> scenario,
> you can display all of the occurances and muse over them before making
> a
> global change or you can do a repeat find/replace and select the
> desired
> changes.
>
> I have long used this feature when I needed to locate a program or some
> bit
> of code. In Windows, I can now give programs much more meaningful names
> than
> on Z/OS but sometimes the names still do not help. On Z/OS, there is a
> tool
> called Proc Source which will copy each file in a directory (there are
> some
> nuances to that I am going to ignore) and place them in a single text
> file
> along with a line that identifies the original program. When you open
> this
> "flat" file with the editor, you may
>
> 1) exclude all the lines
> 2) start doing "find all" for the program name lines as well as title
> lines
> or some other suitable strings.
> 3) Once you find the programs with the items suggest that that they may
> be
> candidates, you may quickly delete the large excluded blocks that are
> not of
> interest and
> 4) unexclude the text and start looking for the desired program or code
> elements.
>
> You may, of course, do a Windows search for specific strings but you
> are
> then presented with a list of files that must be opened one a time. A
> while
> back, I wrote a PC SAS program that will extract the SAS programs from
> one
> or more folders and dump them in much the same manner as Proc Source.
> Using
> the "flat" file, I can either use the DM find feature or move the file
> to
> the mainframe where the really useful tools live.
>
> For those who might find the program of interest, here 'tis.
>
> Birdies: Are you listening?
>
> Nat
>
> dm 'log;clear;output;clear;pgm;';
>
> ** note that the member name is not correct. the scan function is not
> recognizing the slashes
>
> ** as delimiter. reported as us 6125345
>
> ** note that the ./ add card may need a bit of editing on the mainframe
>
> ** SearchText
>
> ** Nat Wooding
>
> ** Dominion Virginia Power
>
> ** March, 2003
>
> **
>
> ** this code lists the statements of any files in a given path that
> have the
> extension
>
> ** SAS. The code is displayed in the sas log and the name of the file
> is
> also shown.
>
> ** The output window may be searched for specific strings using the
> find
> option in the
>
> ** SAS pull down Find command;
>
>
>
> data a;
>
> filename txt ('u:\my sas files\natsas\pgms 2005\*.sas',
>
> 'u:\my sas files\natsas\pgms 200\8*.sas');
>
> length file $ 50;
>
> infile txt filename=file lrecl=80 truncover;
>
> informat txt $72.;
>
> input txt $ 1-72;;
>
> if index(file,'7bdat') then delete;* do not include sas data sets;
>
> source=file;
>
>
> source=scan(source,-2 );
>
> /* now, add a break between the code from each file. */
>
> run;
>
>
>
> data a;
>
> set a;
>
> by notsorted source;
>
> if first.source then do;
>
> hold=txt;
>
> txt='./add name ='||source ;* the ./add here is for use if I want to
>
> * to move the programs to the mainframe.
>
> * note that it also sticks the file name at the
>
> *front ;
>
> output;
>
> txt=hold;
>
> output;
>
> end;
>
> else output;
>
> drop hold;
>
> * you now have your sas code in a sas data set which you could write to
> a
> flat file
>
> and edit if you wish.
>
> *;
>
> data a;
>
> set;
>
> filename new 'c:\park\upload storage.txt';
>
> file new;
>
> put txt;
>
> proc print;
>
> run;
| 
