It's Regular Expressions to the rescue! [pre] data _null_; infile datalines; retain r; if _n_ = 1 then r = prxparse("/^0*(.*)$/"); input address $80.; if prxmatch(r, address) then do; call prxposn(r, 1, start, length); address = substr(address, start, length); end; put address; datalines; 000000000000000000000000000123 Elm 000000000000000000000000123 Elm St 00000000000000000000000123 Elm St. 00000000000000000000123 Elm Street 000000000000000000000000001050 Elm 0000001234 NE 1st Street, Extended 1111 Twisted Turny Long Avenue, NE ;;;; [/pre] produces [pre] 42 data _null_; 43 infile datalines; 44 retain r; 45 if _n_ = 1 then 46 r = prxparse("/^0*(.*)$/"); 47 input address $80.; 48 if prxmatch(r, address) then do; 49 call prxposn(r, 1, start, length); 50 address = substr(address, start, length); 51 end; 52 put address; 53 datalines; 123 Elm 123 Elm St 123 Elm St. 123 Elm Street 1050 Elm 1234 NE 1st Street, Extended 1111 Twisted Turny Long Avenue, NE NOTE: DATA statement used (Total process time): real time 0.01 seconds cpu time 0.01 seconds 61 ;;;; [/pre] ... View more

This looks like something Tech Support should know about. You can submit a problem to Tech Support here: http://support.sas.com/ctx/supportform/index.jsp. ... View more

Sivas, When I look at an across variable in PDF, what I see is that the variable is in its own cell surrounded by borders on all 4 sides. What do you want that's different from this? Do you want the variable name underlined? [pre] +---------------------------------------------+ | Sex | +---------------------------------------------+ | M | F | Other | +---------------------------------------------+ [/pre] I'm using SAS 9.1.3 SP4 with the default style. ... View more

The whole statement is a formatted PUT in the form [pre] put (variable-list) (format-list); [/pre] Within (format-list) you can put formats, character strings, and pointer controls. You can read about it in the Put Statement, Formatted section of the SAS Language Reference: Dictionary online: http://support.sas.com/onlinedoc/913/docMainpage.jsp. See especially the example "Writing a Character between Formatted Values". The DSD option puts commas between the values by default. The +(-1) pointer control is an trick to prevent that by backing up the column pointer 1 column. I think I must've learned about it on sas-l or something. ... View more

Well, the doc says "Even though a character string or character variable is accepted, only the first character of the string or variable is used as the output delimiter." However, there's no reason you can't make up your own CSV output. Would this (or some variation of this) work? [pre] data; input field1 field2 field3 field4; datalines; 2 123 432 0.0101 2 456 . 0.01 2 . 876 0.0356 2 98 654 0.0112 ;;;; data _null_; set; file log dsd; put field1 (field2--field4) (+(-1) "||"); run; [/pre] ... View more

Sorry, John. There's no way to do this with REPORT. Then nonbreaking spaces are deliberate. REPORT really wants each LINE statement to create exactly one line of output. Possibly you could post-process the RTF output but that's out of my league. ... View more

Quite possibly, but I'd have to have more details about exactly what it is you want to do. ... View more

You say "resolution," that is, pixels per inch or centimeter, but it sounds like you mean "dimensions," the number of columns and rows. In either case, check out ImageMagick at http://www.imagemagick.org. This package includes a command called "identify" that prints the dimensions and resolution of an image along with a lot of other information. Here's a SAS program that runs identify and parses its output: [pre] filename id pipe 'identify myimage.jpg'; data _null_; attrib imgname length=$12 format length=$8 dimensions length=$9; infile id; /* Skip to the 2nd line. The first word in the line is the image filename. The 2nd is the image format, in this case 'JPEG'. The third word are the dimensions, which are displayed in the form CCCCxRRRR. */ input / imgname format dimensions; columns = input(scan(dimensions, 1, 'x'), best.); rows = input(scan(dimensions, 2, 'x'), best.); put _all_; run; [/pre] If you need the resolution use the -verbose option on identify. ImageMagick is free, supports over 100 different image formats, and runs on just about any platform you can imagine. ... View more

Since this is a forum for ODS and the Base Reporting procedures, I doubt very much that there are any authorities on this question here. However, Tech Support is ideally suited for this kind of question. To contact Tech Support, go to http://support.sas.com/ and click "Submit a Problem". ... View more

Here's my solution. Use substr to extract the date part of the string and the time part of the string into two variables. Then use the input function with the mmddyy10. informat to convert the date string to a SAS date value, and with the time8. informat to convert the time string to a SAS time value. Given a SAS date value and a SAS time value, the SAS documentation for the DHMS function explains how to combine them into a SAS datetime value. Here's a data step to test my algorithm: [pre] data _null_; input str $20.; dtstr = substr(str, 1, 10); tmstr = substr(str, 12, 8); dt = input(dtstr, mmddyy10.); tm = input(tmstr, time8.); dttm = dhms(dt, 0, 0, tm); put dttm datetime19.; datalines; 10/16/2007 22:37:30 05/01/1953 11:15:27 07/04/1976 23:59:59 12/31/2000 00:00:00 run;[/pre] Please test this for yourself to make sure it works. ... View more

As this is a forum for questions about ODS and Base Reporting, there aren't likely to be a lot of people here who are experts with this problem. Your best bet is Tech Support. To find out how to contact SAS Tech Support, refer to: http://support.sas.com/techsup/contact/index.html. ... View more

The best doc for tagsets is here: http://support.sas.com/rnd/base/topics/odsmarkup/tagsets.html. You'll find an explanation of "eval" there. ... View more

Hi, This is not the right forum for this question. In fact, this is really a question that is best answered by Tech Support. To find out how to contact SAS Tech Support, refer to: http://support.sas.com/techsup/contact/index.html. ... View more

How about this? [pre]data; input component $8. close; datalines; Apples 8 Apples 12 Oranges 10 Apples 15 Bananas 4 Bananas 3 Pears 17 Oranges 14 Pears 23 Apples 12 Bananas 3 Limes 22 ;;;; proc report nowd; columns component n close; define component / group width=10; define n / 'Number of Obs/Per Product' width=13; define close / mean format=3.1 width=5; run;[/pre] ... View more