Hi,   I have written the following code (just a simple test case of a much bigger program). I have a dataset that has a variable of type character string but in fact its values are timestamp. I need to add another variable as some temporary variable for some further calculations in this data step with the real SAS timestamp data type so that I can use SAS corresponding functions for timestamp/date type (extract year, hour, min, etc.)   Here is the code:   data test; input timestamp_str :$17.; datalines; 18-01-12T14:22:31 19-05-17T20:34:10 19-07-04T14:24:43 run; data test; set test; timestamp_value = input(timestamp_str, ANYDTDTM17.); run; Years in the dataset test include only two digits so for example the 2nd observation which is 19-05-17T20:34:10 is to be interpreted in the following way:   year = 2019 month = 05 (May) day = 17 hour = 20 minute = 34 seconde = 10 The problem is once the program has been run the new variable timestamp_value is empty. I don't get any error message but I suspect that this should be related to the format I've specified: ANYDTDTM17. If that is the case could someone kindly guide me about the correct format?   thanks in advance ... View more

Hi,   SAS version: 9.4   I would like to know whether it is possible to select specific columns from a pipe delimited text file and insert them into a SAS dataset. I have a huge file (several gigabytes) with 100 columns (= variables from SAS point of view), but I need to read only specific columns from that file. For example: the 10th, 23rd, 42th, 51th and 75th columns from the file.   I believe that similar questions have already been asked but the answers I found while I was googling, were mostly for column aligned data with fixed lengths and as a result fixed start and end positions for each column, that is, using  the @ operator with appropriate column number as the start reading position, followed by appropriate offset numbers from that point, to continue reading by skiping the unwanted columns.   But my problem is different in a sense that there is no assumption on a fixed equal length per column (apart from the max possible length) and it is the delimiter "|" which separates the data and it is the sole criterion that allows to distinguish columns one from another.   Therefore, I would like to ask whether it is possible to do this and select specific (varying size) columns from a pipe delimited text file and insert them into SAS.   Thanks in advance, ... View more

Hi,   SAP version 9.4 Data file (data.txt) George Albert Bob My SAS program to read the data above: DATA Work.names; INFILE '/folders/myfolders/exercices/data/data.txt' missover ; INPUT name :$ 6.; RUN; PROC PRINT DATA = Work.names; TITLE 'Names'; RUN; It seems to me that: INPUT name :$ 6.; And (putting the colon after the $ character) INPUT name $: 6.; Both give the same result. As I'm a beginner I just wanted to know why. Is it just the fact that both syntaxes are accepted by SAS or there is some concept that I've missed?   Thanks in advance, ... View more