alternatively select test_values format= $quote200. into: test separated by ',' deals with trailing spaces as required and highlights that we can apply useful formats when SELECT-ing INTO :macrovars ... View more

Original q "in general, how do I see what the default value of the optional parameter" Usually the doc standards are good and the default is explained. Here it is just cross referenced ... View more

Sounds like you want a grouping variable that collects (to aggregate) data in the 6 month window. That grouping variable would be used in a 'group by' clause. Each source row would contribute to multiple(6) 'tracts'. Judging by the topic and use of SQL and 6-fold expanding of the data, it might be wise to make this expansion in a VIEW or views, prior to aggregation. ... View more

thanks Scott Probably put it down to my ARADD (age related attention deficit disorder)😎 ... View more

Hi Kevin My suggestion was only for readkng contents of the files. I wouldn't use it for file names. However, with some scenarios (e.g. expecting only small files) your proposal works, but excludes empty files. I don't remember what happens for subfolders. Peter  ... View more

Hi Kevin Any extension of this simple routine should be cautious about handling binary files like .xlsx format. The simple precision of the DSD option in handling "xxxxx separated values" won't work for binary data. However, it only needs a DLM= '09'x added to the INFILE statement to switch to tab separated values. If there is acceptable standard that .csv files need DLM= ',' and .tsv files need DLM= '09'x then these could be dynamically supported in folders with a the mix of file types, because the DLM= value can name a variable which you change depending on the file type/extension. Timing takes care and probably, practise. My guess is the added complexity is not well rewarded. By the way, the DLM= value can hold more than one character. These then are alternative delimiters. So, if tabs only appear as delimiters in .csv files and likewise commas are protected (by quotes) when not to be treated as delimiters .... then the DLM could be that combination '2c09'x ... View more

I know It's been done (to death😎) But for such a simple request it's nice to provide the simple solution: Read all csv files in one folder into one dataset..... Data resultDS( compress= yes) ; Length file filename col1-col100 $300 ; INFILE "yourfolder/?*.csv" lrecl= 32000 dsd filename= file truncover ; Input col1-col100 ; Filename=file ; Run ; ... View more