About Maze

Maze · ‎10-14-2019

Thanks everyone for the pointers, doc is pretty clear, I don´t know why I did not see that. Why do you want to read the file twice? I have a dataset with table names that include a structure tag. Based on that tag I read the according structure file to generate a DDL script (structures may repeat!).

Maze · ‎10-11-2019

Hi everyone, I´m having trouble reading the same file twice inside a datastep. data WORK.OUT; infile datalines; length INP $40; input INP $; infile dummy filevar=INP end=EOF; do while(^EOF); input VAR1 VAR2 VAR3; output; end; datalines; /path/to/file/file1.txt /path/to/file/file2.txt /path/to/file/file2.txt /path/to/file/file3.txt ; Each file is read and output once even though file2.txt should have been output twice. The code is just exemplary. The real code depends on a list of items in a dataset for which (depending on a parameter) different external files should be read. If the same file occurs two times in a row it is not read again. If files alternate like below, they are read every time: /path/to/file/file1.txt /path/to/file/file2.txt /path/to/file/file3.txt /path/to/file/file2.txt What am I missing? Do I need to reset a pointer or the infile? Or is this not intended? Any help is appreciated! /edit: formatting...

Maze · ‎06-23-2017

Hey everyone, I have a problem regarding file transfer via SAS Portal between plattforms. It´s a quite simple Task: The user presses a button on my SAS Portal page which calls a Stored Process. That STP creates 1-n files in a loop and puts them in an archive. ODS PACKAGE(LISTZIP) open nopf; //for every id provided, create an xml file and add it to the package ODS PACKAGE(LISTZIP) add file="&filepath._&i." mimetype="text/xml" TEXT; ODS PACKAGE(LISTZIP) publish archive properties (archive_name="&l_archiv." archive_path="&l_work."); ODS PACKAGE(LISTZIP) close; After file creation, I have tried either data set to _webout or proc stream to return the zip-file to the user %LET RV = %SYSFUNC(appsrv_header(Content-disposition,%STR(attachment; filename=&fname.))); %LET RV = %SYSFUNC(appsrv_header(Content-Type,application/octet-stream)); filename out "&outfile."; proc stream outfile=_webout; begin &streamdelim. readfile outfile; ;;;; run; /* or data step */ data _NULL_; file _WEBOUT; infile "&outfile."; input; put _INFILE_; run; I download the archive on a windows client and open it, but it only shows one out of three files and if I try to open that file I get a CRC-error, file corrupted. Weirdly enough, this only happens if the files inside the archive are larger than a certain size. Further can I access the produced archive on the Unix filesystem, transfer it to my machine wie ftp client and open it and its contents without any problems at all. When hex-comparing the downloaded version with the ftp´d one, I can see there are a few '0A's in the downloaded version that are not in the ftp´d one and a few '0D's in the ftp´d version that are not in the downloaded one. Having all these hints leads me to believe I have some issue regarding linebreak-formatting in transfer (0D Carriage-Return, 0A Line-Feed (Unix), 0D0A (Windows)). But I am not sure how to adress this issue. The files are generated and archived on the Unix-Server, how do I stream the archive to a client in SAS Portal to be accessible? I´m working with SAS 9.4 Kind regards, Mat PS: As a workaround I´m sending the archive via Mail and it works without a problem. Strengthens my focus on the issue beeing transfer to the client via browser.

Maze · ‎12-12-2014

just a heads up: I fixed the problem by manually inserting a CR/LF (carriage return/line feed) character ('0D0A'X) at the end of every line. The plain file now has an empty row between every relevant row, but the file inside the package is correctly formatted, which is what matters to me. Explanation: Unix uses LF or in hex '0A'x to indicate the end of a line. Windows uses CR/LF or in hex '0D0A'x. (Macintosh uses CR '0D'x) It seems like windows just ignores the '0A'x. I´m working on a cleaner solution that replaces all LF with CR/LF instead of adding a CR/LF to an existing LF.

Maze · ‎12-10-2014

thanks for your suggestion, I´ll try the @ variant. although I think the problem is, that the end-of-line value gets messed up between unix and windows. I realized it looks like the line is filled until lrecl is reached instead of having a line for each put-statement. changing the output format is no option, I´d never do it like I have to if I were given the choice

Maze · ‎12-10-2014

hey there, I ran into a problem I can´t seem to resolve on my own: I´m creating output in a .txt file on a unix server. The file is then beeing archived (ODS PACKAGE) and sent to several people using the mail-engine. When I open the file on unix directly, it is perfectly aligned, every record has it´s own line and the same values start in the same columns. Opening the same file from the mail on an windows environment shows that there is a offset (7 characters, can´t find the reason for the number) which every line is offset from the next one. Example: Unix Output (correct): values that are only on the first row this is the first row 123 randomstring this is the second row 472 anotherstring this is the third row 555 laststringvalue values that are only on the last row Windows Output: values that are only on the first row this is the first row 123 randomstring this is the second row 472 anothe rstring this is the third row 555 laststring value values that are only on the last row It seems like there is a problem between the static first row and the first dynamic row generated from column values The code looks something like this: data WORK.OUT end=EOF; file = "/bla/bla/out.txt"; if _N_ = 1 then put 'values that are only on the first row'; put STR1 $26. STR2 $7. STR3 $15.; if EOF then put 'values that are only on the last row'; run; I´d appreciate it, if someone could point me in the right direction. Thank you.

Maze · ‎10-17-2014

Hi! I´m having a problem regarding naming conventions in SAS. I working with a table that I have no control over and one columns identifier contains a hashtag (ID_XXX#). To insert rows via 'proc sql', I have to omit the insert column listing because SAS can´t handle the # in the one column´s name. Since omitting this list is bad practice and may crash my program if there are changes made to the table, I´d like to find a way to handle the problem. How to handle column names containing hashtags

Online Status	Offline
Date Last Visited	‎10-14-2019 05:08 AM

Re: read same file twice inside datastep

read same file twice inside datastep

Cross-plattform archive transfer / SAS Portal

Re: Create file on unix, open it on windows

Re: Create file on unix, open it on windows

Create file on unix, open it on windows

Hashtag (#) in foreign table´s columnname

Re: read same file twice inside datastep

Re: read same file twice inside datastep

Re: read same file twice inside datastep

Re: Hashtag (#) in foreign table´s columnname

Cross-plattform archive transfer / SAS Portal

Re: read same file twice inside datastep

read same file twice inside datastep

Cross-plattform archive transfer / SAS Portal

Re: Create file on unix, open it on windows

Re: Create file on unix, open it on windows

Create file on unix, open it on windows

Hashtag (#) in foreign table´s columnname

CoDe SAS German