The Viya 4 SAS Studio git integration is imho much more streamlined. With the future in mind (it's actually already present) I'd be looking into the migration path of SAS EG to SAS Studio and the git integration capabilities of SAS Studio.  The latest EG versions are now also compatible with SAS Viya but they don't offer the same richness in functionality. I'm especially fond of SAS Studio Flow and the growing number of SAS Studio Custom Steps available in public git repositories that one can use to build such flows.   ... View more

@cashaowan wrote: I run the codes you provided. SAS/Access Interface to PC Files is installed. The dataset I created has less than 5000 observations. Would appreciate if you can suggest the next SAS procedure.  Below one option: proc export data=sashelp.class outfile='c:\temp\myExcel.xlsx' dbms=xlsx replace ; sheet='Class'; run; And here some more options: https://blogs.sas.com/content/sasdummy/2012/01/25/export-to-excel-got-easier/ https://blogs.sas.com/content/sasdummy/2012/02/11/export-excel-methods/   Note: The path to the Excel file must be accessible from the location where the SAS process executes. If this is a SAS Server under Linux then it would be some Linux path.   ... View more

sas_execfilepath is an environment variable set and updated when you're using "PC SAS". If you're running in batch mode then SAS can't know in advance what program you want to run and though won't create this environment variable.  The -set option in the batch command @Ksharp shared lets you define and populate an environment variable that you then can query within your program. This then allows you to execute the exactly same code in batch.  Alternatively you can also query the value of SYSIN as it will also contain the path to the .sas file you're executing in batch. ... View more

The comments about using Proc Import have already been made so no more going there.  With Proc Import and text files as source: Use also GUESSINGROWS=MAX; to ensure SAS scans as many source rows as possible to determine type and length of the required SAS variables. And of course define the appropriate delimiter. ... View more

Are there any Error messages in the SAS log prior to the one you've shared? If so then please share.   For example if I execute your code then my SAS log looks like:   Do you have a locally installed SAS or are you connecting to a remote SAS Server via EG or Studio? ... View more

You don't need MS Office installed on the server to create MS Excel files using SAS. You do need SAS Module SAS Access Interface to PC files installed and licensed.    To check if SAS Access Interface to PC Files is licensed and installed run below commands and then check in the SAS log if it appears. /* licensed modules */ proc setinit; run; /* installed modules */ proc product_status; run; To further support you please let us first know if SAS Access Interface to PC Files is available to you.   Btw: What is a "large SAS dataset"? Excel got a limit of 1M rows per sheet. ... View more

@PaigeMiller wrote: @Patrick wrote: @PaigeMiller The usual question is like always: "What's different between the two runs?" 1. Is this running identical code? 2. Are you using the same or different SAS Servers? 3. What's the database? 4. Is it a distributed database with potentially region specific views?   I feel @Tom is onto something here.  You could via explicit pass-through verify if the data type is the same on the DB side as well as get the db local.  Then on the SAS side I'd compare again settings like the local. If there are no differences then you would likely need to look into the drivers and the like.  1. Yes 2. No SAS server, this is base SAS 9.4 3. SQL Server 4. I doubt it, but can't say for sure.   Will try passthru There must be a difference either on the Server side (SQL Server), connection (driver) or client side (SAS).  I'm not aware of any such differences that would cause a SQL Server date column to become Character on the SAS side so I'd also first verify that the data types on the SQL server side are the same using code similar to below: libname yourlib sqlsrv ....; proc sql; connect using yourlib; select * from connection to yourlib ( SELECT * FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = '<schema>' AND TABLE_NAME = '<table>' and COLUMN_NAME='<column>' ); disconnect from yourlib; quit; I'd be running this test via SAS using the exactly same libname definition as in the code with the issues to ensure we're accessing the exact same table or view. ... View more

This error looks to me like it can't find the JDBC driver.   I've never tried this myself so can't really be of further help. If you don't get sufficient answers here in the forum then contact SAS TechSupport directly.   Below what Copilot suggested - which might or might not be correct.   To connect SAS to Databricks on AWS using the LIBNAME statement, you can use the JDBC interface. Here’s an example of how you might set this up: libname mydblib jdbc classpath="/path/to/jdbc/driver" class="cdata.jdbc.databricks.DatabricksDriver" URL="jdbc:databricks://<hostname>:443;httpPath='/sql/1.0/warehouses/<warehouse_id>';ssl=1;AuthMech=3;UID=<your_username>;PWD=<your_password>"; Replace the placeholders with your actual Databricks server details: <hostname> : Your Databricks server name. <warehouse_id> : The ID of your Databricks SQL warehouse. <your_username> : Your Databricks username. <your_password> : Your Databricks password. Make sure the JDBC driver is correctly installed and the classpath points to the location of the driver. If you encounter any issues, double-check the connection properties and ensure they are correctly formatted. For example, the httpPath should not have single quotes around it, and ssl should be set to 1 if using SSL. Establishing a libname connection to Databricks from SAS Viya 4 on AWS How to Speed Up Data Flow Between Databricks and SAS   ... View more

I've never done it so don't really know. But just based on the documentation found here it looks to me that at least some of the connection info needs to be passed via parameter URL. Connecting to Databricks in Microsoft Azure This example accesses Spark data in Databricks on Microsoft Azure. The Databricks JDBC driver is recommended for accessing Spark data in Databricks. libname spkdbk spark classpath='/JAR-path' driverclass="com.databricks.client.jdbc.Driver" bulkload=no url=' jdbc:spark://server:port/schema;transportMode=http;ssl=1;HTTPPath=myHttpPath; AuthMech=3;defaultStringColumnLength=255;useNativeQuery=1;<options>' user=token password=mytoken ; ... View more

Just as an observation: Wouldn't below semicolons not need to be within the single quotes so that the first one doesn't terminate the SAS libname statement?     ... View more

@PaigeMiller The usual question is like always: "What's different between the two runs?" 1. Is this running identical code? 2. Are you using the same or different SAS Servers? 3. What's the database? 4. Is it a distributed database with potentially region specific views?   I feel @Tom is onto something here.  You could via explicit pass-through verify if the data type is the same on the DB side as well as get the db local.  Then on the SAS side I'd compare again settings like the local. If there are no differences then you would likely need to look into the drivers and the like.  ... View more

For macros and included code what could work is using options. options mautolocdisplay source2; These two options write the paths to the source code into the log for any code that has been executed. You then could parse the log for these paths and copy the source code to your report.   Example: I've stored below attached .sas files under c:\temp and then run below code. /* add c:\temp to SAS Autocall Facility */ options insert=(sasautos='C:\temp'); /* set option that writes macro source path to SAS log */ options mautolocdisplay; %demo(Hello World); /* set option that writes location and code of included scripts to SAS log */ options source2; %include 'C:\temp\demo3.sas';   This will create SAS log that contains the paths to both the macro and included program source code.  The approach will also work for nested macro calls and for "utility" macros that get compiled as part of a call to another macro. It will also work for "overlay" macros where the source code exists in more than one location. And it will also work for nested %include statements (=an included program containing an %include to another program). 31 /* set option that writes macro source path to SAS log */ 32 options mautolocdisplay; MAUTOLOCDISPLAY(DEMO): This macro was compiled from the autocall file C:\temp\demo.sas 33 %demo(Hello World); MAUTOLOCDISPLAY(DEMO_UTIL): This macro was compiled from the autocall file C:\temp\demo.sas MAUTOLOCDISPLAY(DEMO2): This macro was compiled from the autocall file C:\temp\demo2.sas demo: Hello World demo_util: Hello Again demo2: Hello from another place NOTE: DATA statement used (Total process time): real time 0.00 seconds cpu time 0.01 seconds 34 35 /* set option that writes location and code of included scripts to SAS log */ 36 options source2; 37 %include 'C:\temp\demo3.sas'; NOTE: %INCLUDE (level 1) file C:\temp\demo3.sas is file C:\temp\demo3.sas. 38 +%put '2nd call of macro %demo'; '2nd call of macro %demo' 39 +%demo(Hello World once more); MAUTOLOCDISPLAY(DEMO): This macro was compiled from the autocall file C:\temp\demo.sas MAUTOLOCDISPLAY(DEMO_UTIL): This macro was compiled from the autocall file C:\temp\demo.sas 2 The SAS System Thursday, October 10, 2024 11:34:00 AM MAUTOLOCDISPLAY(DEMO2): This macro was compiled from the autocall file C:\temp\demo2.sas demo: Hello World once more demo_util: Hello Again demo2: Hello from another place What you won't get is the path to source code that has been called in code but never has been executed (i.e. a macro call that didn't get executed because the call was within an %IF condition that was false).   I can't think of a way how you could directly get the source code of custom formats stored in some permanent library. You could search for the format along the paths stored in option FMTSEARCH and if found use proc format cntlout= to write the format definition to a SAS dataset.   If you also got custom functions (proc fcomp) then I believe a similar approach could be possible.     ... View more

Thank you! I tried changing the registry but didn't have sufficient permissions so replacing the .cfg it is. ... View more