SAS Transport Files

People who have used SAS on other systems may well want to continue using the same SAS Data Sets on new platforms to which they will be moving. This document outlines and gives examples of the process of creating SAS Transport Files out of SAS Data Sets, moving these Transport Files to a new system, then reconverting the Transport Files into true SAS Data Sets on the new system.

NOTE: When moving to UNIX, it is possible to store SAS Transport Files as such in UniTree, then access them directly from there. This is especially convenient for large data sets that would be impossible to store on disk and/or cumbersome to move back and forth from UniTree to UNIX disk for analysis. Details can be found in the general SAS help document or in the specific "SAS and UniTree" excerpt in

http://www.usc.edu/its/doc/statistics/help/sasunitree/

When moving between systems (not using SAS/CONNECT), it is necessary to prepare SAS Data Sets for the move, as simple transfer of true SAS Data Sets between operating systems will corrupt them. Most types of SAS files, in fact, can be transferred from system to system, including Catalogs (such as Format Catalogs) as well as SAS Data Sets. Catalogs, however, cannot be transferred from one Release to a lower- numbered release (such as from 6.12 on UNIX to 6.08 on a mainframe). Catalogs going to lower-numbered releases must be re-created on the target system using SAS code. This document outlines the procedures for transferring SAS Data Sets, but the procedures for transfer of Catalogs, for example, are similar. Consult the appropriate SAS Manuals for details.

To accomplish the transfer of SAS Data Sets from one operating system to another it is necessary to

1. convert the data set into "transport" format
2. move, send or carry the data in "binary" (i.e., non-translated) format to the operating system running the other Version
3. convert the transport data set back into a SAS Data Set

For details on the procedures outlined below, users are encouraged to use SAS Language: Reference, Version 6 Edition. Also helpful is a special SAS Technical Report, No. P-195, available only from SAS Institute, and titled, "Transporting SAS Files Between Host Systems".

Some data transfer operations will require the use of documents that go beyond the one you are now reading. For one thing, it is important to note that the examples in this section assume moderate-sized data sets. Larger data sets may require increased space allocations or other manipulations when sending and/or receiving. Such manipulations are specific to the method of transfer or operating system(s) involved, and since the possibilities are quite numerous they cannot be documented here in detail. See related User's Manuals and documents for assistance.

Step One: Preparing the Data for the Move: Transport Format

Example 1: (Sending from UNIX)

libname olddata '/home/almaak/user12/sasdata';  **this is 
                                   the directory where the 
                                   SAS Data Sets are;
filename guava '/home/almaak/user12/portfile.xpt';
proc cport data=olddata.mydata1 file=guava;

NOTE: as usual, if the file portfile.xpt is in the same directory where you invoked SAS, you don't need the path shown in the example. Also, you can substitute ~/ for your home directory in any path expression.

Example 2: (Sending from SAS for Windows)

libname olddata 'c:\sasdata';  **this is the directory
                          where the SAS Data Sets are;
filename guava 'a:portfile.xpt';
proc cport data=olddata.mydata1 file=guava;

In these examples the location of the SAS Data Sets to be transferred is designated as olddata and the file to which the transport version of the data is to be written is designated as guava, which is just a nickname (or FILEREF) for the actual file to be created on the A: drive, called portfile.xpt.

Step Two: Moving a Transport File to the Target System

Once the data are in transport format, the file may be moved by most methods appropriate for sending, copying or otherwise transferring ASCII data, provided that the sending method allows binary transfer (i.e., no translation takes place). The file may be uploaded or downloaded between the PC and any system with pc-to- host communication packages, sent via eMail or bitnet, copied over PC-NFS, ftp'd, and so forth.

Step Three: Reconverting the Data to SAS Data Sets

Example 1: (Receiving at Windows)

If the receiving system is Windows, the transport file is identified by a FILENAME statement, and the permanent library is identified by a LIBNAME statement:

libname olddata 'c:\my\sasdata';  **SAS Data Sets
                                     will be written here;
filename cumquat 'a:portfile.xpt';
proc cimport library=olddata infile=cumquat;

The file from which the transport version of the data is to be read is designated as cumquat, which is just a nickname (or FILEREF) for the actual file portfile.xpt which for this example is shown on the diskette in A:\. The location in which the SAS Data Set(s) will be written is designated as olddata. PROC CIMPORT allows specifications other than LIBRARY to designate the output SAS Data Set location and name. See the SAS documentation on PROC CIMPORT for details.

NOTE: Moving SAS data sets or catalogs from one Windows machine to another Windows machine requires no transport files. Files may simply be placed on a diskette or ZIP disk and carried or sent to the target machine. If you will be moving the files from one Windows machine to another Windows machine using eMail or ftp, however, transport files should be used anyway to avoid possible corruption.

Example 2: (Receiving at UNIX)

If the receiving system is UNIX, the transport file is identified by a FILENAME statement, and the permanent library is identified by a LIBNAME statement:

libname olddata '/home/almaak/username';  **SAS Data Sets
                                     will be written here;
filename cumquat '/home/almaak/username/portfile.xpt';
proc cimport library=olddata infile=cumquat;

This example shows file specifications in which the files and SAS data sets will reside in the "home" directory. (The "home" directory also may be represented with ~/.) The file from which the transport version of the data is to be read is designated as cumquat, which is just a nickname (or FILEREF) for the actual file previously placed in the "home" directory, and called portfile.xpt. The location in which the SAS Data Set(s) will be written is designated as olddata. PROC CIMPORT allows specifications other than LIBRARY to designate the output SAS Data Set location and name. See the SAS documentation on PROC CIMPORT for details.

Transport files created with PROC COPY

Transport files in previous versions of SAS often were created using PROC COPY or a DATA step with an XPORT (or SASV5XPT) engine on a LIBNAME statement. Files created in this way must be reconstructed using the same method.

Example 1: (Receiving at Windows, using PROC COPY):

libname  olddata  'c:\sasdata'; **SAS Data Sets
                              are written  here;
libname  pear  xport  'a:portfile.xpt';
proc copy in=pear out=olddata; select peach guava;

Example 2: (Receiving at UNIX, using PROC COPY):

libname  olddata  '/home/almaak/username'; **SAS Data Sets  
                                          are written here;
libname  pear  xport  '/home/almaak/username/portfile.xpt';
proc copy in=pear out=olddata; select peach guava;

In Examples 1 and 2, "peach" and "guava" are SAS Data Sets "peach.sd2" and "guava.sd2" that will be written to the PC subdirectory "c:\sasdata" or "peach.ssd01" and "guava.ssd01" that will be written to the UNIX root directory. The transported file is the Windows file "portfile.xpt" or the UNIX file "portfile.xpt", which is regarded as a SAS library with one or more (in this case two) members.

Example 3: (Receiving at Windows, using DATA step):

libname  olddata  'c:\sasdata'; **SAS  Data  Sets  
                              are  written   here;
libname  pear  xport  'a:portfile.xpt';
data olddata.apple; set pear.peach;

libname  olddata  '/home/almaak/username'; **SAS Data Sets  
                                          are written here;
libname  pear  xport  '/home/almaak/username/portfile.xpt';
data olddata.apple; set pear.peach;

USC Students, Faculty and Staff may use computers and documentation in the Public User Areas in Waite Phillips Hall, King Olympic Hall and Salvatori Computer Science Center, as well as the Leavey Library Information Commons (Ground Floor).

Consulting. The ITS Consultants may be familiar with the language and general operations of this software, but it may be necessary to make an appointment with a member of the full time staff in order to receive detailed help. Support of statistics software is the responsibility of the ITS Statistics Consultant with the participation of other full-time ITS staff. These people may be contacted through the ITS Customer Support Center as shown here.

Customer Support. USC students, staff or faculty who would like information about ITS Workshops or about obtaining site-licensed software or other computing-related questions should visit the Customer Support Center in Leavey Library Lower Commons, or call 213.740.5555, or send eMail to <consult@usc.edu>

Documentation. This document, and many others on a variety of topics, are available in the ITS Customer Support Help System, available on the World Wide Web at:

http://www.usc.edu/its/

You can find Statistics Software Help Documents through the search engine at this same URL, or go directly to them at:

http://www.usc.edu/its/doc/statistics/help/

Newsgroups. Another source of information, this one providing the opportunity to exchange thoughts with other users, is the newsgroup 'usc.comp.all.stat.users'. For more information about reading news, subscribing to newsgroups, and related topics, visit

  
http://www.usc.edu/its/doc/internet/news/