reading ACCESS

shiyiming

When reading ACCESS tables to SAS, I find that, for example,

[code:keisxull]proc import out = aaa_ datatable = “AccessTableName” dbms = access replace;
                database = "dataBase";
run;[/code:keisxull]
where AccessTableName has to be no more than 32 characters long.  
Anyone has an idea on that? How to solve this problem without changing the table name in ACCESS.

Many thanks to you,

JingJu

shiyiming

try VALIDVARNAME=ANY system option
default is VALIDVARNAME=V7, which limits to 32 alphanumeric chars

shiyiming

Thank you for the reply.
In fact, as my understanding, option ‘validVarName =’ is working on SAS variable names, but nothing to do with ACCESS table name. I am thinking if we can try other importing schemes.
JingJu

shiyiming

quote SAS manual:
"VALIDVARNAME=ANY allows any characters in DBMS column names to appear as valid characters in SAS variable names. Symbols, such as the equal sign (=) and the asterisk (*), must be contained in a 'variable-name'n construct. You must use ANY whenever you want to read DBMS column names that do not follow the SAS naming conventions."

I don't have ACCESS on my machine therefore I can't test it, but, you can give the solution a try and you will know whether it works or not.

shiyiming

I may not illustrate my question explicitly. Sorry about that. In fact, I am asking about ACCESS table name instead of variable name. I don't know how to do it thus I ask you. I did try the way (=any) as you mentioned and unfortunately it did not work. Moreover, I am not sure if ‘validvarname = any’ can increase the maximum length of the variable. I have ACCESS and I tried it, please trust me.
By the way I dropped by your blog, such amazing code you have written! Very nice.

shiyiming

okay, I see. Sorry, I don't know the answer. A better place for you is SAS-L, more SAS gurus there. I am not good at pure SAS programming details

Thanks for your complimentary words on my blog. I am preparing two SAS recipe books for modern data mining algorithms, complements, respectively, to Matrix Methods in Data Mining & Pattern Recognition, and The Elements of Statistical Learning.

If you have time, give me a vote at <!-- m --><a class="postlink" href="http://ires.ku.edu/%7Eipsr/SGF2010/royvote.htm">http://ires.ku.edu/%7Eipsr/SGF2010/royvote.htm</a><!-- m -->

shiyiming

ACCESS是高科技