Look up what %include does - and just check return codes from time to time. If you want to execute a program within a data step then you might want to learn about 'call execute' You can always wrap a SAS Base program into a %macro ... %mend and then call the macro later on in your program (just link the objects in EG and have the code with %macro...%mend executed before you call the macros. HTH Patrick Message was edited by: Patrick ... View more

For point 1 and 2: Query the dictionary tables (or the corresponding views in SASHELP). The dictionary table is either 'tables' or 'columns' - never sure about this. ... View more

Use the function 'translate()' for your ID variable. If you don't know all possible special characters then a more generic solution would be possible using Regular Expressions - but it's also more difficult to code... HTH Patrick ... View more

data Vall / view=Vall; set a(in=a) b(in=b) c(in=c) d(in=d) e(in=e) f(in=f); if a then show='a'; else if b then show='b'; else if.... run; proc tabulate data=Vall; .....and then work out what list you really want. HTH Patrick ... View more

This is a SQL variant: proc sql; create table ObsAndVarsCommon as select * from Table1 intersect corr all select * from Table2 intersect corr all select * from Table3 .....and so on and so on ; quit; ... View more

Please describe how the data look like, what the problem is, and what the desired output is. ... View more

Hi FUN Following this thread I suddenly remembered that I have something in my box which needed only a bit of change to cover your requirements. Have a look at the code below. I changed your 'modate' part in the clause to 'coalesce(modate,crdate)' as I saw in the dictionary table that there are entries with a blank modate. HTH Patrick %macro CleanUpLib(CleanupLib=); options mprint; %local dropmems; proc sql noprint nowarn; select catx('.',libname,memname) into :dropmems separated by ',' from dictionary.tables where libname =%upcase("&CleanupLib") and memtype="DATA" and today()-datepart(coalesce(modate,crdate)) > 2; %if &dropmems ne %then %do; %put ************ Tables will be dropped ******; %put Dropmems=&dropmems; drop table &dropmems; %end; %else %do; %put ************ No tables to be dropped ******; %end; quit; run; %mend; %CleanUpLib(CleanupLib=MyLib) Message was edited by: Patrick ... View more

As Linus mentiones: There are bulk-load options doing the job. The only problem I've seen so far with bulk-loading to Oracle was that if a bulk-load fails Oracle doesn't always send a proper error code back to SAS - what results in a SAS job ending without errors but data not loaded... The work-around we had to implement was a little macro which analyses the bulk-load log (searching for ora-errors in the log) and aborting the SAS job if an error message is found. HTH Patrick ... View more

Hi GR This here is a SAS forum with SAS users - not the official SAS Tech Support site. Two things you could do: - Check if there is enough disk space on your server. - Restart the metadata server If this doesn't help: Contact SAS Tech Support HTH Patrick ... View more

Hi Wayne66 Cynthia gave you already lotz of interesting information. From what you tell I assume that these XML files are stored in SQL in a field with type CLOB. The first question you have to solve is how to read CLOB data >32K with SAS. Well: I couldn't find an answer to that. I was thinking about streaming but couldn't find an example. Hopefully someone has an answer to that. As Cynthia mentioned XML files are text. Would it be possible that your IT guys give you access to the XML files before they load it into SQL - or that they export these XML files later from SQL to a text file? A soon as you have these XML files as text files (that's what an XML file normally is: A text file with extension XML in a defined structure which can be interpreted by an XML parser - i.e. InternetExplorer) you can read/convert/interprete it with SAS. Cynthia pointed you in the right direction for that. As I understand it this combining of XML files is more a combining of data these XML files contain. In my understanding it wouldn't make sense to concatenate the XML files - what you might want to do is to read all the XML files one after the other and concatenate the resuls into one single SAS dataset (proc append?). Just guessing! Is it that what you're after? Cheers, Patrick As soon as ... View more

The values in the dictionary tables are upper case:....libname='TEMP'.... and it will work. ... View more

proc sql; select * from dictionary.macros; quit; OR: proc print data=sashelp.vsmacro; run; And to get what you asked for: proc sql; create table MyTable as select name as x from dictionary.macros; quit; OR: data MyTable; set sashelp.vsmacro(keep=name); rename name=x; run; Have a look at the link I've sent already: There you see the names and description of the fields in the dictionary table. sashelp.vsmacro is a view on dictionary.macros. Using PC SAS or SAS EG you can also browse sashelp.vsmacro HTH Patrick ... View more

I don't know what "prepare time series data" does but from what you tell I have the suspicion that your merge doesn't work because the keys are different. The format of your 'date variable' doesn't matter as formats are only used for printing values. A 'SAS date variable' is numeric and contains the DAYS since 1/1/1960. A 'SAS datetime variable' is also numeric and contains the SECONDS since 1/1/1960. And here could be the problem: You write about datetime. informat (so the variable should contain a number representing a SAS datetime) but then the key variable has in one case a weekdatx. format (used for date variables; days since 1/1/1960) and in the other case a datetime format (=seconds since 1/1/1960). Check the values of your keys (i.e. with 'put myvar best32.;) This would then also explain your join results: Inner Join: Null result Left Join: Only data from left table Right Join: Only data from right table If you want to create a denormalised dataset by joining weekly data with monthly ones then I assume that you have to transform your keys before merging - i.e. by: KeyForMerge=put(MyDateTimeVar,dtmonyy7.); (for datetime vars) and KeyForMerge=put(MyDateTimeVar,monyy7.); (for date vars). Alternatively you could use the function 'intnx' to transform your keys to a common basis. HTH Patrick Message was edited by: Patrick ... View more

select myvar1 length=$20, myvar2 format=$15. HTH Patrick ... View more

Hi sas@kumar please keep your posts in one forum only. It's a bit annoying to answer a question only to see later on that there was already quite a discussion going on in another forum. Cheers, Patrick ... View more