About kas

kas · ‎04-29-2012

I came across this thread because I was trying to work out the same thing. So, I tried it and the presentation turned out perfect. I had always avoided this method because of the ERROR generated in the NOTE in the log. Is there a way to change the message in the NOTE? My site gets all worked up when 'ERROR' appears in a log. Perhaps the NOTE could read 'NOTE: The ANNOTATE dataset is not happy with the values outside the boundaries but it is fine' Something like that?

kas · ‎11-30-2011

Yes, that is what I was looking for. I did not realize proc sql pre-dated the ANSI standards; interesting. This is good information. Is it in the documentation?

kas · ‎11-27-2011

Interesting input from everyone. Thanks all for your responses. art297 - and I think the Oracle SQL definitions are clearer and that SAS had more conflict in usage. Joeldw - yes, that documentation is helpful but not precise. In practice, after a column is assignged an alias, the alias can be used or the original column name. This is actually pretty cool (but not documented to my knowledge). The data step processing makes sense. The rename executes as the data is read and the original variable name is no longer valid. In SQL, either can be used; I see that as more in conflict than Oracle. Anyways, I think SAS would better serve its customer base by documenting the 'undocumented features' that we often encounter.

kas · ‎11-25-2011

Thank you both for you responses; I have a couple of additional questions/comments for anyone who might know the answer. Perhaps it is not clear, but this does not follow ANSI standards for SQL and so I thought SAS would have this documented. I have not been able to find any yet and this is my question - where is it documented? Someone who has experience writing SQL for Oracle, would find the interchangeability of the alias to original column peculiar. It is different than a data step because in your example (for MikeZdeb), you cannot interchange sex/gender in your where statement. The variable has been renamed upon reading and can no longer be referenced by the old value (i.e. sex no longer exists). In my example, either column (source/alias) can be identified in the where clause.

kas · ‎11-25-2011

I as surprised to find that I could use a column alias in my where statement. Example create table xxx as select one, two, three from dset where one =1 ; create table yyy as select one as flag, two, three from dset where flag = 1; Do these statements both yield these same result? If soon, is this documented somewhere. I have found these documentation for using caculated values in a where statements but not for using these column alias in these where statements.

kas · ‎05-20-2010

For those who are browsing this thread and not the OP who has found a solution. Setting the infile to 2 will not exclude any observations. Since this discussion is dealing with a csv file, the first line is the header row and the actual data lines start on line 2. You can test this for yourself to verify. The OP may have a different csv definition, not really sure why it did not work. I used it to define the csv table and saw that this became the default on the properties. Message was edited by: kas

kas · ‎05-19-2010

Agreed - it does not appear that would be the solution but it works. I'm still trying to determine the 'why' and figure out the pieces. When I was testing and defined the csv from an existing file, it defaulted to this. I'm concluding it was because ine one was selected for the header definitions on the infile. This then becomes the default for writing the file. That's all I've got; still pretty new to DI Studio. Thanks for the options statement; I knew it was an options but could not remember /find the correct one. It worked nicely.

kas · ‎05-19-2010

I was searching for a solution to this same problem and found this thread. I was able to find a partial solution. I'm using DI Studio 4.21 on a windows server. To get the header line to appear in the csv file follow these steps: open the properties of the csv table select the 'File Parameters' tab select 'Advanced' Find the 'External file records' section and change the 'Start record:' to 2 This should get you the header row. My problem is that my file is very wide and I've changed the lrecl option to match but I'm getting a warning when the header row is written. It is too wide. This is the file statement: file 'C:\files\updates\file_output.csv' lrecl=2000 dlm=','; And this is the warning: WARNING: The quoted string currently being processed has become more than 262 characters long. You may have unbalanced quotation marks. So, if your file is not too wide, then this will work. Otherwise, you are stuck with me (unless somebody can solve this part). Karen

Online Status	Offline
Date Last Visited	‎09-01-2015 07:11 AM

Re: How to create uneven tick marks on graph axis?

SQL where statement using alias column

SQL where statement using alias column

SQL where statement using alias column

SQL where statement using alias column

Re: DI Studio and external file

Re: DI Studio and external file

Re: DI Studio and external file

Re: DI Studio and external file

Re: DI Studio and external file

Re: How to create uneven tick marks on graph axis?

SQL where statement using alias column

SQL where statement using alias column

SQL where statement using alias column

SQL where statement using alias column

Re: DI Studio and external file

Re: DI Studio and external file

Re: DI Studio and external file