EG should understand windows paths

[tomrvincent]

I have a drive accessible to both Linux and Windows. 

So, I should be able to use 'S:/_Pennsylvania/Claims Processing/'

But no...I have to use '/mnt/nfs/prod/Teams/statutory/REGREP/_Pennsylvania/Claims Processing/' instead.

Also, if EG can import from my F drive (a Windows only drive), I should be able to use F in a proc import statement.  Same with export.

[tomrvincent]

And what I'm asking is that EG, when it hits something like 

proc import datafile="f:\accounting\2017\test.csv"...; 

does what happens when you use the 'import data' wizard and copy the file to the work location (or some other location that SAS knows), sending the interpreted code on to the SAS engine to be run.  I don't see that as complicated.

[ChrisHemedinger]

I think we understand the request.  Responses from @Kurt_Bremser and me fall into the realm of "that's not how that works" -- as EG does not parse/process SAS code.  EG simply sends code to SAS for processing and there is no opportunity for EG to scan the code for potential "upload" actions.

But from your point of view, you would like it to work that way -- and I can understand that.  I think we've gathered enough info here to discuss with the product development team.

[tomrvincent]

And that's why I suggested it.  I know EG is mostly a shell and I didn't claim it parsed or processed the code.

Improving the handshaking between EG and the SAS engine is ripe with opportunities.  Conditional prompt screens, prompt screens with interactivity, building prompt screens 'on-the-fly', for example, are 3 ideas that come to mind.

 "that's not how that works" (as irrelevant as it is in this suggestion) should be the *starting* point of a discussion, not the *ending* point.

[Quentin]

Just for a counter-argument, I definitely wouldn't want EG to start interpreting my SAS code.  When I use EG as an IDE, I'm writing .sas files that I expect to be able to run in EG or as a stored process or as a batch job etc etc.  So if EG starts interpreting my SAS code (i.e. interpeting "c:\my documents\whatever.dat") and doing something special with it, that would cause me problems.  I wouldn't want it to magically upload a file to work.  That's fine for tasks (which I don't use : ) , but not code.

If the argument is that, say, linux SAS should be able to understand "c:\my documents\whatever.dat", I suppose that would be some interesting magic.  No clue how it might work.  We have linux directories mounted to windows via samba.  I suppose somewhere in there samba knows that "/home/quentin" is mapped on windows to "d:\Quentin"?  Or maybe not, maybe linux just makes /home/quentin available to windows but doesn't know which users are mapping to it or which windows drive letters have been assigned.  Or maybe Samba knows the UNC?  In fact, since one user might map "/home/quentin" to "D:" and another user might map "/home/quentin" to "Q:", it's even messier.  Art Carpenter had a paper with a macro to get UNCs from drive letters.  Head starting to hurt...

I like the macro variable route, standardized directory structures, relative paths, etc.

[tomrvincent]

@Quentin all I'm asking for is a way to call the import/export tasks from within code, since EG knows about my windows paths but SAS doesn't.