Hi David, I think it's a two-stage operation. When you have a single source dataset what you can do at that point is look in the navigation pane of Workbench for the 'source dataset' setting. Double-click that and you'll be presented with a dialog with two buttons.
One of these buttons is the regular selection tool [...] - you can use this to select multiple files in the same folder. The [+] button is the Swizzler - you can use it to select an entire folder or multiple files in different folders. The Swizzler is also available on the 'create workspace' dialog so you can set up a workspace like this from the very start. Once you have done this there is the risk of not having the correct Feature Types. In your case the 'directory' is classed as a 'Dataset' and each TAB file within it is a 'Feature Type'. FME uses the name of a file as the name of the Feature Type - so as it stands when you have one file added to the workspace you have only one Feature Type. Since the other files (Feature Types) have different names they won't match the one in the workspace and so won't get processed. If you run it now all of the files except the first are read but then dumped - look in the log file for the 'unexpected input remover' which tells you how many features were lost due to this problem. To solve this you could add all the files (Feature Types) as you did, but for a neat workspace I'd suggest setting a merge Feature Type filter - find this in the properties dialog for your single source Feature Type - tick the box and enter * into the filter (or any regular expression). Now all the data will be allowed to pass. You can find more information about this at... http://www.fmepedia.com/index.php/Workbench_Intermediate I'm not sure whether I'm supposed to mention the fmepedia ahead of an official announcement but it's up and running and has a whole set of FAQ so go for it... Hope this helps. I know it's a long email but this is a good concept to be aware of in FME. It will help you create workspaces that are tidier and are able to process different datasets. Regards Mark Mark Ireland, Product Support Engineer Safe Software Inc. Surrey, BC, CANADA [EMAIL PROTECTED] http://www.safe.com Solutions for Spatial Data Translation, Distribution and Access --- In [email protected], "Eagle, David A" <[EMAIL PROTECTED]> wrote: > Hello... > > I have a directory that contains a number of TAB files, each one is > identical in structure and contains the same geometry but they each cover a > different geographical area and have a different name accordingly. Each file > contains a number of map objects classified by type. I have constructed a > workspace to filter out specific 'types' of object using the attributefilter > these are then route back into a single TAB file. > > Using a source file that points at the directory within which all the files > reside, my translation only filters data from the first file in the folder. > So I have now added each individual file as a source and used the > connections view to quickly link all 20 files to the transformer. That seems > to work great but I end up with a fairly untidy and maybe inefficient > workspace. > > So, how can I input feature types with exactly the same format and schema > through a single feature type definition? Is that called a Swizzler and if > so how do I do it? > > Any suggestions gratefully received... > > David > Get the maximum benefit from your FME, FME Objects, or SpatialDirect via our Professional Services team. Visit www.safe.com/services for details. Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/fme/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
