Question

Esri SDE writer: How to use it compatible for SDE/Oracle and SDE/Postgresql

  • 8 August 2019
  • 2 replies
  • 10 views
mhab
Rank
Badge +9
  • mhab Contributor
  • Contributor
  • 101 replies

I have a workflow which writes data to Esri SDE (Enterprise Geodatabase) via the Geodatabase Writer.

It is dynamic and reads schema definition from a CSV file.

In the Geodatabase the target schema is already defined at FME write time. It is created beforehand in ArcCatalog by importing an XML workspace document.

This workflow produces compatible results for Personal GeoDB, FileGDB and SDE with identically configured writers for MDB, FileGDB and SDE.

Up to now I used SDE based on SQL Server or ORACLE. So far so good.

 

Now my problems start when I try the same workflow on an SDE based on Postgresql.

The symptoms I notice are:

1) In ArcCatalog: After Import of the XML workspace document all Table names are as expected (mixed case) but all column names are converted to lower case in contradiction to what happens on ORACLE or other environments.

2) Writing data gives No WARNING, NO ERROR not any hint something went wrong but all columns are empty

 

After studying Esri SDE Postgresql docu this is not really a suprise (:-(() but how is the solution for this dilemma ?

Is there any way I can achieve a compatible solution for all flavours of SDE ?

 

Thanks for any hints

Michael

 

2 replies

mhab
Rank
Badge +9

So with a lot of frustration I worked my way through this and spent several days struggling with "Featuretypes to Read" parameters in SDE readers, as well as SDE Schema readers, had to rework some duplicate detection, and had to add StringCaseChangers and BulkAttributeRenamers in order to get dynamic writing to SDE/PostgreSQL to the desired result ...

 

 

... and all this because PostgreSQL defaults to lower case identifiers ?

 

 

My colleagues working with C# on ArcObjects level don't understand my problems with this and commented "FME must have a problem here ..."

So I am done with this and have the strong opinion I had to work around some undone homeworks on the underlying interfaces !!!

 

 

I tend to blame Esri's way of SDE/PostgreSQL integration or is it PostgreSQL itself ?

Let me know if you have deeper insight.

 

 

Michael
nampreetatsafe
Rank
Userlevel 1
Badge +10

Hi Michael: Thanks for clearly outlining your challenges with working with SDE on PostgreSQL and cases. I'm sorry that you had to go through some frustrating moments while working on this!

After doing a bit of digging and consulting with some of my more seasoned colleagues, there seems to have been chatter about similar struggles in the past.

We found a ticket in our system that mirrors your exact experience, but it didn't gain enough traction then. I'll re-open the ticket and link this post to it in hopes that we can help improve the experience with working with cases in FME (internal reference: FMEENGINE-4563). If there is any progress on this ticket, we'll post updates on this thread.

It would be great to hear if there are others out there who have experienced similar frustrations... if so, feel free to post a comment here.

Reply


Community Stats

28,670
Posts
109,365
Replies
36,531
Members
Terms & ConditionsCookie settings
A product of
Safe Software respectfully acknowledges that we live, learn and work on the traditional and unceded territories of the Kwantlen, Katzie, and Semiahmoo First Nations.