Bulk / auto connecting Reader to writer

Found it, define a sample writer, then select all your readers and right click / "Duplicate on writer" But I have still 139 autoconnect clicks + lowercase! Can't upload image in comment.

itay wrote:

my suggestion would be to avoid all the feature types by using the dynamic mode when creating the workspace.

This is great indeed. But it's a bit like a black box to me. But it works, after you create the corresponding schema in your writer db.

That's what I obtain by letting it fully automatic. You just have to create the corresponding schema on your db writer. Static workspace generation is great too.

 

 

There's no way to lowercase but that's an other issue over here:

 

https://knowledge.safe.com/questions/1785/lowercase-postgis-table-names.html

 

mark2atsafe wrote:

For connecting feature types, try the Feature Type Connections window (View > Windows > Feature Type Connections)

For connecting upper-lower case attributes, try the BulkAttributeRenamer.

Hope that helps.

Ty!

 

For lowercase feature types like table and schema names, there's an other issue there https://knowledge.safe.com/questions/1785/lowercase-postgis-table-names.html

ponceta wrote:

This is great indeed. But it's a bit like a black box to me. But it works, after you create the corresponding schema in your writer db.

That's what I obtain by letting it fully automatic. You just have to create the corresponding schema on your db writer. Static workspace generation is great too.

 

 

There's no way to lowercase but that's an other issue over here:

 

https://knowledge.safe.com/questions/1785/lowercase-postgis-table-names.html

 

You can lowercase your target table and attribute names, but you need to first include a SchemaSetter then a StringCaseChanger to set "fme_feature_type" to lower case. On the output feature type, set schema source to "Schema from schema feature". Works for me.

 

 

ponceta wrote:

This is great indeed. But it's a bit like a black box to me. But it works, after you create the corresponding schema in your writer db.

That's what I obtain by letting it fully automatic. You just have to create the corresponding schema on your db writer. Static workspace generation is great too.

 

 

There's no way to lowercase but that's an other issue over here:

 

https://knowledge.safe.com/questions/1785/lowercase-postgis-table-names.html

 

You can create the output schema beforehand, but the is not necessary since FME will act according to the settings in the writer and it's feature type and can be used to create the schema prior to writing.

david_r wrote:

You can lowercase your target table and attribute names, but you need to first include a SchemaSetter then a StringCaseChanger to set "fme_feature_type" to lower case. On the output feature type, set schema source to "Schema from schema feature". Works for me.

 

 

This seems to work but for me it won't since it generates some python issues on table names :

 

 

 

 

I must probably update my version of FME to get the last builds:

 

But this is strange since table names have no accents and it works for the first entities.

itay wrote:

You can create the output schema beforehand, but the is not necessary since FME will act according to the settings in the writer and it's feature type and can be used to create the schema prior to writing.

How? I've not found this option in the writer feature type options.

 

 

ponceta wrote:

This seems to work but for me it won't since it generates some python issues on table names :

 

 

 

 

I must probably update my version of FME to get the last builds:

 

But this is strange since table names have no accents and it works for the first entities.

You probably have non-ascii characters in an attribute name somewhere. Yes, maybe try using the latest FME 2017 RC.

ponceta wrote:

How? I've not found this option in the writer feature type options.

 

 

See tanle handling :

 

https://docs.safe.com/fme/html/FME_Desktop_Documentation/FME_ReadersWriters/postgis/format_parameters_w.htm

ponceta wrote:

This seems to work but for me it won't since it generates some python issues on table names :

 

 

 

 

I must probably update my version of FME to get the last builds:

 

But this is strange since table names have no accents and it works for the first entities.

I got it to work by using FME 2017 and setting the Python compatibility to version 3.4+. This is becuase Python 3+ has vastly improved support for international characters (unicode, more specifically):

 

ponceta wrote:

How? I've not found this option in the writer feature type options.

 

 

Did not find where. For tables its obvious but automatic schema creation no idea how to do it.

ponceta wrote:

This seems to work but for me it won't since it generates some python issues on table names :

 

 

 

 

I must probably update my version of FME to get the last builds:

 

But this is strange since table names have no accents and it works for the first entities.

FME 2017, python 3.4+ and still errors like this amazing one:

(when you are in fully automated workspace <3)

david_r wrote:

I got it to work by using FME 2017 and setting the Python compatibility to version 3.4+. This is becuase Python 3+ has vastly improved support for international characters (unicode, more specifically):

 

That's not a Python error, that's a database error, your string literal is too long (>255 characters).