The preloaded data size exceeds the profiler treshold

[Silvio Domingos]

Hi,

I'm trying to load a huge ttl fiile into semantic turkey.

See: ftp://ftp.nlm.nih.gov/online/mesh/rdf/

But I'm getting this error: The preloaded data size exceeds the profiler treshold 1048576 bytes

I did find what configuration to change or if there is a connector to load files from system.

I saw you have something to load from local file here:

https://bitbucket.org/art-uniroma2/semantic-turkey/src/master/st-core-framework/src/main/java/it/uniroma2/art/semanticturkey/ontology/impl/OntologyManagerImpl.java

But, is it the correct place to look at?

Bwt,

1) The unit tests from this issue: https://groups.google.com/forum/?utm_medium=email&utm_source=footer#!topic/semanticturkey-user/3CulrKE29Sg

Are now passing after your last(s) commit: https://bitbucket.org/art-uniroma2/semantic-turkey/commits/

Could you share exactly what was changed?

2) The last commit is nothing find the jar from  https://bitbucket.org/art-uniroma2/coda.git  in the maven central, versoin 1.7.0

Are you guys soon uploading it?

3) What do you guys think to include https://editor.swagger.io/ in the project? It can be configured direct on spring mvc.

I may open a pull request for it.

4) I'm almost finishing the docker image for the back end. I just figured out that the deployment with karaf is a little bit different.

Best Regards,

SIlvio

[Armando Stellato]

Dear Silvio,

 

concerning the error on preloaded data size:

 

1. We have in v.7.0 enabled the possibility to change the max size of data loaded through the preloader.
2. However, pls consider that this is still a feature meant to facilitate the creation of a project by inspecting the file before it is loaded into the real repository backing the project and automatically setting all fields such as baseuri, model, lexicalization model, etc..
   The preloading happens in memory and it would consume a big amount of memory with a huge file.
   So, for cases like this, I suggest to just plainly setup the information in the project manually and then load the data through the “load data” functionality.

 

Concerning the sources, most probably you accessed while we were uploading all recent stable releases of CODA, LIME API etc.. on Maven thus you didn’t find the dependency. All should be ok now.

 

Swagger…is surely very useful, but we have a code generation module in ST which, starting from the java methods, produces the service code. Swagger goes the other way round. We’ll see in the future if there’s still something we can do about it.

 

Kind Regards,

 

Armando

 

 

--
You received this message because you are subscribed to the Google Groups "semanticturkey-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to semanticturkey-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/semanticturkey-user/02ecc15f-173d-4632-b277-54fb5f27b5dd%40googlegroups.com.