[jira] [Created] (ATLAS-2992) Rdbms model

[charles shen (JIRA)]

charles shen created ATLAS-2992:
---

 Summary: Rdbms model
 Key: ATLAS-2992
 URL: https://issues.apache.org/jira/browse/ATLAS-2992
 Project: Atlas
  Issue Type: Bug
  Components:  atlas-core
Affects Versions: 1.0.0
Reporter: charles shen


When creating a rdbms_db with referenced entity rdbms_intance, both rdbms_db 
and rdbms_instance will be created, but when creating another rdbms_db with the 
same rdbms_instance, the rdbms_instance will be treated as update though they 
are indeed the same. Looking into the rdbms model, rdbms_instance has attribute 
"databases" and rdbms_db has attribute "instance" constrained with inverseref 
to "databases" attribute to rdbms_instance.

So for first creation, rdbms_db will populate the "databases" attribute for 
rdbms_instance, for the second creation, altas will consider the rdbms_instance 
not equal since the "databases" attributes are different. This sounds like a 
model bug.

 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

[jira] [Created] (ATLAS-2991) Create/update behavior inconsistent between super types and sub types with same unique attribute

[charles shen (JIRA)]

charles shen created ATLAS-2991:
---

 Summary: Create/update behavior inconsistent between super types 
and sub types with same unique attribute
 Key: ATLAS-2991
 URL: https://issues.apache.org/jira/browse/ATLAS-2991
 Project: Atlas
  Issue Type: Bug
  Components:  atlas-core
Affects Versions: 1.0.0
Reporter: charles shen


Create/update behavior inconsistent between super types and sub types with same 
unique attribute.

If creating a super type first, then create a sub type with the same unique 
attribute, there will be two entities created. But if creating a sub type 
first, then create a super type with the same unique attribute, it's an update 
to the existing sub type. Why is this designed?



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Re: Review Request 69511: Atlas OMRS Connector and Event Mapper

[David Radley]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/69511/#review211091
---




open-metadata/README.md
Lines 19 (patched)


I was expecting the instructions on how to configure and run the Atlas 
connector to be in this readme.



open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/eventmapper/AtlasOMRSRepositoryEventMapper.java
Lines 64 (patched)


I am wondering what happens if you run 2 Atlas instances on the same 
machine - I assume we would need to have the ability to specify different topic 
names.



open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/AtlasEntityMapper.java
Lines 194 (patched)


remove?



open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/AtlasEntityMapper.java
Lines 344 (patched)


remove?



open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/LocalAtlasOMRSMetadataCollection.java
Lines 18 (patched)


License here twice



open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/LocalAtlasOMRSMetadataCollection.java
Lines 474 (patched)


Shouldn't we catch and throw the specific exception types? This pattern is 
used in all the methods. I am wondering why we need this try catch at all, I 
wonder if we should remove all these try catches; it seems to be there to just 
add debug tracing, if we are going to trace shouldn't we be putting out error 
tracing for the error - or is this done elsewhere? Similar to method 
processAtlasClassificationDef.



open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/LocalAtlasOMRSMetadataCollection.java
Lines 11945 (patched)


I am not sure what the CTOR is



open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/LocalAtlasOMRSMetadataCollection.java
Lines 15188 (patched)


is this right it looks like it is processing as a Boolean? Most of the 
other debug steing do not look right in this method.



open-metadata/src/main/java/org/apache/atlas/openmetadata/admin/server/spring/OpenMetadataAdminResource.java
Lines 429 (patched)


Should we remove this ?



open-metadata/src/main/java/org/apache/atlas/openmetadata/admin/server/spring/OpenMetadataAdminResource.java
Lines 454 (patched)


should we remove this?



open-metadata/src/main/java/org/apache/atlas/openmetadata/admin/server/spring/OpenMetadataAdminResource.java
Lines 477 (patched)


remove?



open-metadata/src/main/java/org/apache/atlas/openmetadata/admin/server/spring/OpenMetadataAdminResource.java
Lines 496 (patched)


remove?



open-metadata/src/main/java/org/apache/atlas/openmetadata/admin/server/spring/OpenMetadataAdminResource.java
Lines 516 (patched)


remove?


- David Radley


On Dec. 5, 2018, 3:52 p.m., Graham Wallis wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/69511/
> ---
> 
> (Updated Dec. 5, 2018, 3:52 p.m.)
> 
> 
> Review request for atlas.
> 
> 
> Repository: atlas
> 
> 
> Description
> ---
> 
> ATLAS-1773: Atlas OMRS conenctor and event mapper
> 
> 
> Diffs
> -
> 
>   open-metadata/README.md PRE-CREATION 
>   open-metadata/pom.xml PRE-CREATION 
>   
> open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/eventmapper/AtlasOMRSRepositoryEventMapper.java
>  PRE-CREATION 
>   
> open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/eventmapper/AtlasOMRSRepositoryEventMapperProvider.java
>  PRE-CREATION 
>   
> open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/AtlasAttributeDefMapper.java
>  PRE-CREATION 
>   
> open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/AtlasAttributeMapper.java
>  PRE-CREATION 
>   
> open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/AtlasBaseTypeDefMapper.java
>  PRE-CREATION 
>   
> open-metadata/src/main/java/org/apache/atlas/openmetadata/adapters/repositoryconnector/Atl