Dennis Hendriksen comments

Results 66 comments of


                                            Dennis Hendriksen

MolgenisChangePasswordFilter creates a database transaction for each servlet request it filters

@fdlk *bump*

One_to_many's with autoids cannot be imported

Absent entity identifiers are automatically generated in the data layer. The source entity stream doesn't generate any identifiers which results in two separate identifier sets being generated. As a workaround...

One_to_many's with autoids cannot be imported

We could do something like: ```java private void persistEntitiesSecondPass(EntityType entityType, Stream entities) { if (entityType.hasMappedByAttributes()) { String entityTypeId = entityType.getId(); dataService.update(entityTypeId, entities.filter(entity -> hasMappedByData(entityType, entity))); } } private boolean hasMappedByData(EntityType...

The time to commit metadata changes increases dramatically with the amount of metadata present in the system

Cannot reproduce on my local machine. What are the ```max_locks_per_transaction``` and ```shared_buffers``` values of the PostgreSQL database of this MOLGENIS instance? Should be 1024 and 1GB.

The time to commit metadata changes increases dramatically with the amount of metadata present in the system

> Is this still present in 8.2RC? Yes

Tomcat 9 chokes on data explorer attribute selection URLs

I've reported this Firefox bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1621013

Tomcat 9 chokes on data explorer attribute selection URLs

@joerivandervelde this is a configuration issue on our recent MOLGENIS VMs and can easily be fixed as described in https://github.com/bohnman/squiggly/issues/42#issuecomment-414673274. @erikschaberg recently fixed this for http://molgenis.org/gavin.

Sort order for reference columns always sorts on ID

The resulting data explorer query is of the form ```/api/v2/eu_bbmri_eric_collections?sort=biobank```. REST API v2 doesn't allow queries such as ```/api/v2/eu_bbmri_eric_collections?sort=biobank.name``` and probably it will require some work to propagate such a...

Sort order for reference columns always sorts on ID

Reported by a colleague of @cpavanrun in 8.4.1

Query of type 'in' on nullable mref incorrect results

> I suppose this is necessary for our new biobank explorer release? nope