Merging from FAIR Data Station #2017
Conversation
| @@ -0,0 +1,14 @@ | |||
| <%= render :partial => "general/page_title",:locals=>{:title=>"Update from FAIR Data Station for #{ t('investigation') } #{link_to(@investigation.title, @investigation)}".html_safe} %> | |||
There was a problem hiding this comment.
use new style, not :x =>
| @@ -1,4 +1,24 @@ | |||
| <%= render :partial => "general/page_title",:locals=>{:title=>"Import ISA from FAIRData Station for #{ t('Project') } #{link_to(@project.title, @project)}".html_safe} %> | |||
| <%= render :partial => "general/page_title",:locals=>{:title=>"Import ISA from FAIR Data Station for #{ t('Project') } #{link_to(@project.title, @project)}".html_safe} %> | |||
| @@ -33,7 +53,14 @@ def reset_data_file_cache | |||
| @data_file_cache = {} | |||
| end | |||
|
|
|||
| def build_assay(datastation_assay, contributor, policy, projects, sample, study) | |||
| def preload_data_file_cache(data_files) | |||
There was a problem hiding this comment.
question: why do you need to do this?
There was a problem hiding this comment.
data files can be connected in multiple places, so may be encountered more than once. The cache is used to preven't multiple duplicates of the data files being created. For the import this is built up whilst as they are created, but for the update the cache needs preloading, otherwise new ones will start being created again.
| investigation | ||
| end | ||
|
|
||
| def update_isa(investigation, datastation_inv, contributor, projects, policy) |
There was a problem hiding this comment.
I understand that the current update_isa method only supports fair_data_station resource types derived from SEEK resources. In the future, could this method be adapted for use with SEEK ISA elements as well? A good use case could be creating and updating a whole ISA through the API.
There was a problem hiding this comment.
I believe that once SEEK resources have an external_identifier, we’ll be one step closer to enabling resource migration between different SEEK instance
There was a problem hiding this comment.
yes, although thinking more in terms of supporting RO-Crates. The FAIR data station TTL is very similar to what an RO-Crate may look like.
For SEEK to SEEK migration the UUID's may be safer to use than the external identifiers, it's what they were originally added for.
thanks. Yes, the UI of the results is something I also think will be needed, but as a later task. I'm also not sure how much this will be used by a human rather than via an API. |
Supports updating Investigation->Study->ObsUnit->Sample->Assays with new information from FAIR Data Station.
Supports
It doesn't currently support deleting items, as there may be cases where things are being updated from a new template with partial information (like a PATCH). An option to delete may be added in the future.
There is some future work needed to improve error handling and reporting, to be treated separately