Features/veen 3 loading tag types #72
base: projects/VEEN
Are you sure you want to change the base?
Conversation
gijskant
commented
Apr 12, 2016
- Exporting browse tags.
- Loading i2b2 tags.
- Conversion script from browse tag types to i2b2 tag types.
I just had a cursory look, but for now a few comments:
|
void writeHeader(Writer writer) throws IOException { | ||
writer.write(['concept_key', 'tag_title', 'tag_description', 'index'].join('\t')) | ||
} | ||
} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In Groovy this is done with fieldExtractor: { ['\\', it.value.type.displayName, it.value.description, 0] } as FieldExtractor
and headerCallback: { it.write(['concept_key', 'tag_title', 'tag_description', 'index'].join('\t')) } as FlatFileHeaderCallback
.
Plus, there's no need to have a wrapper class just for instantiating a FlatFileItemWriter
. Either do this instantiation directly in the @Configuration
class, or, if you really want to have a separate class, implement FactoryBean
.
Please rebase on master, as it has a fix for a connection leak on Oracle which causes spurious failures. |
bd1de97
to
ffda9d1
Compare
|
50de07b
to
8511629
Compare
@cataphract Comments addressed. Build passes for Postgres, not for Oracle, because there are no Oracle table definitions for tag types in transmart-data. |
@Override | ||
void write(List<? extends TagType> items) throws Exception { | ||
insertTagTypeService.deleteAllTagsForTagTypes() | ||
insertTagTypeService.deleteAllTagTypes() |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@gijskant The cleanup of the tables in this place doesn't look right. The write
method would be executed multiple times (for each chunk) during the upload. At the end only items inserted at the last chunk would remain.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Okay, I'll make it a separate step/tasklet.
BTW Coverage report is here: https://codecov.io/github/thehyve/transmart-batch/commit/851162933c82e4a64421b3838edb593efcdc41d4 It's not been posted automaticaly because of the build failure. There's a bunch of uncovered code in |
It seems you only allow type |
Also the convention I used for the field names are that they are lowercase except for predefined ones like the catch-all |
@@ -0,0 +1,3 @@ | |||
node_type title solr_field_name value_type shown_if_empty values index | |||
STUDY Test tag TEST_TAG ANALYZED_STRING Y Test option 1,Test option 2,Test option 3 1 | |||
STUDY Programming language PROGRAMMING_LANGUAGE ANALYZED_STRING N Java,C,R,Python,Javascript,Pascal,Haskell 2 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Rename this tsv.
c362a54
to
30cf873
Compare
5f7a3e3
to
36d0612
Compare
Upload of log transformed data points.
+ Some minor code improvements
428f001
to
48243a9
Compare
Age will be floor-ed for storing it in PatientDimension (just as we humans do). Age as Observation is unchanged.
…or-patientdimension Ft 1819 truncate/round age down for patientdimension
224415a
to
c3c40fc
Compare
miRNA data upload pipeline implementation.
Example of encoded path: Biomart_Data+MRNA (_ => space, + => \) - Reuse this parsing from two places: column mapping file and subject sample mapping file read. - We do not implicitly drop empty node names anymore (e.g. Biomart_Data\\MRNA\Lung; two slashes points to the empty node name) - It's alowed now to do not start ConceptFragment string with slash. The complete ConceptPath still has to start with slash.
c3c40fc
to
ebafd72
Compare
ebafd72
to
28fe984
Compare