Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ala - ignore processed recordedby #987

Merged
merged 2 commits into from
Dec 7, 2023
Merged

ala - ignore processed recordedby #987

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
39dec79
ala - ignore processed recordedby
Nov 21, 2023
6e927e1
Merge branch 'dev' into la-no-processed-recordedby
adam-collins Dec 7, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 4 additions & 6 deletions livingatlas/pipelines/src/main/java/au/org/ala/pipelines/beam/ClusteringPipeline.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -137,8 +137,7 @@ public void processElement(
List<String> otherCatalogNumbers =
source.getMultiValues().get(DwcTerm.otherCatalogNumbers.simpleName());

List<String> recordedBy =
source.getMultiValues().get(DwcTerm.recordedBy.simpleName());
String recordedBy = source.getStrings().get(DwcTerm.recordedBy.simpleName());

Long eventDateL = source.getLongs().get(DwcTerm.eventDate.simpleName());
String eventDate = "";
Expand All @@ -162,7 +161,7 @@ public void processElement(
.withDay(day)
.withEventDate(eventDate)
.withTypeStatus(typeStatus)
.withRecordedBy(recordedBy)
.withRecordedBy(Collections.singletonList(recordedBy))
.withFieldNumber(fieldNumber)
.withRecordNumber(recordNumber)
.withCatalogNumber(catalogNumber)
Expand Down Expand Up @@ -229,9 +228,8 @@ public void processElement(

// 3. taxonKey|year|recordedBy hashkeys
if (Strings.isNotEmpty(taxonKey) && year != null && recordedBy != null) {
for (String r : recordedBy) {
out.output(builder.withHashKey(taxonKey + "|" + year + "|" + r).build());
}
out.output(
builder.withHashKey(taxonKey + "|" + year + "|" + recordedBy).build());
}
}
}));
Expand Down
2 changes: 1 addition & 1 deletion livingatlas/pipelines/src/main/java/au/org/ala/pipelines/converters/CoreCsvConverter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,7 @@ public class CoreCsvConverter {
.addKeyTermFn(DwcTerm.institutionCode, getString(DwcTerm.institutionCode))
.addKeyTermFn(DwcTerm.recordNumber, getString(DwcTerm.recordNumber))
.addKeyTermFn(DwcTerm.basisOfRecord, getString(DwcTerm.basisOfRecord), "HumanObservation")
.addKeyTermFn(DwcTerm.recordedBy, getMultivalue(DwcTerm.recordedBy))
.addKeyTermFn(DwcTerm.recordedBy, getString(DwcTerm.recordedBy))
.addKeyTermFn(DwcTerm.occurrenceStatus, getString(DwcTerm.occurrenceStatus))
.addKeyTermFn(DwcTerm.individualCount, getInt(DwcTerm.individualCount))
.addKeyTermFn(DwcTerm.scientificName, getString(DwcTerm.scientificName))
Expand Down
8 changes: 6 additions & 2 deletions ...ngatlas/pipelines/src/main/java/au/org/ala/pipelines/transforms/IndexRecordTransform.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -198,7 +198,7 @@ public static IndexRecord createIndexRecord(
"degreeOfEstablishment"); // GBIF treats it as a JSON, but ALA needs a String which is
// defined
skipKeys.add(DwcTerm.typeStatus.simpleName());
skipKeys.add(DwcTerm.recordedBy.simpleName());
skipKeys.add(DwcTerm.recordedBy.simpleName()); // Do not use processed recordedBy
skipKeys.add(DwcTerm.identifiedBy.simpleName());
skipKeys.add(DwcTerm.preparations.simpleName());
skipKeys.add(DwcTerm.datasetID.simpleName());
Expand Down Expand Up @@ -610,7 +610,6 @@ private static void applyBasicRecord(BasicRecord br, IndexRecord.Builder indexRe
addTermWithAgentsSafely(
indexRecord, DwcTerm.identifiedByID.simpleName(), br.getIdentifiedByIds());
addMultiValueTermSafely(indexRecord, DwcTerm.typeStatus.simpleName(), br.getTypeStatus());
addMultiValueTermSafely(indexRecord, DwcTerm.recordedBy.simpleName(), br.getRecordedBy());
addMultiValueTermSafely(indexRecord, DwcTerm.identifiedBy.simpleName(), br.getIdentifiedBy());
addMultiValueTermSafely(indexRecord, DwcTerm.preparations.simpleName(), br.getPreparations());
addMultiValueTermSafely(indexRecord, DwcTerm.datasetID.simpleName(), br.getDatasetID());
Expand Down Expand Up @@ -850,6 +849,11 @@ public static Set<String> getAddedValues() {
.addAll(
BasicRecord.getClassSchema().getFields().stream()
.map(Field::name)
.filter(
name ->
!DwcTerm.recordedBy
.simpleName()
.equals(name)) // Do not use the processed recordedBy
.collect(Collectors.toList()))
.addAll(
TemporalRecord.getClassSchema().getFields().stream()
Expand Down
4 changes: 2 additions & 2 deletions ...ngatlas/pipelines/src/test/java/au/org/ala/pipelines/converters/CoreTsvConverterTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,7 +52,7 @@ public void converterTest() {
"\"raw_er_institutionCode\"", // DwcTerm.institutionCode
"\"raw_er_recordNumber\"", // DwcTerm.recordNumber
"\"br_basisOfRecord\"", // DwcTerm.basisOfRecord
"\"br_recordedBy_1|br_recordedBy_2\"", // DwcTerm.recordedBy
"\"raw_er_recordedBy\"", // DwcTerm.recordedBy
"\"br_occurrenceStatus\"", // DwcTerm.occurrenceStatus
"\"222\"", // DwcTerm.individualCount
"\"atxr_ScientificName\"", // DwcTerm.scientificName
Expand Down Expand Up @@ -940,7 +940,7 @@ public void converterDefaultTest() {
"\"raw_er_institutionCode\"", // DwcTerm.institutionCode
"\"raw_er_recordNumber\"", // DwcTerm.recordNumber
"\"HumanObservation\"", // DwcTerm.basisOfRecord
"\"\"", // DwcTerm.recordedBy
"\"raw_er_recordedBy\"", // DwcTerm.recordedBy
"\"\"", // DwcTerm.occurrenceStatus
"\"\"", // DwcTerm.individualCount
"\"\"", // DwcTerm.scientificName
Expand Down
4 changes: 1 addition & 3 deletions livingatlas/solr/conf/managed-schema
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -368,7 +368,7 @@
<field name="preparations" type="string" docValues="true" indexed="true" multiValued="true"/>
<field name="previousIdentifications" type="string" docValues="true" indexed="true"/>
<field name="recordNumber" type="string" docValues="true" indexed="true"/>
<field name="recordedBy" type="string" docValues="true" indexed="true" multiValued="true"/>
<field name="recordedBy" type="string" docValues="true" indexed="true"/>
<field name="relativeOrganismQuantity" type="double" docValues="true" indexed="true"/>
<field name="references" type="string" docValues="true" indexed="true"/>
<field name="relatedResourceID" type="string" docValues="true" indexed="true"/>
Expand Down Expand Up @@ -661,7 +661,6 @@
<field name="raw_publishingOrganizationKey" type="string" docValues="true" indexed="true" />
<field name="raw_taxonRank" type="string" docValues="true" indexed="true" />
<field name="raw_taxonRankID" type="string" docValues="true" indexed="true" />
<field name="raw_recordedBy" type="string" docValues="true" indexed="true" />
<field name="raw_recordedByID" type="string" docValues="true" indexed="true" />
<field name="raw_references" type="string" docValues="true" indexed="true" />
<field name="raw_relativeOrganismQuantity" type="string" docValues="true" indexed="true" />
Expand Down Expand Up @@ -751,7 +750,6 @@
<copyField source="recordedBy" dest="text"/>
<copyField source="identifiedBy" dest="text_identifiedBy"/>
<copyField source="recordedBy" dest="text_recordedBy"/>
<copyField source="raw_recordedBy" dest="text_recordedBy"/>
<copyField source="scientificName" dest="matched_name"/>
<copyField source="scientificName" dest="text"/>
<copyField source="species" dest="text"/>
Expand Down