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

[ALS-4947] Add anyRecordOfMulti field to the query object #77

Merged
merged 4 commits into from
Aug 30, 2023
Merged

[ALS-4947] Add anyRecordOfMulti field to the query object #77

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
5912b01
ALS-4947: Change filter logic to be AND not OR between 2 any record o…
ramari16 Aug 24, 2023
5bdb474
ALS-4947: Allow multiple sets of anyRecordOf filters
ramari16 Aug 25, 2023
b71a5af
ALS-4947: Minor cleanup
ramari16 Aug 28, 2023
ff5f762
ALS-4947: Removed redundant merging handled by getter
ramari16 Aug 28, 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
16 changes: 14 additions & 2 deletions client-api/src/main/java/edu/harvard/hms/dbmi/avillach/hpds/data/query/Query.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,6 +37,7 @@ public Query(Query query) {
private List<String> fields = new ArrayList<>();
private List<String> requiredFields = new ArrayList<>();
private List<String> anyRecordOf = new ArrayList<>();
private List<List<String>> anyRecordOfMulti = new ArrayList<>();
private Map<String, DoubleFilter> numericFilters = new HashMap<>();
private Map<String, String[]> categoryFilters = new HashMap<>();
private List<VariantInfoFilter> variantInfoFilters = new ArrayList<>();
Expand All @@ -62,6 +63,14 @@ public List<String> getRequiredFields() {
public List<String> getAnyRecordOf() {
return anyRecordOf;
}
public List<List<String>> getAnyRecordOfMulti() {
return anyRecordOfMulti;
}
public List<List<String>> getAllAnyRecordOf() {
srpiatt marked this conversation as resolved.
Show resolved Hide resolved
List<List<String>> anyRecordOfMultiCopy = new ArrayList<>(anyRecordOfMulti);
anyRecordOfMultiCopy.add(anyRecordOf);
return anyRecordOfMultiCopy;
}

public Map<String, DoubleFilter> getNumericFilters() {
return numericFilters;
Expand Down Expand Up @@ -98,6 +107,9 @@ public void setRequiredFields(Collection<String> requiredFields) {
public void setAnyRecordOf(Collection<String> anyRecordOf) {
this.anyRecordOf = anyRecordOf != null ? new ArrayList<>(anyRecordOf) : new ArrayList<>();
}
public void setAnyRecordOfMulti(Collection<List<String>> anyRecordOfMulti) {
this.anyRecordOfMulti = anyRecordOfMulti != null ? new ArrayList<>(anyRecordOfMulti) : new ArrayList<>();
}

public void setNumericFilters(Map<String, DoubleFilter> numericFilters) {
this.numericFilters = numericFilters != null ? new HashMap<>(numericFilters) : new HashMap<>();
Expand Down Expand Up @@ -191,7 +203,7 @@ public String toString() {
writePartFormat("Numeric filters", numericFilters, builder);
writePartFormat("Category filters", categoryFilters, builder);
writePartFormat("Variant Info filters", variantInfoFilters, builder, false);
writePartFormat("Any-Record-Of filters", anyRecordOf, builder, true);
writePartFormat("Any-Record-Of filters", getAllAnyRecordOf(), builder, true);

return builder.toString();
}
Expand Down Expand Up @@ -234,7 +246,7 @@ private static void showTopLevelValues(Collection varList, StringBuilder builder

Integer count = countMap.get(firstLevel);
if(count == null) {
count = new Integer(1);
count = 1;
} else {
count = count + 1;
}
Expand Down
49 changes: 25 additions & 24 deletions ...essing/src/main/java/edu/harvard/hms/dbmi/avillach/hpds/processing/AbstractProcessor.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
import java.util.Map.Entry;
import java.util.concurrent.*;
import java.util.stream.Collectors;
import java.util.stream.Stream;
import java.util.zip.GZIPInputStream;

import com.google.common.util.concurrent.UncheckedExecutionException;
Expand Down Expand Up @@ -195,24 +196,27 @@ protected Set<Integer> applyBooleanLogic(List<Set<Integer>> filteredIdSets) {
* @return
*/
protected List<Set<Integer>> idSetsForEachFilter(Query query) {
ArrayList<Set<Integer>> filteredIdSets = new ArrayList<Set<Integer>>();
final ArrayList<Set<Integer>> filteredIdSets = new ArrayList<>();

try {
addIdSetsForAnyRecordOf(query, filteredIdSets);
query.getAllAnyRecordOf().forEach(anyRecordOfFilterList -> {
addIdSetsForAnyRecordOf(anyRecordOfFilterList, filteredIdSets);
});
addIdSetsForRequiredFields(query, filteredIdSets);
addIdSetsForNumericFilters(query, filteredIdSets);
addIdSetsForCategoryFilters(query, filteredIdSets);
} catch (InvalidCacheLoadException e) {
log.warn("Invalid query supplied: " + e.getLocalizedMessage());
filteredIdSets.add(new HashSet<Integer>()); // if an invalid path is supplied, no patients should match.
filteredIdSets.add(new HashSet<>()); // if an invalid path is supplied, no patients should match.
}

//AND logic to make sure all patients match each filter
if(filteredIdSets.size()>1) {
filteredIdSets = new ArrayList<Set<Integer>>(List.of(applyBooleanLogic(filteredIdSets)));
List<Set<Integer>> processedFilteredIdSets = new ArrayList<>(List.of(applyBooleanLogic(filteredIdSets)));
return addIdSetsForVariantInfoFilters(query, processedFilteredIdSets);
} else {
return addIdSetsForVariantInfoFilters(query, filteredIdSets);
}

return addIdSetsForVariantInfoFilters(query, filteredIdSets);
}

/**
Expand Down Expand Up @@ -260,22 +264,19 @@ private void addIdSetsForRequiredFields(Query query, ArrayList<Set<Integer>> fil
}
}

private void addIdSetsForAnyRecordOf(Query query, ArrayList<Set<Integer>> filteredIdSets) {
if(!query.getAnyRecordOf().isEmpty()) {
Set<Integer> patientsInScope = new ConcurrentSkipListSet<Integer>();
VariantBucketHolder<VariantMasks> bucketCache = new VariantBucketHolder<VariantMasks>();
query.getAnyRecordOf().parallelStream().forEach(path->{
if(patientsInScope.size()<Math.max(
phenotypeMetaStore.getPatientIds().size(),
variantService.getPatientIds().length)) {
if(VariantUtils.pathIsVariantSpec(path)) {
addIdSetsForVariantSpecCategoryFilters(new String[]{"0/1","1/1"}, path, patientsInScope, bucketCache);
} else {
patientsInScope.addAll(getCube(path).keyBasedIndex());
}
private void addIdSetsForAnyRecordOf(List<String> anyRecordOfFilters, ArrayList<Set<Integer>> filteredIdSets) {
if(!anyRecordOfFilters.isEmpty()) {
VariantBucketHolder<VariantMasks> bucketCache = new VariantBucketHolder<>();
Set<Integer> anyRecordOfPatientSet = anyRecordOfFilters.parallelStream().flatMap(path -> {
if (VariantUtils.pathIsVariantSpec(path)) {
TreeSet<Integer> patientsInScope = new TreeSet<>();
addIdSetsForVariantSpecCategoryFilters(new String[]{"0/1", "1/1"}, path, patientsInScope, bucketCache);
return patientsInScope.stream();
} else {
return (Stream<Integer>) getCube(path).keyBasedIndex().stream();
Comment on lines -263 to +276
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This used to add all patients matching any path to patientsInScope which is a hacky way of doing OR.

The new logic creates a set with all patients matching any of the paths (i.e., they have any record of that path) and adds it to filteredIdSets, which is a list of sets that eventually get ANDed with each other. We can then call this method multiple times with multiple different lists of anyRecordOf filters to support AND/OR functionality in this very specific way.

}
});
filteredIdSets.add(patientsInScope);
}).collect(Collectors.toSet());
filteredIdSets.add(anyRecordOfPatientSet);
}
}

Expand All @@ -289,9 +290,9 @@ private void addIdSetsForNumericFilters(Query query, ArrayList<Set<Integer>> fil

private void addIdSetsForCategoryFilters(Query query, ArrayList<Set<Integer>> filteredIdSets) {
if(!query.getCategoryFilters().isEmpty()) {
VariantBucketHolder<VariantMasks> bucketCache = new VariantBucketHolder<VariantMasks>();
Set<Set<Integer>> idsThatMatchFilters = (Set<Set<Integer>>)query.getCategoryFilters().entrySet().parallelStream().map(entry->{
Set<Integer> ids = new TreeSet<Integer>();
VariantBucketHolder<VariantMasks> bucketCache = new VariantBucketHolder<>();
Set<Set<Integer>> idsThatMatchFilters = query.getCategoryFilters().entrySet().parallelStream().map(entry->{
Set<Integer> ids = new TreeSet<>();
if(VariantUtils.pathIsVariantSpec(entry.getKey())) {
addIdSetsForVariantSpecCategoryFilters(entry.getValue(), entry.getKey(), ids, bucketCache);
} else {
Expand Down