Skip to content

Improve search results #716

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

Open
wants to merge 9 commits into
base: master
Choose a base branch
from
Open

Improve search results #716

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
45904fb
Don't use wildcard instructions for searchable content
axlewin Aug 13, 2025
25c0314
Require stricter matching for searching by id
axlewin Aug 13, 2025
e210166
Restore slight boost to searchable content
axlewin Aug 13, 2025
b810ce6
Prioritise exact matches on id/title/subtitle
axlewin Aug 15, 2025
93ce11e
Prioritise exact matches for question finder search
axlewin Aug 19, 2025
06180ad
Add null check for QF search string
axlewin Aug 19, 2025
2b3da1c
Change access modifier
axlewin Aug 19, 2025
39780ad
Fix indentation
axlewin Aug 19, 2025
2891e46
Fix indentation
axlewin Aug 19, 2025
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
31 changes: 28 additions & 3 deletions src/main/java/uk/ac/cam/cl/dtg/segue/dao/content/GitContentManager.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -349,9 +349,20 @@ public final ResultsWrapper<ContentDTO> siteWideSearch(
// Restrict content types
.includeContentTypes(contentTypes)

// High priority matches on untokenised search string
.searchFor(new SearchInField(Constants.ID_FIELDNAME + "." +
Constants.UNPROCESSED_SEARCH_FIELD_SUFFIX, Collections.singleton(searchString))
.priority(Priority.HIGH).strategy(Strategy.SIMPLE))
.searchFor(new SearchInField(Constants.TITLE_FIELDNAME + "." +
Constants.UNPROCESSED_SEARCH_FIELD_SUFFIX, Collections.singleton(searchString))
.priority(Priority.HIGH).strategy(Strategy.SIMPLE))
.searchFor(new SearchInField(Constants.SUBTITLE_FIELDNAME + "." +
Constants.UNPROCESSED_SEARCH_FIELD_SUFFIX, Collections.singleton(searchString))
.priority(Priority.HIGH).strategy(Strategy.SIMPLE))

// Fuzzy search term matches
.searchFor(new SearchInField(Constants.ID_FIELDNAME, searchTerms)
.priority(Priority.HIGH).strategy(Strategy.FUZZY))
.priority(Priority.HIGH).strategy(Strategy.DEFAULT))
.searchFor(new SearchInField(Constants.TITLE_FIELDNAME, searchTerms)
.priority(Priority.HIGH).strategy(Strategy.FUZZY))
.searchFor(new SearchInField(Constants.SUBTITLE_FIELDNAME, searchTerms)
Expand All @@ -361,7 +372,7 @@ public final ResultsWrapper<ContentDTO> siteWideSearch(
.searchFor(new SearchInField(Constants.TAGS_FIELDNAME, searchTerms)
.priority(Priority.HIGH).strategy(Strategy.FUZZY))
.searchFor(new SearchInField(Constants.PRIORITISED_SEARCHABLE_CONTENT_FIELDNAME, searchTerms)
.priority(Priority.HIGH).strategy(Strategy.FUZZY))
.strategy(Strategy.FUZZY))
.searchFor(new SearchInField(Constants.SEARCHABLE_CONTENT_FIELDNAME, searchTerms)
.strategy(Strategy.FUZZY))

Expand Down Expand Up @@ -451,10 +462,24 @@ public final ResultsWrapper<ContentDTO> questionSearch(
.searchFor(new SearchInField(Constants.TAGS_FIELDNAME, searchTerms)
.priority(Priority.HIGH).strategy(Strategy.SUBSTRING))
.searchFor(new SearchInField(Constants.PRIORITISED_SEARCHABLE_CONTENT_FIELDNAME, searchTerms)
.priority(Priority.HIGH).strategy(Strategy.SUBSTRING))
.strategy(Strategy.SUBSTRING))
.searchFor(new SearchInField(Constants.SEARCHABLE_CONTENT_FIELDNAME, searchTerms)
.strategy(Strategy.SUBSTRING));

if (searchString != null && !searchString.isBlank()) {
// High priority matches on untokenised search string
searchInstructionBuilder
.searchFor(new SearchInField(Constants.ID_FIELDNAME + "." +
Constants.UNPROCESSED_SEARCH_FIELD_SUFFIX, Collections.singleton(searchString))
.priority(Priority.HIGH).strategy(Strategy.SIMPLE))
.searchFor(new SearchInField(Constants.TITLE_FIELDNAME + "." +
Constants.UNPROCESSED_SEARCH_FIELD_SUFFIX, Collections.singleton(searchString))
.priority(Priority.HIGH).strategy(Strategy.SIMPLE))
.searchFor(new SearchInField(Constants.SUBTITLE_FIELDNAME + "." +
Constants.UNPROCESSED_SEARCH_FIELD_SUFFIX, Collections.singleton(searchString))
.priority(Priority.HIGH).strategy(Strategy.SIMPLE));
}

// FIXME: Make this and PageFacade agnostic
// It doesn't need to know about books, just have required tags
// It doesn't need to know about subject, field or topic, just have atLeastOne tags
Expand Down
2 changes: 1 addition & 1 deletion src/main/java/uk/ac/cam/cl/dtg/segue/etl/ElasticSearchIndexer.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,7 +58,7 @@ class ElasticSearchIndexer extends ElasticSearchProvider {
@Inject
public ElasticSearchIndexer(RestHighLevelClient searchClient) {
super(searchClient);
rawFieldsListByType.put("content", Lists.newArrayList("id", "title"));
rawFieldsListByType.put("content", Lists.newArrayList("id", "title", "subtitle"));
rawFieldsListByType.put("school", Lists.newArrayList("urn"));
nestedFieldsByType.put("content", Lists.newArrayList("audience"));
}
Expand Down
33 changes: 19 additions & 14 deletions src/main/java/uk/ac/cam/cl/dtg/segue/search/IsaacSearchInstructionBuilder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,14 +50,16 @@ public class IsaacSearchInstructionBuilder {
private static final float PRIORITY_CONTENT_BOOST = 5L;

private List<SearchInField> searchesInFields;
public static final Long NO_BOOST = 1L;
private static final Long NO_BOOST = 1L;
private static final Long FIELD_BOOST = 5L;
private static final Long FIELD_BOOST_FUZZY = 1L;
private static final Long WILDCARD_FIELD_BOOST = 1L;

private static final Long HIGH_PRIORITY_FIELD_BOOST = 10L;
private static final Long HIGH_PRIORITY_FIELD_BOOST_FUZZY = 3L;
private static final Long HIGH_PRIORITY_WILDCARD_FIELD_BOOST = 2L;
private static final Long HIGH_PRIORITY_FIELD_BOOST_FUZZY = 8L;
private static final Long HIGH_PRIORITY_WILDCARD_FIELD_BOOST = 5L;

private static final Long SEARCHABLE_CONTENT_FIELD_BOOST = 2L;

private static final Long EVENT_ADDRESS_FIELD_BOOST = 3L;
private static final Long EVENT_ADDRESS_FIELD_BOOST_FUZZY = 1L;
Expand Down Expand Up @@ -359,25 +361,28 @@ private void addSearchesInFieldsToInstruction(final BooleanInstruction instructi

} else if (searchInField.getStrategy() == Strategy.SUBSTRING) {
Long boost = searchInField.getPriority() == Priority.HIGH
? HIGH_PRIORITY_FIELD_BOOST : FIELD_BOOST;
? HIGH_PRIORITY_WILDCARD_FIELD_BOOST : WILDCARD_FIELD_BOOST;

generatedSubInstructions.add(
new MatchInstruction(searchInField.getField(), term, boost, false));
generatedSubInstructions.add(
new WildcardInstruction(searchInField.getField(), "*" + term + "*", boost));

} else if (searchInField.getStrategy() == Strategy.FUZZY) {
Long boost = searchInField.getPriority() == Priority.HIGH
? HIGH_PRIORITY_WILDCARD_FIELD_BOOST : WILDCARD_FIELD_BOOST;

boolean isSearchableContentField = searchInField.getField().equals(Constants.SEARCHABLE_CONTENT_FIELDNAME) ||
searchInField.getField().equals(Constants.PRIORITISED_SEARCHABLE_CONTENT_FIELDNAME);
Long boost = isSearchableContentField
? SEARCHABLE_CONTENT_FIELD_BOOST : searchInField.getPriority() == Priority.HIGH
? HIGH_PRIORITY_WILDCARD_FIELD_BOOST : WILDCARD_FIELD_BOOST;
generatedSubInstructions.add(new MatchInstruction(searchInField.getField(), term, boost, true));
generatedSubInstructions.add(
new WildcardInstruction(searchInField.getField(), "*" + term + "*", boost));
// Use a multi-match instruction, and ensure multi-match instructions for a particular term are
// grouped together
multiMatchSearchesGroupedByTerm.putIfAbsent(term, Sets.newHashSet());
multiMatchSearchesGroupedByTerm.get(term).add(searchInField.getField());

if (!isSearchableContentField) {
generatedSubInstructions.add(new WildcardInstruction(searchInField.getField(), "*" + term + "*", boost));
// Use a multi-match instruction, and ensure multi-match instructions for a
// particular term are grouped together
multiMatchSearchesGroupedByTerm.putIfAbsent(term, Sets.newHashSet());
multiMatchSearchesGroupedByTerm.get(term).add(searchInField.getField());
}

} else if (searchInField.getStrategy() == Strategy.SIMPLE) {
Long boost = searchInField.getPriority() == Priority.HIGH
? HIGH_PRIORITY_FIELD_BOOST : NO_BOOST;
Expand Down
Loading