msokolov commented on code in PR #12829:
URL: https://github.com/apache/lucene/pull/12829#discussion_r1401297534
##########
lucene/core/src/test/org/apache/lucene/index/TestIndexSorting.java:
##########
@@ -3173,4 +3173,184 @@ public void
testSortDocsAndFreqsAndPositionsAndOffsets() throws IOException {
reader.close();
dir.close();
}
+
+ public void testBlockIsMissingParentField() throws IOException {
+ try (Directory dir = newDirectory()) {
+ IndexWriterConfig iwc = new IndexWriterConfig(new
MockAnalyzer(random()));
+ String parentField = "parent";
+ Sort indexSort = new Sort(parentField, new SortField("foo",
SortField.Type.INT));
+ iwc.setIndexSort(indexSort);
+ try (IndexWriter writer = new IndexWriter(dir, iwc)) {
+ List<Runnable> runnabels =
+ Arrays.asList(
+ () -> {
+ IllegalArgumentException ex =
+ expectThrows(
+ IllegalArgumentException.class,
+ () -> {
+ writer.addDocuments(Arrays.asList(new Document(),
new Document()));
+ });
+ assertEquals(
+ "the last document in the block must contain a numeric
doc values field named: parent",
+ ex.getMessage());
+ },
+ () -> {
+ IllegalArgumentException ex =
+ expectThrows(
+ IllegalArgumentException.class,
+ () -> {
+ Document doc = new Document();
+ doc.add(new NumericDocValuesField("parent", 0));
+ writer.addDocuments(Arrays.asList(doc, new
Document()));
+ });
+ assertEquals(
+ "only the last document in the block must contain a
numeric doc values field named: parent",
+ ex.getMessage());
+ },
+ () -> {
+ IllegalArgumentException ex =
+ expectThrows(
+ IllegalArgumentException.class,
+ () -> {
+ writer.addDocuments(Arrays.asList(new Document()));
+ });
+ assertEquals(
+ "the last document in the block must contain a numeric
doc values field named: parent",
+ ex.getMessage());
+ });
+ Collections.shuffle(runnabels, random());
+ for (Runnable runnable : runnabels) {
+ runnable.run();
+ }
+ }
+ }
+ }
+
+ public void testIndexWithSortIsCongruent() throws IOException {
+ try (Directory dir = newDirectory()) {
+ IndexWriterConfig iwc = new IndexWriterConfig(new
MockAnalyzer(random()));
+ String parentField = "parent";
+ Sort indexSort = new Sort(parentField, new SortField("foo",
SortField.Type.INT));
+ iwc.setIndexSort(indexSort);
+ try (IndexWriter writer = new IndexWriter(dir, iwc)) {
+ Document child1 = new Document();
+ child1.add(new StringField("id", Integer.toString(1), Store.YES));
+ Document child2 = new Document();
+ child2.add(new StringField("id", Integer.toString(1), Store.YES));
+ Document parent = new Document();
+ parent.add(new StringField("id", Integer.toString(1), Store.YES));
+ parent.add(new NumericDocValuesField(parentField, 0));
Review Comment:
Is the value `0` significant in any way?
##########
lucene/core/src/test/org/apache/lucene/index/TestIndexSorting.java:
##########
@@ -3173,4 +3173,184 @@ public void
testSortDocsAndFreqsAndPositionsAndOffsets() throws IOException {
reader.close();
dir.close();
}
+
+ public void testBlockIsMissingParentField() throws IOException {
+ try (Directory dir = newDirectory()) {
+ IndexWriterConfig iwc = new IndexWriterConfig(new
MockAnalyzer(random()));
+ String parentField = "parent";
+ Sort indexSort = new Sort(parentField, new SortField("foo",
SortField.Type.INT));
+ iwc.setIndexSort(indexSort);
+ try (IndexWriter writer = new IndexWriter(dir, iwc)) {
+ List<Runnable> runnabels =
+ Arrays.asList(
+ () -> {
+ IllegalArgumentException ex =
+ expectThrows(
+ IllegalArgumentException.class,
+ () -> {
+ writer.addDocuments(Arrays.asList(new Document(),
new Document()));
+ });
+ assertEquals(
+ "the last document in the block must contain a numeric
doc values field named: parent",
+ ex.getMessage());
+ },
+ () -> {
+ IllegalArgumentException ex =
+ expectThrows(
+ IllegalArgumentException.class,
+ () -> {
+ Document doc = new Document();
+ doc.add(new NumericDocValuesField("parent", 0));
+ writer.addDocuments(Arrays.asList(doc, new
Document()));
+ });
+ assertEquals(
+ "only the last document in the block must contain a
numeric doc values field named: parent",
Review Comment:
I think this will prevent us from handling multiple levels of nested child
documents -- with a sort. I guess that would require multiple bitsets so you
can do different kinds of joins. Seems kind of involved, but I think it is
possible today?
##########
lucene/core/src/test/org/apache/lucene/index/TestIndexSorting.java:
##########
@@ -3173,4 +3173,184 @@ public void
testSortDocsAndFreqsAndPositionsAndOffsets() throws IOException {
reader.close();
dir.close();
}
+
+ public void testBlockIsMissingParentField() throws IOException {
+ try (Directory dir = newDirectory()) {
+ IndexWriterConfig iwc = new IndexWriterConfig(new
MockAnalyzer(random()));
+ String parentField = "parent";
+ Sort indexSort = new Sort(parentField, new SortField("foo",
SortField.Type.INT));
+ iwc.setIndexSort(indexSort);
+ try (IndexWriter writer = new IndexWriter(dir, iwc)) {
+ List<Runnable> runnabels =
+ Arrays.asList(
+ () -> {
+ IllegalArgumentException ex =
+ expectThrows(
+ IllegalArgumentException.class,
+ () -> {
+ writer.addDocuments(Arrays.asList(new Document(),
new Document()));
+ });
+ assertEquals(
+ "the last document in the block must contain a numeric
doc values field named: parent",
+ ex.getMessage());
+ },
+ () -> {
+ IllegalArgumentException ex =
+ expectThrows(
+ IllegalArgumentException.class,
+ () -> {
+ Document doc = new Document();
+ doc.add(new NumericDocValuesField("parent", 0));
+ writer.addDocuments(Arrays.asList(doc, new
Document()));
+ });
+ assertEquals(
+ "only the last document in the block must contain a
numeric doc values field named: parent",
+ ex.getMessage());
+ },
+ () -> {
+ IllegalArgumentException ex =
+ expectThrows(
+ IllegalArgumentException.class,
+ () -> {
+ writer.addDocuments(Arrays.asList(new Document()));
+ });
+ assertEquals(
+ "the last document in the block must contain a numeric
doc values field named: parent",
+ ex.getMessage());
+ });
+ Collections.shuffle(runnabels, random());
+ for (Runnable runnable : runnabels) {
+ runnable.run();
+ }
+ }
+ }
+ }
+
+ public void testIndexWithSortIsCongruent() throws IOException {
+ try (Directory dir = newDirectory()) {
+ IndexWriterConfig iwc = new IndexWriterConfig(new
MockAnalyzer(random()));
+ String parentField = "parent";
+ Sort indexSort = new Sort(parentField, new SortField("foo",
SortField.Type.INT));
+ iwc.setIndexSort(indexSort);
+ try (IndexWriter writer = new IndexWriter(dir, iwc)) {
+ Document child1 = new Document();
+ child1.add(new StringField("id", Integer.toString(1), Store.YES));
+ Document child2 = new Document();
+ child2.add(new StringField("id", Integer.toString(1), Store.YES));
+ Document parent = new Document();
+ parent.add(new StringField("id", Integer.toString(1), Store.YES));
+ parent.add(new NumericDocValuesField(parentField, 0));
Review Comment:
I wonder if we could use the value to indicate different parent/child
relations?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
