PyLucene >=9.12 required.

PyLucene 10 compatibility.

CHANGELOG.md

@@ -4,6 +4,7 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
 ## Unreleased
+* PyLucene >=9.12 required
 
 ## [3.2](https://pypi.org/project/lupyne/3.2/) - 2024-07-07
 ### Changed

README.md

@@ -45,7 +45,7 @@ PyLucene is not `pip` installable.
 * [Homebrew](https://brew.sh) formula: `brew install coady/tap/pylucene`
 
 ## Dependencies
-* PyLucene >=9.6
+* PyLucene >=9.12
 * strawberry-graphql (if graphql option)
 * fastapi (if rest option)
 

lupyne/engine/__init__.py

@@ -12,4 +12,4 @@
 from .indexers import IndexSearcher, MultiSearcher, IndexWriter, Indexer  # noqa
 
 version = tuple(map(int, lucene.VERSION.split('.')))
-assert version >= (9, 6), version
+assert version >= (9, 12), version

lupyne/engine/analyzers.py

@@ -7,7 +7,6 @@
 from org.apache.lucene import analysis, queryparser, search, util
 from org.apache.lucene.search import uhighlight
 from org.apache.pylucene.analysis import PythonAnalyzer, PythonTokenFilter
-from org.apache.pylucene.queryparser.classic import PythonQueryParser
 
 
 class TokenStream(analysis.TokenStream):
@@ -155,11 +154,7 @@ def parse(self, query: str, field='', op='', parser=None, **attrs) -> search.Que
             setattr(parser, name, value)
         if isinstance(parser, queryparser.classic.MultiFieldQueryParser):
             return parser.parse(parser, query)
-        try:
-            return parser.parse(query)
-        finally:
-            if isinstance(parser, PythonQueryParser):
-                parser.finalize()
+        return parser.parse(query)
 
     def highlight(self, query: search.Query, field: str, content: str, count: int = 1) -> str:
         """Return highlighted content.

lupyne/engine/documents.py

@@ -348,7 +348,7 @@ def __init__(self, searcher, scoredocs: Sequence, count=0, fields=None):
         self.searcher, self.scoredocs = searcher, scoredocs
         if hasattr(count, 'relation'):
             cls = int if count.relation == search.TotalHits.Relation.EQUAL_TO else float
-            count = cls(count.value)
+            count = cls(count.value() if lucene.VERSION.startswith('10.') else count.value)
         self.count, self.fields = count, fields
 
     def select(self, *fields: str):
@@ -450,8 +450,12 @@ def __len__(self):
     def __getitem__(self, index):
         hits = groupdocs = self.groupdocs[index]
         if isinstance(groupdocs, grouping.GroupDocs):
-            hits = Hits(self.searcher, groupdocs.scoreDocs, groupdocs.totalHits)
-            hits.value = convert(groupdocs.groupValue)
+            if lucene.VERSION.startswith('10.'):  # pragma: no cover
+                hits = Hits(self.searcher, groupdocs.scoreDocs(), groupdocs.totalHits())
+                hits.value = convert(groupdocs.groupValue())
+            else:
+                hits = Hits(self.searcher, groupdocs.scoreDocs, groupdocs.totalHits)
+                hits.value = convert(groupdocs.groupValue)
         hits.fields = self.fields
         return hits
 

lupyne/engine/indexers.py

@@ -273,7 +273,7 @@ def positions(self, name: str, value, payloads=False, offsets=False) -> Iterator
             yield doc, list(positions)
 
     def vector(self, id, field):
-        terms = self.getTermVector(id, field)
+        terms = self.termVectors().get(id, field)
         termsenum = terms.iterator() if terms else index.TermsEnum.EMPTY
         terms = map(operator.methodcaller('utf8ToString'), util.BytesRefIterator.cast_(termsenum))
         return termsenum, terms
@@ -402,12 +402,12 @@ def collector(self, count=None, sort=None, reverse=False, scores=False, mincount
         count = min(count, self.maxDoc() or 1)
         mincount = max(count, mincount)
         if sort is None:
-            return search.TopScoreDocCollector.create(count, mincount)
+            return search.TopScoreDocCollectorManager(count, mincount).newCollector()
         if isinstance(sort, str):
             sort = self.sortfield(sort, reverse=reverse)
         if not isinstance(sort, search.Sort):
             sort = search.Sort(sort)
-        return search.TopFieldCollector.create(sort, count, mincount)
+        return search.TopFieldCollectorManager(sort, count, mincount).newCollector()
 
     def search(
         self,
@@ -417,7 +417,6 @@ def search(
         reverse=False,
         scores=False,
         mincount=1000,
-        timeout=None,
         **parser,
     ) -> Hits:
         """Run query and return [Hits][lupyne.engine.documents.Hits].
@@ -432,17 +431,11 @@ def search(
             reverse: reverse flag used with sort
             scores: compute scores for candidate results when sorting
             mincount: total hit count accuracy threshold
-            timeout: stop search after elapsed number of seconds
             **parser: [parse][lupyne.engine.analyzers.Analyzer.parse]` options
         """
         query = Query.alldocs() if query is None else self.parse(query, **parser)
         results = cache = collector = self.collector(count, sort, reverse, scores, mincount)
-        counter = search.TimeLimitingCollector.getGlobalCounter()
-        if timeout is not None:
-            results = search.TimeLimitingCollector(collector, counter, int(timeout * 1000))
-        with suppress(search.TimeLimitingCollector.TimeExceededException):
-            super().search(query, results)
-            timeout = None
+        super().search(query, results)
         if isinstance(cache, search.CachingCollector):
             collector = search.TotalHitCountCollector()
             cache.replay(collector)

lupyne/engine/queries.py

@@ -1,4 +1,4 @@
-from collections.abc import Callable, Iterable, Iterator
+from collections.abc import Callable, Iterable
 import lucene  # noqa
 from java.lang import Double, Integer, Long
 from java.util import Arrays
@@ -27,7 +27,7 @@ def term(cls, name: str, value) -> 'Query':
     @classmethod
     def terms(cls, name: str, values) -> 'Query':
         """Return lucene TermInSetQuery, optimizing a SHOULD BooleanQuery of many terms."""
-        return cls(search.TermInSetQuery, name, list(map(util.BytesRef, values)))
+        return cls(search.TermInSetQuery, name, Arrays.asList(list(map(util.BytesRef, values))))
 
     @classmethod
     def boolean(cls, occur, *queries, **terms):
@@ -274,8 +274,8 @@ def __getitem__(self, id: int):
 
     class SortedSet(Sorted):
         def __getitem__(self, id: int):
-            ords: Iterator = iter(self.docvalues.nextOrd, self.docvalues.NO_MORE_ORDS)
             if self.docvalues.advanceExact(id):
+                ords = (self.docvalues.nextOrd() for _ in range(self.docvalues.docValueCount()))
                 return tuple(self.type(self.docvalues.lookupOrd(ord)) for ord in ords)
 
 

lupyne/engine/utils.py

@@ -23,7 +23,7 @@ def suppress(exception):
     """Suppress specific lucene exception."""
     try:
         yield
-    except lucene.JavaError as exc:
+    except lucene.JavaError as exc:  # pragma: no cover
         if not exception.instance_(exc.getJavaException()):
             raise
 

tests/test_engine.py

@@ -398,10 +398,6 @@ def test_grouping(tempdir, indexer, zipcodes):
     assert all(grouping.search(indexer.indexSearcher, Q.alldocs()).facets.values())
     assert len(grouping) == len(list(grouping)) > 100
     assert set(grouping) > set(facets)
-    hits = indexer.search(query, timeout=-1)
-    assert not hits and not hits.count and math.isnan(hits.maxscore)
-    hits = indexer.search(query, timeout=10)
-    assert len(hits) == hits.count == indexer.count(query) and hits.maxscore == 1.0
     directory = store.ByteBuffersDirectory()
     query = Q.term('state', 'CA')
     size = indexer.copy(directory, query)
@@ -465,9 +461,6 @@ def test_fields(indexer, constitution):
         engine.Field('', stored='invalid')
     with pytest.raises(AttributeError):
         engine.Field('', invalid=None)
-    with pytest.raises(lucene.JavaError):
-        with engine.utils.suppress(search.TimeLimitingCollector.TimeExceededException):
-            document.Field('name', 'value', document.FieldType())
     assert str(engine.Field.String('')) == str(
         document.StringField('', '', document.Field.Store.NO).fieldType()
     )