More 10.0.x updates.

Author: hscells

install_pylucene.sh

@@ -33,9 +33,7 @@ export PATH="$PATH:$(pwd)/ant/bin"
 # https://lucene.apache.org/pylucene/jcc/install.html
 cd pylucene
 pushd jcc
-# IMPORTANT: Need java 17 (21 for pylucene v10.0.0)
-#export JCC_INCLUDES=/opt/homebrew/Cellar/openjdk@17/17.0.15/libexec/openjdk.jdk/Contents/Home/include:/opt/homebrew/Cellar/openjdk@17/17.0.15/libexec/openjdk.jdk/Contents/Home/include/darwin
-export JCC_INCLUDES=/opt/homebrew/Cellar/openjdk@21/21.0.9/libexec/openjdk.jdk/Contents/Home/include:/opt/homebrew/Cellar/openjdk@21/21.0.9/libexec/openjdk.jdk/Contents/Home/include/darwin
+export JCC_INCLUDES=/opt/homebrew/Cellar/openjdk@21/21.0.10/libexec/openjdk.jdk/Contents/Home/include:/opt/homebrew/Cellar/openjdk@21/21.0.10/libexec/openjdk.jdk/Contents/Home/include/darwin
 
 uv run setup.py build
 uv run setup.py install

pyproject.toml

@@ -13,6 +13,7 @@ dependencies = [
     "furo>=2024.8.6",
     "ir-datasets>=0.5.11",
     "ir-measures>=0.3.7",
+    "jcc>=3.15",
     "jupyter>=1.1.1",
     "jupyterlab>=4.4.4",
     "lucene",
@@ -41,3 +42,4 @@ dev = [
 
 [tool.uv.sources]
 pybool-ir = { workspace = true }
+lucene = { path = "pylucene/dist/lucene-10.0.0-cp313-cp313-macosx_11_0_arm64.whl" }

src/pybool_ir/query/generic/parser.py

@@ -28,7 +28,8 @@
 
 assert lucene.getVMEnv() or lucene.initVM()
 Q = engine.Query
-search.BooleanQuery.setMaxClauseCount(MAX_CLAUSES)  # There is apparently a cap for efficiency reasons.
+# TODO https://lucene.apache.org/core/10_0_0/MIGRATE.html
+# search.BooleanQuery.setMaxClauseCount(MAX_CLAUSES)  # There is apparently a cap for efficiency reasons.
 
 # Makes parsing faster. (?)
 ParserElement.enablePackrat()

src/pybool_ir/query/parser.py

@@ -9,6 +9,7 @@
 
 from pybool_ir.query.ast import ASTNode
 
+# TODO https://lucene.apache.org/core/10_0_0/MIGRATE.html
 MAX_CLAUSES = 60_000
 
 assert lucene.getVMEnv() or lucene.initVM()

src/pybool_ir/query/pubmed/parser.py

@@ -30,7 +30,8 @@
 
 assert lucene.getVMEnv() or lucene.initVM()
 Q = engine.Query
-search.BooleanQuery.setMaxClauseCount(MAX_CLAUSES)  # There is apparently a cap for efficiency reasons.
+# TODO https://lucene.apache.org/core/10_0_0/MIGRATE.html
+#search.BooleanQuery.setMaxClauseCount(MAX_CLAUSES)  # There is apparently a cap for efficiency reasons.
 analyzer = engine.analyzers.Analyzer.standard()
 
 

src/pybool_ir/query/units.py

@@ -13,7 +13,8 @@
 
 assert lucene.getVMEnv() or lucene.initVM()
 Q = engine.Query
-search.BooleanQuery.setMaxClauseCount(MAX_CLAUSES)  # There is apparently a cap for efficiency reasons.
+# TODO https://lucene.apache.org/core/10_0_0/MIGRATE.html
+#search.BooleanQuery.setMaxClauseCount(MAX_CLAUSES)  # There is apparently a cap for efficiency reasons.
 analyzer = engine.analyzers.Analyzer.standard()