Add SPARQL Conformance Testing Framework

.gitignore

@@ -2,4 +2,5 @@ __pycache__
 build/
 dist/
 *.egg-info
+*.memory-log.json
 *.swp

README.md

@@ -21,6 +21,15 @@ On other platforms simply install the `qlever` command-line
 [python package using `pipx`/`uv`](https://docs.qlever.dev/quickstart/#others).
 Note: QLever will be executed in a container which will come with a performance penalty.
 
+# Optional SPARQL conformance integration
+
+The `qlever` package is independent of the SPARQL conformance test harness and
+does not install it. To use the integrated
+`sparql_conformance setup|test|analyze|visualize` workflow, install the
+[`sparql-conformance`](https://github.com/SIRDNARch/sparql-conformance)
+package separately. That package owns the `sparql_conformance` command and
+loads qlever-control only when the integration is used.
+
 # Use with your own dataset
 
 To use QLever with your own dataset, you need a `Qleverfile`, like in the

pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["setuptools>=61.0", "wheel"]
+requires = ["setuptools>=64.0", "wheel"]
 build-backend = "setuptools.build_meta"
 
 [project]
@@ -33,7 +33,13 @@ repository = "https://github.com/ad-freiburg/qlever.git"
 bugtracker = "https://github.com/ad-freiburg/qlever/issues"
 
 [project.scripts]
+"qblazegraph" = "qlever.qlever_main:main"
+"qgraphdb" = "qlever.qlever_main:main"
+"qjena" = "qlever.qlever_main:main"
 "qlever" = "qlever.qlever_main:main"
+"qmdb" = "qlever.qlever_main:main"
+"qoxigraph" = "qlever.qlever_main:main"
+"qvirtuoso" = "qlever.qlever_main:main"
 
 [tool.setuptools]
 package-data = { "qlever" = ["Qleverfiles/*", "evaluation/www/*"] }

src/qblazegraph/Dockerfile

@@ -0,0 +1,23 @@
+# Use an official OpenJDK runtime as a parent image
+FROM docker.io/library/openjdk:26-ea-21-jdk-slim
+
+RUN apt-get update && apt-get install -y wget gzip coreutils curl unzip
+
+WORKDIR /opt
+
+# Download and set up Blazegraph
+RUN wget https://github.com/blazegraph/database/releases/download/BLAZEGRAPH_2_1_6_RC/blazegraph.jar
+
+# Set ownership to the user passed by UID and GID
+ARG UID
+ARG GID
+RUN if [ "${UID:-}" != "" ] && [ "${GID:-}" != "" ]; then \
+        chown -R ${UID}:${GID} /opt; \
+    fi
+
+RUN chmod u+x blazegraph.jar
+
+# Expose Blazegraph port
+# EXPOSE 9999
+
+CMD ["bash"]

src/qblazegraph/RWStore.conformance.properties

@@ -0,0 +1,33 @@
+# ===================== Basic Configuration =====================
+com.bigdata.journal.AbstractJournal.file=blazegraph.jnl
+com.bigdata.journal.AbstractJournal.bufferMode=DiskRW
+com.bigdata.journal.AbstractJournal.initialExtent=209715200
+com.bigdata.journal.AbstractJournal.maximumExtent=200000000000
+com.bigdata.rdf.store.AbstractTripleStore.textIndex=false
+com.bigdata.rdf.store.AbstractTripleStore.justify=false
+com.bigdata.rdf.store.AbstractTripleStore.axiomsClass=com.bigdata.rdf.axioms.NoAxioms
+com.bigdata.rdf.store.AbstractTripleStore.quads=true
+com.bigdata.rdf.store.AbstractTripleStore.statementIdentifiers=false
+com.bigdata.rdf.sail.truthMaintenance=false
+com.bigdata.rdf.store.AbstractTripleStore.geoSpatial=false
+
+# ===================== DataLoader Performance =====================
+com.bigdata.rdf.store.DataLoader.bufferCapacity=1000000
+com.bigdata.rdf.store.DataLoader.queueCapacity=150
+com.bigdata.rdf.store.DataLoader.gzipBufferSize=1000000000
+com.bigdata.rdf.store.DataLoader.closure=None
+
+# ===================== BTree Tuning (example values; adjust based on your dataset) =====================
+# com.bigdata.namespace.kb.lex.BLOBS.com.bigdata.btree.BTree.branchingFactor=576
+# com.bigdata.namespace.kb.lex.ID2TERM.com.bigdata.btree.BTree.branchingFactor=781
+# com.bigdata.namespace.kb.lex.TERM2ID.com.bigdata.btree.BTree.branchingFactor=380
+# com.bigdata.namespace.kb.spo.OSP.com.bigdata.btree.BTree.branchingFactor=737
+# com.bigdata.namespace.kb.spo.POS.com.bigdata.btree.BTree.branchingFactor=917
+# com.bigdata.namespace.kb.spo.SPO.com.bigdata.btree.BTree.branchingFactor=731
+
+# ===================== Optional Namespace Settings =====================
+com.bigdata.namespace.kb.spo.com.bigdata.btree.BTree.branchingFactor=1024
+com.bigdata.rdf.store.AbstractTripleStore.namespace=kb
+
+# ===================== RDF & Parser Configuration =====================
+com.bigdata.rdf.parser.RDFParserFactory.supportsGZip=true

src/qblazegraph/RWStore.properties

@@ -0,0 +1,33 @@
+# ===================== Basic Configuration =====================
+com.bigdata.journal.AbstractJournal.file=blazegraph.jnl
+com.bigdata.journal.AbstractJournal.bufferMode=DiskRW
+com.bigdata.journal.AbstractJournal.initialExtent=209715200
+com.bigdata.journal.AbstractJournal.maximumExtent=200000000000
+com.bigdata.rdf.store.AbstractTripleStore.textIndex=false
+com.bigdata.rdf.store.AbstractTripleStore.justify=false
+com.bigdata.rdf.store.AbstractTripleStore.axiomsClass=com.bigdata.rdf.axioms.NoAxioms
+com.bigdata.rdf.store.AbstractTripleStore.quads=false
+com.bigdata.rdf.store.AbstractTripleStore.statementIdentifiers=false
+com.bigdata.rdf.sail.truthMaintenance=false
+com.bigdata.rdf.store.AbstractTripleStore.geoSpatial=false
+
+# ===================== DataLoader Performance =====================
+com.bigdata.rdf.store.DataLoader.bufferCapacity=1000000
+com.bigdata.rdf.store.DataLoader.queueCapacity=150
+com.bigdata.rdf.store.DataLoader.gzipBufferSize=1000000000
+com.bigdata.rdf.store.DataLoader.closure=None
+
+# ===================== BTree Tuning (example values; adjust based on your dataset) =====================
+# com.bigdata.namespace.kb.lex.BLOBS.com.bigdata.btree.BTree.branchingFactor=576
+# com.bigdata.namespace.kb.lex.ID2TERM.com.bigdata.btree.BTree.branchingFactor=781
+# com.bigdata.namespace.kb.lex.TERM2ID.com.bigdata.btree.BTree.branchingFactor=380
+# com.bigdata.namespace.kb.spo.OSP.com.bigdata.btree.BTree.branchingFactor=737
+# com.bigdata.namespace.kb.spo.POS.com.bigdata.btree.BTree.branchingFactor=917
+# com.bigdata.namespace.kb.spo.SPO.com.bigdata.btree.BTree.branchingFactor=731
+
+# ===================== Optional Namespace Settings =====================
+com.bigdata.namespace.kb.spo.com.bigdata.btree.BTree.branchingFactor=1024
+com.bigdata.rdf.store.AbstractTripleStore.namespace=kb
+
+# ===================== RDF & Parser Configuration =====================
+com.bigdata.rdf.parser.RDFParserFactory.supportsGZip=true

src/qblazegraph/commands/benchmark_queries.py

@@ -0,0 +1,21 @@
+from __future__ import annotations
+
+from qlever.commands.benchmark_queries import (
+    BenchmarkQueriesCommand as QleverBenchmarkQueriesCommand,
+)
+
+
+class BenchmarkQueriesCommand(QleverBenchmarkQueriesCommand):
+    def relevant_qleverfile_arguments(self) -> dict[str, list[str]]:
+        return {
+            "data": ["name"],
+            "server": ["host_name", "port", "timeout"],
+            "ui": ["ui_config"],
+        }
+
+    def execute(self, args) -> bool:
+        if not args.sparql_endpoint:
+            args.sparql_endpoint = (
+                f"{args.host_name}:{args.port}/blazegraph/namespace/kb/sparql"
+            )
+        return super().execute(args)

src/qblazegraph/commands/extract_queries.py

@@ -0,0 +1 @@
+../../qlever/commands/extract_queries.py

src/qblazegraph/commands/get_data.py

@@ -0,0 +1 @@
+../../qlever/commands/get_data.py

src/qblazegraph/commands/index.py

@@ -0,0 +1,173 @@
+from __future__ import annotations
+
+from pathlib import Path
+
+import qlever.util as util
+from qlever.command import QleverCommand
+from qlever.containerize import Containerize
+from qlever.log import log
+
+
+class IndexCommand(QleverCommand):
+    def __init__(self):
+        self.script_name = "qblazegraph"
+
+    def description(self) -> str:
+        return "Build the index for a given RDF dataset"
+
+    def should_have_qleverfile(self) -> bool:
+        return True
+
+    def relevant_qleverfile_arguments(self) -> dict[str, list[str]]:
+        return {
+            "data": ["name", "format"],
+            "index": ["input_files", "jvm_args", "extra_args"],
+            "runtime": ["system", "image", "index_container"],
+        }
+
+    def additional_arguments(self, subparser):
+        subparser.add_argument(
+            "--blazegraph-jar",
+            type=str,
+            default="blazegraph.jar",
+            help=(
+                "Path to blazegraph.jar file (default: blazegraph.jar) "
+                "(this requires that you have Java installed and blazegraph.jar "
+                "downloaded on your machine)"
+            ),
+        )
+        subparser.add_argument(
+            "--rebuild-image",
+            action="store_true",
+            default=False,
+            help="Rebuild the Docker image to get the latest updates",
+        )
+
+    @staticmethod
+    def wrap_cmd_in_container(args, cmd: str) -> str:
+        return Containerize().containerize_command(
+            cmd=cmd,
+            container_system=args.system,
+            run_subcommand="run --rm",
+            image_name=args.image,
+            container_name=args.index_container,
+            volumes=[("$(pwd)", "/opt/index")],
+            working_directory="/opt/index",
+        )
+
+    def execute(self, args, called_from_conformance_test: bool = False) -> bool:
+        system = args.system
+        input_files = args.input_files
+
+        jar_path = (
+            args.blazegraph_jar
+            if args.system == "native"
+            else "/opt/blazegraph.jar"
+        )
+
+        index_cmd = (
+            f"java {args.jvm_args} -cp {jar_path} com.bigdata.rdf.store.DataLoader "
+            f"{args.extra_args} RWStore.properties {input_files}"
+        )
+        # Under a conformance test, redirect all output to the log file so the
+        # BlazeGraph banner and DataLoader output do not flood stdout; otherwise
+        # `tee` keeps showing it live.
+        if called_from_conformance_test:
+            index_cmd += f" > {args.name}.index-log.txt 2>&1"
+        else:
+            index_cmd += f" | tee {args.name}.index-log.txt"
+
+        image_id = build_cmd = ""
+        if args.system == "native":
+            cmd_to_show = index_cmd
+        else:
+            index_cmd = self.wrap_cmd_in_container(args, index_cmd)
+            dockerfile_dir = Path(__file__).parent.parent
+            dockerfile_path = dockerfile_dir / "Dockerfile"
+            build_cmd = (
+                f"{system} build -f {dockerfile_path} -t {args.image} --build-arg "
+                f"UID=$(id -u) --build-arg GID=$(id -g) {dockerfile_dir}"
+            )
+            image_id = util.get_container_image_id(system, args.image)
+            cmd_to_show = (
+                f"{build_cmd}\n\n{index_cmd}"
+                if not image_id or args.rebuild_image
+                else index_cmd
+            )
+
+        # Show the command line.
+        self.show(cmd_to_show, only_show=args.show)
+        if args.show:
+            return True
+
+        # Check if all of the input files exist.
+        if not util.input_files_exist(input_files):
+            return False
+
+        # When running natively, check if the binary exists and works.
+        if args.system == "native":
+            try:
+                util.run_command("java --help")
+            except Exception as e:
+                log.error(f"Java not found on the machine! - {e}")
+                log.info(
+                    "Blazegraph needs Java to execute the blazegraph.jar file"
+                )
+                return False
+            if not Path(args.blazegraph_jar).exists():
+                jar_link = (
+                    "https://github.com/blazegraph/database/releases/download/"
+                    "BLAZEGRAPH_2_1_6_RC/blazegraph.jar"
+                )
+                log.error(
+                    "Couldn't find the blazegraph.jar in specified path: "
+                    f"{Path(args.blazegraph_jar).absolute()}\n"
+                )
+                log.info(
+                    "Are you sure you downloaded the blazegraph.jar file? "
+                    f"blazegraph.jar can be downloaded from {jar_link}"
+                )
+                return False
+        else:
+            if Containerize().is_running(args.system, args.index_container):
+                log.info(
+                    f"{args.system} container {args.index_container} is still up, "
+                    "which means that data loading is in progress. Please wait..."
+                )
+                return False
+            # A previously interrupted run can leave a container with this name
+            # in a non-running (Created/Exited) state, which makes the
+            # `docker/podman run --name` below fail with exit code 125. Remove
+            # any such stale container first.
+            Containerize.stop_and_remove_container(
+                args.system, args.index_container
+            )
+
+            if not image_id or args.rebuild_image:
+                build_successful = util.build_image(
+                    build_cmd, system, args.image
+                )
+                if not build_successful:
+                    return False
+            else:
+                log.info(f"{args.image} image present on the system\n")
+
+        index_jnl = Path("blazegraph.jnl")
+        if index_jnl.exists():
+            log.error(
+                "Blazegraph journal blazegraph.jnl found in current working "
+                "directory which shows presence of a previous index\n"
+            )
+            log.info("Aborting the index operation...")
+            return False
+
+        # Run the index command.
+        try:
+            util.run_command(
+                index_cmd, show_output=not called_from_conformance_test
+            )
+        except Exception as e:
+            log.error(f"Building the index failed: {e}")
+            return False
+
+        return True