Import py-pyspark-3.0.1 to wip

py-pyspark/DESCR

@@ -0,0 +1,8 @@
+Spark is a unified analytics engine for large-scale data
+processing. It provides high-level APIs in Scala, Java,
+Python, and R, and an optimized engine that supports
+general computation graphs for data analysis. It also
+supports a rich set of higher-level tools including Spark
+SQL for SQL and DataFrames, MLlib for machine learning,
+GraphX for graph processing, and Structured Streaming
+for stream processing.

py-pyspark/Makefile

@@ -0,0 +1,55 @@
+# $NetBSD$
+
+DISTNAME=	pyspark-${VERS}
+PKGNAME=	${PYPKGPREFIX}-${DISTNAME:S/py//}
+CATEGORIES=	devel python
+MASTER_SITES=   ${MASTER_SITE_PYPI:=p/pyspark/}
+
+MAINTAINER=	kamel.derouiche@gmail.com
+HOMEPAGE=	https://spark.apache.org/
+COMMENT=	Apache Spark Python API
+LICENSE=	apache-2.0
+
+VERS=		3.0.1
+USE_TOOLS+=	bash
+
+PYTHON_VERSIONS_INCOMPATIBLE=	27
+
+DEPENDS+=	${PYPKGPREFIX}-pandas>=0.23.0:../../math/py-pandas
+DEPENDS+=	${PYPKGPREFIX}-py4j>=0.10.9.1:../../wip/py-py4j
+
+USE_JAVA=	run
+USE_JAVA2=	8
+
+REPLACE_BASH+=	\
+	deps/bin/beeline		\
+	deps/bin/docker-image-tool.sh	\
+	deps/bin/find-spark-home	\
+	deps/bin/load-spark-env.sh	\
+	deps/bin/pyspark		\
+	deps/bin/run-example		\
+	deps/bin/spark-class		\
+	deps/bin/spark-shell		\
+	deps/bin/spark-sql		\
+	deps/bin/spark-submit		\
+	deps/bin/sparkR
+
+REPLACE_BASH+=	\
+	deps/sbin/spark-daemon.sh	 \
+	deps/sbin/start-history-server.sh \
+	deps/sbin/stop-history-server.sh
+
+
+SUBST_CLASSES+=                 py_interpreter
+SUBST_STAGE.py_interpreter=     post-extract
+SUBST_FILES.py_interpreter+=	deps/bin/find-spark-home
+SUBST_SED.py_interpreter=       -e "s,python,python${PYVERSSUFFIX},g"
+
+post-extract:
+	${RM} -f ${WRKSRC}/deps/bin/*.cmd
+
+.include "../../math/py-numpy/buildlink3.mk"
+.include "../../lang/python/application.mk"
+.include "../../lang/python/egg.mk"
+.include "../../mk/java-vm.mk"
+.include "../../mk/bsd.pkg.mk"

py-pyspark/distinfo

@@ -0,0 +1,6 @@
+$NetBSD$
+
+SHA1 (pyspark-3.0.1.tar.gz) = d540c283f22c492761e49744dd58f287d948fdf7
+RMD160 (pyspark-3.0.1.tar.gz) = d7c1cb855f861a2100ed948d1b68d7ab64c95672
+SHA512 (pyspark-3.0.1.tar.gz) = 5e56e228a636110e0650af172e69b109feba5586250e41f3e313b770672a8d5df62dab850581e7aa3da0da9bc215be636a769dd558f0415a65dce72b62665e5e
+Size (pyspark-3.0.1.tar.gz) = 204241512 bytes