From 5d625ea754f1a022287afaddf38305977a5b6bf3 Mon Sep 17 00:00:00 2001
From: Oliver Bertuch <o.bertuch@fz-juelich.de>
Date: Mon, 3 Jan 2022 15:06:44 +0100
Subject: [PATCH] feat(solr): add Makefile to create Dataverse ConfigSet #7662

Simple Makefile to download Solr, extract the default configset
and create a Dataverse flavored one.

- Uses Maven to find the Solr distribution version to download.
- Uses xsltproc to apply our XSLT transformations to sorlconfig.xml
- Replaces the managed-schema with the static one we provide
- Zips the configset to make it distributable as artifact
---
 conf/solr/.gitignore |  1 +
 conf/solr/Makefile   | 80 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 81 insertions(+)
 create mode 100644 conf/solr/.gitignore
 create mode 100644 conf/solr/Makefile

diff --git a/conf/solr/.gitignore b/conf/solr/.gitignore
new file mode 100644
index 00000000000..53c37a16608
--- /dev/null
+++ b/conf/solr/.gitignore
@@ -0,0 +1 @@
+dist
\ No newline at end of file
diff --git a/conf/solr/Makefile b/conf/solr/Makefile
new file mode 100644
index 00000000000..45ba7eee937
--- /dev/null
+++ b/conf/solr/Makefile
@@ -0,0 +1,80 @@
+DIST_DIR           := ./dist
+DOWNLOAD_DIR       := $(DIST_DIR)/download
+EXTRACT_DIR        := $(DIST_DIR)/extract
+POM_LOCATION       := ../../pom.xml
+XSLT_LOCATION      := ./config
+XSLT_FILES         := $(wildcard $(XSLT_LOCATION)/*.xslt)
+SOLR_VERS_FILE     := $(DIST_DIR)/solr-version
+SOLR_TARGZ         := $(DOWNLOAD_DIR)/solr.tar.gz
+SOLR_DEFAULT_CS    := $(EXTRACT_DIR)/_default
+SOLR_CONFIGSET     := $(DIST_DIR)/dataverse_cs
+SOLR_CFG_SRC       := $(SOLR_DEFAULT_CS)/conf/solrconfig.xml
+SOLR_CFG_PATCHED   := $(SOLR_CONFIGSET)/conf/solrconfig.xml
+SOLR_CFG_TMP       := $(DIST_DIR)/solrconfig.xml.tmp
+SOLR_SCHEMA_DST    := $(SOLR_CONFIGSET)/conf/schema.xml
+SOLR_SCHEMA_SRC    := ./schema/schema.xml
+SOLR_CONFIGSET_ZIP := $(DIST_DIR)/solr-configset.zip
+
+.PHONY: all clean clean-all get-solr prepare-solr patch-solr package
+
+all: patch-solr
+
+clean:
+	rm -rf $(EXTRACT_DIR) $(SOLR_CONFIGSET) $(SOLR_CONFIGSET_ZIP)
+
+clean-all:
+	rm -rf $(DIST_DIR)
+
+get-solr: $(DOWNLOAD_DIR) $(SOLR_TARGZ)
+
+prepare-solr: $(EXTRACT_DIR) get-solr $(SOLR_CONFIGSET)
+
+patch-solr: prepare-solr $(SOLR_CFG_PATCHED) $(SOLR_SCHEMA_DST)
+
+package: patch-solr $(SOLR_CONFIGSET_ZIP)
+
+$(SOLR_CONFIGSET_ZIP): $(SOLR_CFG_PATCHED) $(SOLR_SCHEMA_DST)
+	$(info Zipping the Dataverse Solr ConfigSet for distribution)
+	cd "$(SOLR_CONFIGSET)" && zip -9Tru "$(abspath $(SOLR_CONFIGSET_ZIP))" *
+
+$(SOLR_SCHEMA_DST): $(SOLR_SCHEMA_SRC)
+	$(info Delete managed-schema and copy static schema into Dataverse ConfigSet)
+	-rm $(SOLR_CONFIGSET)/conf/managed-schema
+	cp "$(SOLR_SCHEMA_SRC)" "$(SOLR_SCHEMA_DST)"
+
+$(SOLR_CFG_PATCHED): $(XSLT_FILES)
+	$(info Dataverse ConfigSet: Copy default solrconfig.xml and patch with XSLT)
+	cp "$(SOLR_CFG_SRC)" "$(SOLR_CFG_PATCHED)"
+	for XSLT in $(XSLT_FILES); do xsltproc "$$XSLT" "$(SOLR_CFG_PATCHED)" > "$(SOLR_CFG_TMP)" && mv "$(SOLR_CFG_TMP)" "$(SOLR_CFG_PATCHED)"; done
+
+$(SOLR_CONFIGSET): $(SOLR_DEFAULT_CS)
+	$(info Copy default ConfigSet into a Dataverse flavored one)
+	[ -d "$(SOLR_CONFIGSET)" ] && rm -rf "$(SOLR_CONFIGSET)" || true
+	cp -a "$(SOLR_DEFAULT_CS)" "$(SOLR_CONFIGSET)"
+	rm "$(SOLR_CFG_PATCHED)" # initialy delete the config file - we will add a patched one in the next step
+
+$(SOLR_DEFAULT_CS): $(SOLR_TARGZ) $(SOLR_VERS_FILE)
+	$(info Extracting default ConfigSet from distribution)
+	$(eval SOLR_VERSION := $(shell cat "$(SOLR_VERS_FILE)"))
+	$(eval SOLR_TARGZ_SUBPATH := solr-$(SOLR_VERSION)/server/solr/configsets/_default)
+	tar -xmzf "$(SOLR_TARGZ)" -C "$(EXTRACT_DIR)" "$(SOLR_TARGZ_SUBPATH)"
+	cp -r "$(EXTRACT_DIR)/$(SOLR_TARGZ_SUBPATH)" "$(EXTRACT_DIR)"
+	rm -rf "$(EXTRACT_DIR)/solr-$(SOLR_VERSION)"
+
+$(SOLR_TARGZ): $(SOLR_VERS_FILE)
+	$(info Downloading Solr binary distribution)
+	$(eval SOLR_VERSION := $(shell cat "$(SOLR_VERS_FILE)"))
+	$(eval SOLR_DOWNLOAD_URL := https://archive.apache.org/dist/lucene/solr/$(SOLR_VERSION)/solr-$(SOLR_VERSION).tgz)
+	curl -fsS "$(SOLR_DOWNLOAD_URL)" -o "$(SOLR_TARGZ)"
+
+$(SOLR_VERS_FILE): $(POM_LOCATION)
+	$(info Extracting Solr version from Maven POM)
+	[ -d "$(DIST_DIR)" ] || mkdir -p "$(DIST_DIR)"
+	mvn -f "$(POM_LOCATION)" help:evaluate -Dexpression=solr.version -q -DforceStdout > "$(SOLR_VERS_FILE)"
+
+$(DOWNLOAD_DIR):
+	[ -d "$(DOWNLOAD_DIR)" ] || mkdir -p "$(DOWNLOAD_DIR)"
+
+$(EXTRACT_DIR):
+	[ -d "$(EXTRACT_DIR)" ] || mkdir -p "$(EXTRACT_DIR)"
+