Merge pull request #132 from scrapnode/feature/upgrade-to-es-8.x

upgrade: docker-compose with version 8.x

.env.sample

@@ -1 +1 @@
-ES_VERSION=7.5.1
+ES_VERSION=8.3.3

Dockerfile

@@ -0,0 +1,48 @@
+ARG ES_VERSION=8.3.3
+ARG DEBIAN_FRONTEND=noninteractive
+
+# thanks to https://github.com/cpfriend1721994/docker-es-cococ-tokenizer
+FROM docker.elastic.co/elasticsearch/elasticsearch:$ES_VERSION as builder
+ARG ES_VERSION
+ARG DEBIAN_FRONTEND
+USER root
+
+RUN apt-get update -y && apt-get install -y software-properties-common build-essential
+RUN gcc --version
+RUN apt-get update -y && \
+    apt-get install -y make cmake pkg-config wget git openjdk-17-jdk
+ENV JAVA_HOME=/usr/lib/jvm/java-17-openjdk-amd64
+
+RUN cd /tmp && wget https://dlcdn.apache.org/maven/maven-3/3.8.8/binaries/apache-maven-3.8.8-bin.tar.gz && \
+tar xvf apache-maven-3.8.8-bin.tar.gz && \
+mkdir -p /usr/share/maven && cd /usr/share/maven && \
+cp -r /tmp/apache-maven-3.8.8/* .
+ENV PATH=/usr/share/maven/bin:$PATH
+
+WORKDIR /
+
+COPY pom.xml .
+RUN mvn verify clean --fail-never
+
+RUN git clone https://github.com/coccoc/coccoc-tokenizer.git
+
+RUN mkdir /coccoc-tokenizer/build
+WORKDIR /coccoc-tokenizer/build
+RUN cmake -DBUILD_JAVA=1 ..
+RUN make install
+
+COPY . /elasticsearch-analysis-vietnamese
+WORKDIR /elasticsearch-analysis-vietnamese
+RUN mvn package -Dmaven.test.skip -e
+
+FROM docker.elastic.co/elasticsearch/elasticsearch:$ES_VERSION
+ARG ES_VERSION
+
+COPY --from=builder /coccoc-tokenizer/dicts/tokenizer /usr/local/share/tokenizer/dicts
+COPY --from=builder /coccoc-tokenizer/dicts/vn_lang_tool /usr/local/share/tokenizer/dicts
+COPY --from=builder /coccoc-tokenizer/build/libcoccoc_tokenizer_jni.so /usr/lib
+COPY --from=builder /coccoc-tokenizer/build/multiterm_trie.dump /usr/local/share/tokenizer/dicts
+COPY --from=builder /coccoc-tokenizer/build/nontone_pair_freq_map.dump /usr/local/share/tokenizer/dicts
+COPY --from=builder /coccoc-tokenizer/build/syllable_trie.dump /usr/local/share/tokenizer/dicts
+COPY --from=builder /elasticsearch-analysis-vietnamese/target/releases/elasticsearch-analysis-vietnamese-$ES_VERSION.zip /
+RUN echo "Y" | /usr/share/elasticsearch/bin/elasticsearch-plugin install --batch file:///elasticsearch-analysis-vietnamese-$ES_VERSION.zip

README.md

@@ -108,6 +108,28 @@ The above example produces the following terms:
 
 ```
 
+## Use Docker
+
+Make sure you have installed both Docker & docker-compose as well
+
+### Use Docker Compose
+
+```bash
+# build the image
+cp .env.sample .env
+docker compose build
+docker compose up
+
+# test
+curl -k http://elastic:changemenow@localhost:9200/_analyze -H 'Content-Type: application/json' -d '
+{
+  "analyzer": "vi_analyzer",
+  "text": "Cộng hòa Xã hội chủ nghĩa Việt Nam"
+}'
+# example result
+{"tokens":[{"token":"cộng hòa","start_offset":0,"end_offset":8,"type":"<WORD>","position":0},{"token":"xã hội","start_offset":9,"end_offset":15,"type":"<WORD>","position":1},{"token":"chủ nghĩa","start_offset":16,"end_offset":25,"type":"<WORD>","position":2},{"token":"việt nam","start_offset":26,"end_offset":34,"type":"<WORD>","position":3}]}                                                                                     
+```
+
 ## Build from Source
 ### Step 1: Build C++ tokenizer for Vietnamese library
 ```sh
@@ -136,7 +158,7 @@ Optionally, edit the `elasticsearch-analysis-vietnamese/pom.xml` to change the v
 
 ```xml
 ...
-<version>7.17.1</version>
+<version>8.3.3</version>
 ...
  ```
 
@@ -149,7 +171,7 @@ mvn package
 ### Step 3: Installation the plugin on Elasticsearch
 
 ```sh
-bin/elasticsearch-plugin install file://target/releases/elasticsearch-analysis-vietnamese-7.17.1.zip
+bin/elasticsearch-plugin install file://target/releases/elasticsearch-analysis-vietnamese-8.3.3.zip
 ```
 
 ## Compatible Versions

docker-compose.yaml

@@ -1,15 +1,28 @@
 version: '3.4'
+
 services:
   elasticsearch:
-    image: docker.elastic.co/elasticsearch/elasticsearch:${ES_VERSION}
+    build: .
     restart: on-failure
     ports:
       - "9200:9200"
-    volumes:
-      - ./target/releases/elasticsearch-analysis-vietnamese-${ES_VERSION}.zip:/usr/share/elasticsearch/plugin/elasticsearch-analysis-vietnamese-${ES_VERSION}.zip
-      - ./install-es-plugin.sh:/apps/install-es-plugin.sh
+    ulimits:
+      nofile:
+         soft: 65536
+         hard: 65536
+      memlock:
+        hard: -1
+        soft: -1
     environment:
-      - "ES_VERSION=${ES_VERSION}"
-      - "discovery.type=single-node"
-    entrypoint:
-      - /apps/install-es-plugin.sh
+      ES_JAVA_OPTS: "-Xmx2g -Xms2g"
+      ELASTIC_USERNAME: elastic
+      ELASTIC_PASSWORD: changemenow
+      bootstrap.memory_lock: "true"
+      discovery.type: single-node
+      xpack.security.enabled: true
+    networks:
+      - elastic
+
+networks:
+  elastic:
+    driver: bridge