Skip to content

SPARKNLP-1094 Adding Support to Read Word Files #3772

SPARKNLP-1094 Adding Support to Read Word Files

SPARKNLP-1094 Adding Support to Read Word Files #3772

name: build
on:
push:
paths-ignore:
- 'docs/**'
- 'conda/**'
- 'scripts/**'
- 'examples/**'
- '**.md'
- '**.yaml'
- '**.yml'
branches:
- 'master'
- '*release*'
- 'release/**'
- 'main'
pull_request:
paths-ignore:
- 'docs/**'
- 'conda/**'
- 'scripts/**'
- 'examples/**'
- '**.md'
- '**.yaml'
- '**.yml'
branches:
- 'master'
- '*release*'
- 'release/**'
- 'main'
jobs:
spark34:
if: "! contains(toJSON(github.event.commits.*.message), '[skip test]')"
runs-on: macos-13
env:
TF_CPP_MIN_LOG_LEVEL: 3
JAVA_OPTS: "-Xmx4096m -XX:+UseG1GC"
name: Build and Test on Apache Spark 3.4.x
steps:
- uses: actions/checkout@v3
- uses: actions/setup-java@v4
with:
distribution: 'temurin'
java-version: '8'
cache: 'sbt'
- name: Install Python 3.7
uses: actions/setup-python@v2
with:
python-version: 3.7.7
architecture: x64
- name: Install Python packages (Python 3.7)
run: |
python -m pip install --upgrade pip
pip install pyspark==3.4.0 numpy pytest
- name: Build Spark NLP on Apache Spark 3.4.0
run: |
brew install sbt
sbt -mem 4096 -Dis_spark34=true clean assemblyAndCopy
- name: Test Spark NLP in Scala - Apache Spark 3.4.x
run: |
sbt -mem 4096 coverage test
- name: Upload coverage data to Coveralls
run: sbt coverageReport coveralls
env:
COVERALLS_REPO_TOKEN: ${{ secrets.GITHUB_TOKEN }}
COVERALLS_FLAG_NAME: Apache Spark 3.4.x - Scala 2.12
- name: Test Spark NLP in Python - Apache Spark 3.4.x
run: |
cd python
python3.7 -m pytest -v -m fast
spark35:
if: "! contains(toJSON(github.event.commits.*.message), '[skip test]')"
runs-on: macos-13
env:
TF_CPP_MIN_LOG_LEVEL: 3
JAVA_OPTS: "-Xmx4096m -XX:+UseG1GC"
name: Build and Test on Apache Spark 3.5.x
steps:
- uses: actions/checkout@v3
- uses: actions/setup-java@v3
with:
distribution: 'adopt'
java-version: '8'
cache: 'sbt'
- name: Install Python 3.10
uses: actions/setup-python@v2
with:
python-version: 3.10.12
architecture: x64
- name: Install Python packages (Python 3.10)
run: |
python -m pip install --upgrade pip
pip install pyspark==3.5.0 numpy pytest
- name: Build Spark NLP on Apache Spark 3.5.0
run: |
brew install sbt
sbt -mem 4096 -Dis_spark35=true clean assemblyAndCopy
- name: Test Spark NLP in Scala - Apache Spark 3.5.x
run: |
sbt -mem 4096 test
- name: Test Spark NLP in Python - Apache Spark 3.5.x
run: |
cd python
python3.10 -m pytest -v -m fast
spark33:
if: "! contains(toJSON(github.event.commits.*.message), '[skip test]')"
runs-on: macos-13
env:
TF_CPP_MIN_LOG_LEVEL: 3
JAVA_OPTS: "-Xmx4096m -XX:+UseG1GC"
name: Build and Test on Apache Spark 3.3.x
steps:
- uses: actions/checkout@v3
- uses: actions/setup-java@v3
with:
distribution: 'adopt'
java-version: '8'
cache: 'sbt'
- name: Install Python 3.7
uses: actions/setup-python@v2
with:
python-version: 3.7.7
architecture: x64
- name: Install Python packages (Python 3.7)
run: |
python -m pip install --upgrade pip
pip install pyspark==3.3.1 numpy pytest
- name: Build Spark NLP on Apache Spark 3.3.1
run: |
brew install sbt
sbt -mem 4096 -Dis_spark33=true clean assemblyAndCopy
- name: Test Spark NLP in Scala - Apache Spark 3.3.x
run: |
sbt -mem 4096 test
- name: Test Spark NLP in Python - Apache Spark 3.3.x
run: |
cd python
python3.7 -m pytest -v -m fast