From 34f373a14d85b1e9ad0e3220d861d5cc4dc395a1 Mon Sep 17 00:00:00 2001
From: David Roazen <droazen@broadinstitute.org>
Date: Fri, 8 Dec 2023 15:47:19 -0500
Subject: [PATCH] Update the GATK base image to a newer LTS ubuntu release

---
 Dockerfile                                    |  6 ++--
 scripts/docker/gatkbase/Dockerfile            | 35 +++++++++++--------
 .../gatkbase/stage_docker_base_in_cloud.sh    | 20 +++++++++++
 3 files changed, 44 insertions(+), 17 deletions(-)
 create mode 100755 scripts/docker/gatkbase/stage_docker_base_in_cloud.sh

diff --git a/Dockerfile b/Dockerfile
index a513863128d..31bb4192c6a 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -1,4 +1,4 @@
-ARG BASE_DOCKER=broadinstitute/gatk:gatkbase-3.1.0
+ARG BASE_DOCKER=us.gcr.io/broad-dsde-methods/gatk-base-image-staging-area:3.2.0rc2
 
 # stage 1 for constructing the GATK zip
 FROM ${BASE_DOCKER} AS gradleBuild
@@ -93,8 +93,8 @@ RUN conda env create -n gatk -f /gatk/gatkcondaenv.yml && \
     echo "source activate gatk" >> /gatk/gatkenv.rc && \
     echo "source /gatk/gatk-completion.sh" >> /gatk/gatkenv.rc && \
     conda clean -afy && \
-    find /opt/miniconda/ -follow -type f -name '*.a' -delete && \
-    find /opt/miniconda/ -follow -type f -name '*.pyc' -delete && \
+    find $CONDA_PATH -follow -type f -name '*.a' -delete && \
+    find $CONDA_PATH -follow -type f -name '*.pyc' -delete && \
     rm -rf /root/.cache/pip
 
 CMD ["bash", "--init-file", "/gatk/gatkenv.rc"]
diff --git a/scripts/docker/gatkbase/Dockerfile b/scripts/docker/gatkbase/Dockerfile
index 7f4d8cb6a7b..692d1d44efe 100644
--- a/scripts/docker/gatkbase/Dockerfile
+++ b/scripts/docker/gatkbase/Dockerfile
@@ -1,12 +1,13 @@
 # Using OpenJDK 17
 # This Dockerfile does not require any files that are in the GATK4 repo.
-FROM ubuntu:18.04
+FROM ubuntu:22.04
 
 #### Basic image utilities
 RUN apt-get update && \
     apt-get upgrade -y && \
     apt-get install -y --no-install-recommends \
     python3 \
+    python3-pip \
     wget \
     curl \
     bc \
@@ -15,10 +16,13 @@ RUN apt-get update && \
     less \
     bedtools \
     samtools \
+    bcftools \
     tabix \
+    git \
     gpg-agent \
     build-essential \
     openjdk-17-jdk \
+    vim \
     software-properties-common && \
     apt-get -y clean  && \
     apt-get -y autoclean  && \
@@ -54,17 +58,20 @@ CMD ["bash"]
 ENV JAVA_LIBRARY_PATH /usr/lib/jni
 
 # Install miniconda
-ENV DOWNLOAD_DIR /downloads
-ENV CONDA_URL https://repo.anaconda.com/miniconda/Miniconda3-py310_23.1.0-1-Linux-x86_64.sh 
-ENV CONDA_MD5 = "32d73e1bc33fda089d7cd9ef4c1be542616bd8e437d1f77afeeaf7afdb019787"
-ENV CONDA_PATH /opt/miniconda
+# Following the instructions from https://docs.conda.io/projects/conda/en/latest/user-guide/install/rpm-debian.html
+# Note that this *does* install just miniconda, despite the package being named "conda"
+RUN curl https://repo.anaconda.com/pkgs/misc/gpgkeys/anaconda.asc | gpg --dearmor > /tmp/conda.gpg && \
+    install -o root -g root -m 644 /tmp/conda.gpg /usr/share/keyrings/conda-archive-keyring.gpg && \
+    gpg --keyring /usr/share/keyrings/conda-archive-keyring.gpg --no-default-keyring --fingerprint 34161F5BF5EB1D4BFBBB8F0A8AEB4F8B29D82806 && \
+    rm /tmp/conda.gpg && \
+    echo "deb [arch=amd64 signed-by=/usr/share/keyrings/conda-archive-keyring.gpg] https://repo.anaconda.com/pkgs/misc/debrepo/conda stable main" > /etc/apt/sources.list.d/conda.list && \
+    apt update && \
+    apt install conda && \
+    apt -y clean  && \
+    apt -y autoclean  && \
+    apt -y autoremove && \
+    . /opt/conda/etc/profile.d/conda.sh && \
+    conda -V
+
+ENV CONDA_PATH /opt/conda
 ENV PATH $CONDA_PATH/bin:$PATH
-RUN mkdir $DOWNLOAD_DIR && \
-    wget -nv -O $DOWNLOAD_DIR/miniconda.sh $CONDA_URL && \
-    test "`md5sum $DOWNLOAD_DIR/miniconda.sh | awk -v FS='  ' '{print $1}'` = $CONDA_MD5" && \
-    bash $DOWNLOAD_DIR/miniconda.sh -p $CONDA_PATH -b && \
-    rm $DOWNLOAD_DIR/miniconda.sh && \
-    conda clean -afy && \
-    find /opt/miniconda/ -follow -type f -name '*.a' -delete && \
-    find /opt/miniconda/ -follow -type f -name '*.pyc' -delete && \
-    rm -rf /root/.cache/pip 
diff --git a/scripts/docker/gatkbase/stage_docker_base_in_cloud.sh b/scripts/docker/gatkbase/stage_docker_base_in_cloud.sh
new file mode 100755
index 00000000000..5a2b6fefe40
--- /dev/null
+++ b/scripts/docker/gatkbase/stage_docker_base_in_cloud.sh
@@ -0,0 +1,20 @@
+#!/bin/bash
+
+if [ $# -ne 1 ]; then
+    echo "Usage: $0 docker_image_version"
+    exit 1
+fi
+
+IMAGE_VERSION=$1
+IMAGE_NAME="us.gcr.io/broad-dsde-methods/gatk-base-image-staging-area"
+DOCKER_IMAGE_TAG="${IMAGE_NAME}:${IMAGE_VERSION}"
+
+gcloud builds submit --tag ${DOCKER_IMAGE_TAG} --timeout=24h --machine-type n1_highcpu_32
+
+if [ $? -ne 0 ]; then
+    echo "gcloud builds submit failed"
+    exit 1
+fi
+
+echo "Successfully published image to staging area at ${DOCKER_IMAGE_TAG}"
+exit 0
\ No newline at end of file