#cloud-config
#
# opensearch-init.yaml provides a development environemt for building
# web applications based on PostgreSQL 16, PostgREST 12, Pandoc 3.1,
# and using Python 3's http.server to emulate a front end web server.
#
# This cloud init target Ubuntu's Multipass and assume the image
# available is Jammy (22.04 LTS). Here's an example creating a VM
# with Multipass call "opensearch-machine" and accessing it via Multipass's
# shell.
#
# You need more than the default disk image size for fit all the stuff.
# When you initialize your VM use the following command
#
# ~~~
# multipass launch --name opensearch-machine \
#                  --memory 4G \
#                  --disk 50G \
#                  --cloud-init opensearch-init.yaml
# ~~~
#
# When you're done you can stop the VM with
#
# ~~~
# multipass stop opensearch-machine
# ~~~
#
# You can remove opensearch-machine completely with
#
# ~~~
# multipass stop opensearch-machine
# multipass delete opensearch-machine
# multipass purge opensearch-machine
# ~~~
#
# The first time you run your VM you will need to run the following
# additional setup bash scripts. Then you can choose to install
# to install Haskell (GHCup) so you can compile the latest Pandoc and PostgREST
# and OpenSearch/OpenSearch Dashboard.
#
# 1. 01-setup-scripts.bash, creates "$HOME/bin" and "$HOME/sbin" and adds things ito your .bashrc
# 2. 02-add-python-packages.bash, adds commonly used Python packages to the system
# 3. 03-add-go-and-caltechlibrary-tools.bash, builds and install DLD Go based tools
# 4. 04-ghcup-install.bash, installs Haskell toolchain and manager
# 5. 05-pandoc-build.bash, downloads the latest Pandoc and builds it
# 6. 06-postgrest-build.bash, downloads the latest PostgREST and builds it.
# 7. 07-add-opensearch.bash, downloads the OpenSearch/OpenSearch Dashbaord debs and installs them
#
write_files:
###   - content: |
###       You are now ready to start working with your OpenSearch Machine. Run
###       You then need to run following to configure this virtual machine for
###       development. 
### 
###           bash ./sbin/01-setup-scripts.bash
### 
###       Your may wish to run other scripts in the `$HOME/sbin` directory
###       depending on your development needs.
###       Finnally remember to resource your `.bashrc` file after you install
###       any software to pickup the environment changes defined in it.
### 
###           . .bashrc
###
###     path: /home/ubuntu/README.1st
###     owner: ubuntu:ubuntu
###     permissions: '0664'
  - content: |
      #!/bin/bash
      APP_NAME="$(basename "$0")"
      if [ "$1"  = "" ]; then
        cat <<EOT
      ${APP_NAME}(1) user manual | version 0

      # NAME

      ${APP_NAME}

      # SYNOPSIS

      ${APP_NAME} METHOD URL_PATH

      # DESCRIPTION

      ${APP_NAME} provides a thin wrapper around curl to create
      a simple OpenSearch client.  It requires you to supply a
      HTTP method (i.e. GET, PUT, POST, DELETE) and path part of the
      URL to access the OpenSearch REST API.

      If you are doing a PUT or POST then data will be read from standard
      input. This means you can easily type in the JSON or read it from a
      file using redirection.

      # EXAMPLE

      Retrieve the settings for our OpenSearch instance running in our 
      Multipass VM.

      ~~~shell
      ${APP_NAME} GET _settings
      ~~~

      Create an index called content-type

      ~~~shell
      ${APP_NAME} PUT content-type
      ~~~

      Add a JSON document to the index (i.e. mydata.json with the id "01")

      ~~~shell
      ${APP_NAME} PUT content-type/_doc/01 <mydata.json
      ~~~
      
      EOT
      
      fi

      #
      # OpenSearch Client
      #
      METHOD="$1"
      URL_PATH="$2"
      CONTENT_TYPE="application/json"
      
      case "${METHOD}" in
          put|PUT|post|POST)
          echo "Reading data from standard input"
          curl -k --user admin:admin \
            -H "Content-Type: ${CONTENT_TYPE}" \
            -X "${METHOD}" \
            -d @- \
            "https://localhost:9200/$URL_PATH" | jq .
          ;;
          *)
          curl -k --user admin:admin \
            -X "${METHOD}" \
            -H "Content-Type: ${CONTENT_TYPE}" \
            "https://localhost:9200/$URL_PATH" | jq .
      esac
    owner: root:root
    path: /usr/local/bin/os_client.bash
    permissions: '0775'
  - content: |
      #!/bin/bash
      mkdir -p $HOME/bin
      mkdir -p $HOME/sbin
      echo 'export PATH="$HOME/bin:$HOME/sbin:$PATH"' >>"${HOME}/.bashrc"
      cat <<EOT

      You have added ${HOME}/bin to the PATH in your .bashrc
      file. To update the path without logging out and in
      do
      
              . .bashrc
      
      EOT
    path: /usr/local/sbin/01-setup-scripts.bash
    owner: root:root
    permissions: '0775'
  - content: |
      #!/bin/bash
      python3 -m pip install idutils
      python3 -m pip install sqlfluff
      python3 -m pip install pylint
      python3 -m pip install progressbar2
      python3 -m pip install pybtex
      python3 -m pip install feedgen
      python3 -m pip install python-dateutil
      cat <<EOT >$HOME/bin/ws
      #!/bin/bash
      echo "Enter Ctlr-c to exit web server"
      if [ "$1" != "" ]; then
      	python3 -m http.server "$1"
      else
      	python3 -m http.server
      fi
      EOT
      chmod 775 $HOME/bin/ws
    path: /usr/local/sbin/02-add-python-packages.bash
    owner: root:root
    permissions: '0775'
  - content: |
      #!/bin/bash
      cd
      sudo snap install go --classic
      export GOPATH=$HOME
      mkdir -p "${HOME}/bin"
      mkdir -p "${HOME}/src/github.com/caltechlibrary"
      for REPO in irdmtools datatools dataset newt; do
        cd "${HOME}/src/github.com/caltechlibrary"
        git clone "https://github.com/caltechlibrary/${REPO}"
        cd "${REPO}"
        make
        make install
        cd "${HOME}/src/github.com/caltechlibrary"
      done
      cd 
    path: /usr/local/sbin/03-add-go-and-caltechlibrary-tools.bash
    owner: root:root
    permissions: '0775'
  - content: |
      #!/bin/bash
      curl --proto '=https' --tlsv1.2 -sSf https://get-ghcup.haskell.org | sh
    path: /usr/local/sbin/04-ghcup-install.bash
    owner: root:root
    permissions: '0775'
  - content: |
      #!/bin/bash
      PATH=$HOME/.ghcup/bin:$PATH
      export PATH
      cd
      git clone https://github.com/jgm/pandoc src/pandoc
      cd src/pandoc
      make
      cp -v $(find . -type f -name pandoc) "${HOME}/bin"
      cp -v $(find . -type f -name pandoc) "${HOME}/bin/pandoc-server"
    path: /usr/local/sbin/05-pandoc-build.bash
    owner: root:root
    permissions: '0775'
  - content: |
      #!/bin/bash
      cd
      git clone https://github.com/PostgREST/postgrest src/postgrest
      cd src/postgres 
      stack build --install-ghc --copy-bins --local-bin-path $HOME/bin
      cd
    path: /usr/local/sbin/06-postgrest-build.bash
    owner: root:root
    permissions: '0775'
  - content: |
      #!/bin/bash
      cd
      sudo apt update && sudo apt upgrade -y
      sudo apt -y install curl lsb-release gnupg2 ca-certificates
      curl -fsSL https://artifacts.opensearch.org/publickeys/opensearch.pgp| \
                sudo gpg --dearmor -o /etc/apt/trusted.gpg.d/opensearch.gpg
      echo "deb https://artifacts.opensearch.org/releases/bundle/opensearch/2.x/apt stable main" | \
                 sudo tee /etc/apt/sources.list.d/opensearch-2.x.list
      echo "deb https://artifacts.opensearch.org/releases/bundle/opensearch-dashboards/2.x/apt stable main" |\
                 sudo tee -a /etc/apt/sources.list.d/opensearch-2.x.list
      sudo apt update
      sudo apt list -a opensearch
      sudo apt install opensearch=2.5.0
      sudo systemctl enable opensearch
      sudo systemctl restart opensearch
      sudo systemctl status opensearch
      cat <<EOT
         Connect to opensearch on port 9200. By default it is configured with
         SSL and a self signed cert. You need to access v2.5.0 with the default
         credentials of 'admin:admin'.

             curl -k --user admin:admin https://localhost:9200/_settings

      EOT
      # Now install elasticdump to dump and restore indexes
      sudo apt install nodejs npm
      sudo npm install n -g
      sudo n latest
      sudo npm install elasticdump -g
      elasticdump --version
      if [ -e /var/run/reboot-required ]; then echo "Rebooting VM"; sudo reboot; fi
    path: /usr/local/sbin/07-add-opensearch.bash
    owner: root:root
    permissions: '0775'
package_upgrade: true
# Install additional packages on first boot
#
# Default: none
#
# if packages are specified, then package_update will be set to true
#
# packages may be supplied as a single package name or as a list
# with the format [<package>, <version>] wherein the specific
# package version will be installed.
packages:
#  - ubuntu-desktop
  - curl
  - openjdk-11-jdk
  - lsb-release
  - gnupg2
  - ca-certificates
  - aptitude
  - tmux
  - micro
  - vim
  - emacs
  - tcsh
  - build-essential
  - autotools-dev
  - autoconf
  - automake
  - git
  - curl
  - lynx
  - make
  - pkg-config
  - postgresql-16
  - postgresql-16-postgis
  - postgresql-16-postgis-scripts
  - postgresql-16-mysql-fdw
  - postgresql-plpython3-16
  - postgresql-16-pllua
  - postgresql-16-plpgsql-check
  - postgresql-16-semver
  - pgloader
  - pgtop
  - libncurses-dev
  - libffi-dev
  - libgmp-dev
  - libncurses5
  - zlib1g-dev
  - shellcheck
  - python3
  - python3-pip
  - libffi-dev
  - libffi6
  - libgmp-dev
  - libgmp10
  - libncurses-dev
  - libncurses5
  - libtinfo5
  - npm
  - jq