#!/usr/bin/env bash

# calculate_qc_metrics v4.1.0
# 
# This wrapper script is auto-generated by viash 0.9.7 and is thus a derivative
# work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
# Intuitive.
# 
# The component may contain files which fall under a different license. The
# authors of this component should specify the license in the header of such
# files, or include a separate license file detailing the licenses of all included
# files.
# 
# Component authors:
#  * Dries Schaumont (author)

set -e

if [ -z "$VIASH_TEMP" ]; then
  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TMPDIR}
  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TEMPDIR}
  VIASH_TEMP=${VIASH_TEMP:-$VIASH_TMP}
  VIASH_TEMP=${VIASH_TEMP:-$TMPDIR}
  VIASH_TEMP=${VIASH_TEMP:-$TMP}
  VIASH_TEMP=${VIASH_TEMP:-$TEMPDIR}
  VIASH_TEMP=${VIASH_TEMP:-$TEMP}
  VIASH_TEMP=${VIASH_TEMP:-/tmp}
fi

# define helper functions
# ViashQuote: put quotes around non flag values
# $1     : unquoted string
# return : possibly quoted string
# examples:
#   ViashQuote --foo      # returns --foo
#   ViashQuote bar        # returns 'bar'
#   Viashquote --foo=bar  # returns --foo='bar'
function ViashQuote {
  if [[ "$1" =~ ^-+[a-zA-Z0-9_\-]+=.+$ ]]; then
    echo "$1" | sed "s#=\(.*\)#='\1'#"
  elif [[ "$1" =~ ^-+[a-zA-Z0-9_\-]+$ ]]; then
    echo "$1"
  else
    echo "'$1'"
  fi
}
# ViashRemoveFlags: Remove leading flag
# $1     : string with a possible leading flag
# return : string without possible leading flag
# examples:
#   ViashRemoveFlags --foo=bar  # returns bar
function ViashRemoveFlags {
  echo "$1" | sed 's/^--*[a-zA-Z0-9_\-]*=//'
}
# ViashSourceDir: return the path of a bash file, following symlinks
# usage   : ViashSourceDir ${BASH_SOURCE[0]}
# $1      : Should always be set to ${BASH_SOURCE[0]}
# returns : The absolute path of the bash file
function ViashSourceDir {
  local source="$1"
  while [ -h "$source" ]; do
    local dir="$( cd -P "$( dirname "$source" )" >/dev/null 2>&1 && pwd )"
    source="$(readlink "$source")"
    [[ $source != /* ]] && source="$dir/$source"
  done
  cd -P "$( dirname "$source" )" >/dev/null 2>&1 && pwd
}
# ViashFindTargetDir: return the path of the '.build.yaml' file, following symlinks
# usage   : ViashFindTargetDir 'ScriptPath'
# $1      : The location from where to start the upward search
# returns : The absolute path of the '.build.yaml' file
function ViashFindTargetDir {
  local source="$1"
  while [[ "$source" != "" && ! -e "$source/.build.yaml" ]]; do
    source=${source%/*}
  done
  echo $source
}
# see https://en.wikipedia.org/wiki/Syslog#Severity_level
VIASH_LOGCODE_EMERGENCY=0
VIASH_LOGCODE_ALERT=1
VIASH_LOGCODE_CRITICAL=2
VIASH_LOGCODE_ERROR=3
VIASH_LOGCODE_WARNING=4
VIASH_LOGCODE_NOTICE=5
VIASH_LOGCODE_INFO=6
VIASH_LOGCODE_DEBUG=7
VIASH_VERBOSITY=$VIASH_LOGCODE_NOTICE

# ViashLog: Log events depending on the verbosity level
# usage: ViashLog 1 alert Oh no something went wrong!
# $1: required verbosity level
# $2: display tag
# $3+: messages to display
# stdout: Your input, prepended by '[$2] '.
function ViashLog {
  local required_level="$1"
  local display_tag="$2"
  shift 2
  if [ $VIASH_VERBOSITY -ge $required_level ]; then
    >&2 echo "[$display_tag]" "$@"
  fi
}

# ViashEmergency: log events when the system is unstable
# usage: ViashEmergency Oh no something went wrong.
# stdout: Your input, prepended by '[emergency] '.
function ViashEmergency {
  ViashLog $VIASH_LOGCODE_EMERGENCY emergency "$@"
}

# ViashAlert: log events when actions must be taken immediately (e.g. corrupted system database)
# usage: ViashAlert Oh no something went wrong.
# stdout: Your input, prepended by '[alert] '.
function ViashAlert {
  ViashLog $VIASH_LOGCODE_ALERT alert "$@"
}

# ViashCritical: log events when a critical condition occurs
# usage: ViashCritical Oh no something went wrong.
# stdout: Your input, prepended by '[critical] '.
function ViashCritical {
  ViashLog $VIASH_LOGCODE_CRITICAL critical "$@"
}

# ViashError: log events when an error condition occurs
# usage: ViashError Oh no something went wrong.
# stdout: Your input, prepended by '[error] '.
function ViashError {
  ViashLog $VIASH_LOGCODE_ERROR error "$@"
}

# ViashWarning: log potentially abnormal events
# usage: ViashWarning Something may have gone wrong.
# stdout: Your input, prepended by '[warning] '.
function ViashWarning {
  ViashLog $VIASH_LOGCODE_WARNING warning "$@"
}

# ViashNotice: log significant but normal events
# usage: ViashNotice This just happened.
# stdout: Your input, prepended by '[notice] '.
function ViashNotice {
  ViashLog $VIASH_LOGCODE_NOTICE notice "$@"
}

# ViashInfo: log normal events
# usage: ViashInfo This just happened.
# stdout: Your input, prepended by '[info] '.
function ViashInfo {
  ViashLog $VIASH_LOGCODE_INFO info "$@"
}

# ViashDebug: log all events, for debugging purposes
# usage: ViashDebug This just happened.
# stdout: Your input, prepended by '[debug] '.
function ViashDebug {
  ViashLog $VIASH_LOGCODE_DEBUG debug "$@"
}

# find source folder of this component
VIASH_META_RESOURCES_DIR=`ViashSourceDir ${BASH_SOURCE[0]}`

# find the root of the built components & dependencies
VIASH_TARGET_DIR=`ViashFindTargetDir $VIASH_META_RESOURCES_DIR`

# define meta fields
VIASH_META_NAME="calculate_qc_metrics"
VIASH_META_FUNCTIONALITY_NAME="calculate_qc_metrics"
VIASH_META_EXECUTABLE="$VIASH_META_RESOURCES_DIR/$VIASH_META_NAME"
VIASH_META_CONFIG="$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
VIASH_META_TEMP_DIR="$VIASH_TEMP"



# initialise variables
VIASH_MODE='run'
VIASH_ENGINE_ID='docker'

######## Helper functions for setting up Docker images for viash ########
# expects: ViashDockerBuild

# ViashDockerInstallationCheck: check whether Docker is installed correctly
#
# examples:
#   ViashDockerInstallationCheck
function ViashDockerInstallationCheck {
  ViashDebug "Checking whether Docker is installed"
  if [ ! command -v docker &> /dev/null ]; then
    ViashCritical "Docker doesn't seem to be installed. See 'https://docs.docker.com/get-docker/' for instructions."
    exit 1
  fi

  ViashDebug "Checking whether the Docker daemon is running"
  local save=$-; set +e
  local docker_version=$(docker version --format '{{.Client.APIVersion}}' 2> /dev/null)
  local out=$?
  [[ $save =~ e ]] && set -e
  if [ $out -ne 0 ]; then
    ViashCritical "Docker daemon does not seem to be running. Try one of the following:"
    ViashCritical "- Try running 'dockerd' in the command line"
    ViashCritical "- See https://docs.docker.com/config/daemon/"
    exit 1
  fi
}

# ViashDockerRemoteTagCheck: check whether a Docker image is available 
# on a remote. Assumes `docker login` has been performed, if relevant.
#
# $1                  : image identifier with format `[registry/]image[:tag]`
# exit code $?        : whether or not the image was found
# examples:
#   ViashDockerRemoteTagCheck python:latest
#   echo $?                                     # returns '0'
#   ViashDockerRemoteTagCheck sdaizudceahifu
#   echo $?                                     # returns '1'
function ViashDockerRemoteTagCheck {
  docker manifest inspect $1 > /dev/null 2> /dev/null
}

# ViashDockerLocalTagCheck: check whether a Docker image is available locally
#
# $1                  : image identifier with format `[registry/]image[:tag]`
# exit code $?        : whether or not the image was found
# examples:
#   docker pull python:latest
#   ViashDockerLocalTagCheck python:latest
#   echo $?                                     # returns '0'
#   ViashDockerLocalTagCheck sdaizudceahifu
#   echo $?                                     # returns '1'
function ViashDockerLocalTagCheck {
  [ -n "$(docker images -q $1)" ]
}

# ViashDockerPull: pull a Docker image
#
# $1                  : image identifier with format `[registry/]image[:tag]`
# exit code $?        : whether or not the image was found
# examples:
#   ViashDockerPull python:latest
#   echo $?                                     # returns '0'
#   ViashDockerPull sdaizudceahifu
#   echo $?                                     # returns '1'
function ViashDockerPull {
  ViashNotice "Checking if Docker image is available at '$1'"
  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
    docker pull $1 && return 0 || return 1
  else
    local save=$-; set +e
    docker pull $1 2> /dev/null > /dev/null
    local out=$?
    [[ $save =~ e ]] && set -e
    if [ $out -ne 0 ]; then
      ViashWarning "Could not pull from '$1'. Docker image doesn't exist or is not accessible."
    fi
    return $out
  fi
}

# ViashDockerPush: push a Docker image
#
# $1                  : image identifier with format `[registry/]image[:tag]`
# exit code $?        : whether or not the image was found
# examples:
#   ViashDockerPush python:latest
#   echo $?                                     # returns '0'
#   ViashDockerPush sdaizudceahifu
#   echo $?                                     # returns '1'
function ViashDockerPush {
  ViashNotice "Pushing image to '$1'"
  local save=$-; set +e
  local out
  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
    docker push $1
    out=$?
  else
    docker push $1 2> /dev/null > /dev/null
    out=$?
  fi
  [[ $save =~ e ]] && set -e
  if [ $out -eq 0 ]; then
    ViashNotice "Container '$1' push succeeded."
  else
    ViashError "Container '$1' push errored. You might not be logged in or have the necessary permissions."
  fi
  return $out
}

# ViashDockerPullElseBuild: pull a Docker image, else build it
#
# $1                  : image identifier with format `[registry/]image[:tag]`
# ViashDockerBuild    : a Bash function which builds a docker image, takes image identifier as argument.
# examples:
#   ViashDockerPullElseBuild mynewcomponent
function ViashDockerPullElseBuild {
  local save=$-; set +e
  ViashDockerPull $1
  local out=$?
  [[ $save =~ e ]] && set -e
  if [ $out -ne 0 ]; then
    ViashDockerBuild $@
  fi
}

# ViashDockerSetup: create a Docker image, according to specified docker setup strategy
#
# $1          : image identifier with format `[registry/]image[:tag]`
# $2          : docker setup strategy, see DockerSetupStrategy.scala
# examples:
#   ViashDockerSetup mynewcomponent alwaysbuild
function ViashDockerSetup {
  local image_id="$1"
  local setup_strategy="$2"
  if [ "$setup_strategy" == "alwaysbuild" -o "$setup_strategy" == "build" -o "$setup_strategy" == "b" ]; then
    ViashDockerBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
  elif [ "$setup_strategy" == "alwayspull" -o "$setup_strategy" == "pull" -o "$setup_strategy" == "p" ]; then
    ViashDockerPull $image_id
  elif [ "$setup_strategy" == "alwayspullelsebuild" -o "$setup_strategy" == "pullelsebuild" ]; then
    ViashDockerPullElseBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
  elif [ "$setup_strategy" == "alwayspullelsecachedbuild" -o "$setup_strategy" == "pullelsecachedbuild" ]; then
    ViashDockerPullElseBuild $image_id $(ViashDockerBuildArgs "$engine_id")
  elif [ "$setup_strategy" == "alwayscachedbuild" -o "$setup_strategy" == "cachedbuild" -o "$setup_strategy" == "cb" ]; then
    ViashDockerBuild $image_id $(ViashDockerBuildArgs "$engine_id")
  elif [[ "$setup_strategy" =~ ^ifneedbe ]]; then
    local save=$-; set +e
    ViashDockerLocalTagCheck $image_id
    local outCheck=$?
    [[ $save =~ e ]] && set -e
    if [ $outCheck -eq 0 ]; then
      ViashInfo "Image $image_id already exists"
    elif [ "$setup_strategy" == "ifneedbebuild" ]; then
      ViashDockerBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
    elif [ "$setup_strategy" == "ifneedbecachedbuild" ]; then
      ViashDockerBuild $image_id $(ViashDockerBuildArgs "$engine_id")
    elif [ "$setup_strategy" == "ifneedbepull" ]; then
      ViashDockerPull $image_id
    elif [ "$setup_strategy" == "ifneedbepullelsebuild" ]; then
      ViashDockerPullElseBuild $image_id --no-cache $(ViashDockerBuildArgs "$engine_id")
    elif [ "$setup_strategy" == "ifneedbepullelsecachedbuild" ]; then
      ViashDockerPullElseBuild $image_id $(ViashDockerBuildArgs "$engine_id")
    else
      ViashError "Unrecognised Docker strategy: $setup_strategy"
      exit 1
    fi
  elif [ "$setup_strategy" == "push" -o "$setup_strategy" == "forcepush" -o "$setup_strategy" == "alwayspush" ]; then
    ViashDockerPush "$image_id"
  elif [ "$setup_strategy" == "pushifnotpresent" -o "$setup_strategy" == "gentlepush" -o "$setup_strategy" == "maybepush" ]; then
    local save=$-; set +e
    ViashDockerRemoteTagCheck $image_id
    local outCheck=$?
    [[ $save =~ e ]] && set -e
    if [ $outCheck -eq 0 ]; then
      ViashNotice "Container '$image_id' exists, doing nothing."
    else
      ViashNotice "Container '$image_id' does not yet exist."
      ViashDockerPush "$image_id"
    fi
  elif [ "$setup_strategy" == "donothing" -o "$setup_strategy" == "meh" ]; then
    ViashNotice "Skipping setup."
  else
    ViashError "Unrecognised Docker strategy: $setup_strategy"
    exit 1
  fi
}

# ViashDockerCheckCommands: Check whether a docker container has the required commands
#
# $1                  : image identifier with format `[registry/]image[:tag]`
# $@                  : commands to verify being present
# examples:
#   ViashDockerCheckCommands bash:4.0 bash ps foo
function ViashDockerCheckCommands {
  local image_id="$1"
  shift 1
  local commands="$@"
  local save=$-; set +e
  local missing # mark 'missing' as local in advance, otherwise the exit code of the command will be missing and always be '0'
  missing=$(docker run --rm --entrypoint=sh "$image_id" -c "for command in $commands; do command -v \$command >/dev/null 2>&1; if [ \$? -ne 0 ]; then echo \$command; exit 1; fi; done")
  local outCheck=$?
  [[ $save =~ e ]] && set -e
  if [ $outCheck -ne 0 ]; then
  	ViashError "Docker container '$image_id' does not contain command '$missing'."
  	exit 1
  fi
}

# ViashDockerBuild: build a docker image
# $1                               : image identifier with format `[registry/]image[:tag]`
# $...                             : additional arguments to pass to docker build
# $VIASH_META_TEMP_DIR             : temporary directory to store dockerfile & optional resources in
# $VIASH_META_NAME                 : name of the component
# $VIASH_META_RESOURCES_DIR        : directory containing the resources
# $VIASH_VERBOSITY                 : verbosity level
# exit code $?                     : whether or not the image was built successfully
function ViashDockerBuild {
  local image_id="$1"
  shift 1

  # create temporary directory to store dockerfile & optional resources in
  local tmpdir=$(mktemp -d "$VIASH_META_TEMP_DIR/dockerbuild-$VIASH_META_NAME-XXXXXX")
  local dockerfile="$tmpdir/Dockerfile"
  function clean_up {
    rm -rf "$tmpdir"
  }
  trap clean_up EXIT

  # store dockerfile and resources
  ViashDockerfile "$VIASH_ENGINE_ID" > "$dockerfile"

  # generate the build command
  local docker_build_cmd="docker build -t '$image_id' $@ '$VIASH_META_RESOURCES_DIR' -f '$dockerfile'"

  # build the container
  ViashNotice "Building container '$image_id' with Dockerfile"
  ViashInfo "$docker_build_cmd"
  local save=$-; set +e
  if [ $VIASH_VERBOSITY -ge $VIASH_LOGCODE_INFO ]; then
    eval $docker_build_cmd
  else
    eval $docker_build_cmd &> "$tmpdir/docker_build.log"
  fi

  # check exit code
  local out=$?
  [[ $save =~ e ]] && set -e
  if [ $out -ne 0 ]; then
    ViashError "Error occurred while building container '$image_id'"
    if [ $VIASH_VERBOSITY -lt $VIASH_LOGCODE_INFO ]; then
      ViashError "Transcript: --------------------------------"
      cat "$tmpdir/docker_build.log"
      ViashError "End of transcript --------------------------"
    fi
    exit 1
  fi
}

######## End of helper functions for setting up Docker images for viash ########

# ViashDockerFile: print the dockerfile to stdout
# $1    : engine identifier
# return : dockerfile required to run this component
# examples:
#   ViashDockerFile
function ViashDockerfile {
  local engine_id="$1"

  if [[ "$engine_id" == "docker" ]]; then
    cat << 'VIASHDOCKER'
FROM python:3.13-slim
ENTRYPOINT []
RUN apt-get update && \
  DEBIAN_FRONTEND=noninteractive apt-get install -y procps && \
  rm -rf /var/lib/apt/lists/*

RUN pip install --upgrade pip && \
  pip install --upgrade --no-cache-dir "anndata~=0.12.16" "awkward" "scipy~=1.17.1" "mudata~=0.3.8" "scipy" && \
  python -c 'exec("try:\n  import zarr; from importlib.metadata import version\nexcept ModuleNotFoundError:\n  exit(0)\nelse:  assert int(version(\"zarr\").partition(\".\")[0]) > 2")'

LABEL org.opencontainers.image.authors="Dries Schaumont"
LABEL org.opencontainers.image.description="Companion container for running component qc calculate_qc_metrics"
LABEL org.opencontainers.image.created="2026-05-27T07:38:53Z"
LABEL org.opencontainers.image.source="https://github.com/openpipelines-bio/openpipeline"
LABEL org.opencontainers.image.revision="a6499fddaa8553874585c747370f919f4d7b729c"
LABEL org.opencontainers.image.version="v4.1.0"

VIASHDOCKER
  fi
}

# ViashDockerBuildArgs: return the arguments to pass to docker build
# $1    : engine identifier
# return : arguments to pass to docker build
function ViashDockerBuildArgs {
  local engine_id="$1"

  if [[ "$engine_id" == "docker" ]]; then
    echo ""
  fi
}

# ViashAbsolutePath: generate absolute path from relative path
# borrowed from https://stackoverflow.com/a/21951256
# $1     : relative filename
# return : absolute path
# examples:
#   ViashAbsolutePath some_file.txt   # returns /path/to/some_file.txt
#   ViashAbsolutePath /foo/bar/..     # returns /foo
function ViashAbsolutePath {
  local thePath
  local parr
  local outp
  local len
  if [[ ! "$1" =~ ^/ ]]; then
    thePath="$PWD/$1"
  else
    thePath="$1"
  fi
  echo "$thePath" | (
    IFS=/
    read -a parr
    declare -a outp
    for i in "${parr[@]}"; do
      case "$i" in
      ''|.) continue ;;
      ..)
        len=${#outp[@]}
        if ((len==0)); then
          continue
        else
          unset outp[$((len-1))]
        fi
        ;;
      *)
        len=${#outp[@]}
        outp[$len]="$i"
      ;;
      esac
    done
    echo /"${outp[*]}"
  )
}
# ViashDockerAutodetectMount: auto configuring docker mounts from parameters
# $1                             : The parameter value
# returns                        : New parameter
# $VIASH_DIRECTORY_MOUNTS        : Added another parameter to be passed to docker
# $VIASH_DOCKER_AUTOMOUNT_PREFIX : The prefix to be used for the automounts
# examples:
#   ViashDockerAutodetectMount /path/to/bar      # returns '/viash_automount/path/to/bar'
#   ViashDockerAutodetectMountArg /path/to/bar   # returns '--volume="/path/to:/viash_automount/path/to"'
function ViashDockerAutodetectMount {
  local abs_path=$(ViashAbsolutePath "$1")
  local mount_source
  local base_name
  if [ -d "$abs_path" ]; then
    mount_source="$abs_path"
    base_name=""
  else
    mount_source=`dirname "$abs_path"`
    base_name=`basename "$abs_path"`
  fi
  local mount_target="$VIASH_DOCKER_AUTOMOUNT_PREFIX$mount_source"
  if [ -z "$base_name" ]; then
    echo "$mount_target"
  else
    echo "$mount_target/$base_name"
  fi
}
function ViashDockerAutodetectMountArg {
  local abs_path=$(ViashAbsolutePath "$1")
  local mount_source
  local base_name
  if [ -d "$abs_path" ]; then
    mount_source="$abs_path"
    base_name=""
  else
    mount_source=`dirname "$abs_path"`
    base_name=`basename "$abs_path"`
  fi
  local mount_target="$VIASH_DOCKER_AUTOMOUNT_PREFIX$mount_source"
  ViashDebug "ViashDockerAutodetectMountArg $1 -> $mount_source -> $mount_target"
  echo "--volume=\"$mount_source:$mount_target\""
}
function ViashDockerStripAutomount {
  local abs_path=$(ViashAbsolutePath "$1")
  echo "${abs_path#$VIASH_DOCKER_AUTOMOUNT_PREFIX}"
}
# initialise variables
VIASH_DIRECTORY_MOUNTS=()

# configure default docker automount prefix if it is unset
if [ -z "${VIASH_DOCKER_AUTOMOUNT_PREFIX+x}" ]; then
  VIASH_DOCKER_AUTOMOUNT_PREFIX="/viash_automount"
fi

# initialise docker variables
VIASH_DOCKER_RUN_ARGS=(-i --rm)


# ViashHelp: Display helpful explanation about this executable
function ViashHelp {
  echo "calculate_qc_metrics v4.1.0"
  echo ""
  echo "Add basic quality control metrics to an .h5mu file."
  echo ""
  echo "The metrics are comparable to what scanpy.pp.calculate_qc_metrics output,"
  echo "although they have slightly different names:"
  echo ""
  echo "Var metrics (name in this component -> name in scanpy):"
  echo "  - pct_dropout -> pct_dropout_by_{expr_type}"
  echo "  - num_nonzero_obs -> n_cells_by_{expr_type}"
  echo "  - obs_mean -> mean_{expr_type}"
  echo "  - total_counts -> total_{expr_type}"
  echo ""
  echo "  Obs metrics:"
  echo "    - num_nonzero_vars -> n_genes_by_{expr_type}"
  echo "    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}"
  echo "    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}"
  echo "    - pct_of_counts_in_top_{top_n_vars}_vars ->"
  echo "pct_{expr_type}_in_top_{n}_{var_type}"
  echo "    - total_counts -> total_{expr_type}"
  echo ""
  echo "Inputs:"
  echo "    --input"
  echo "        type: file, required parameter, file must exist"
  echo "        example: input.h5mu"
  echo "        Input data. Can be encoded in H5 or Zarr (v3) format."
  echo ""
  echo "    --modality"
  echo "        type: string"
  echo "        default: rna"
  echo "        Which modality from the input MuData file to process."
  echo ""
  echo "    --layer"
  echo "        type: string"
  echo "        example: raw_counts"
  echo "        Layer from modality to use as input data. If not provided the .X"
  echo "        attribute is used."
  echo ""
  echo "Metrics added to .obs:"
  echo "    --var_qc_metrics"
  echo "        type: string, multiple values allowed"
  echo "        example: ercc,highly_variable,mitochondrial"
  echo "        Keys to select a boolean (containing only True or False) column from"
  echo "        .var."
  echo "        For each cell, calculate the proportion of total values for genes which"
  echo "        are labeled 'True',"
  echo "        compared to the total sum of the values for all genes."
  echo ""
  echo "    --var_qc_metrics_fill_na_value"
  echo "        type: boolean"
  echo "        Fill any 'NA' values found in the columns specified with"
  echo "        --var_qc_metrics to 'True' or 'False'."
  echo "        as False."
  echo ""
  echo "    --top_n_vars"
  echo "        type: integer, multiple values allowed"
  echo "        Number of top vars to be used to calculate cumulative proportions."
  echo "        If not specified, proportions are not calculated. \`--top_n_vars 20;50\`"
  echo "        finds"
  echo "        cumulative proportion to the 20th and 50th most expressed vars."
  echo ""
  echo "    --output_obs_num_nonzero_vars"
  echo "        type: string"
  echo "        default: num_nonzero_vars"
  echo "        Name of column in .obs describing, for each observation, the number of"
  echo "        stored values"
  echo "        (including explicit zeroes). In other words, the name of the column that"
  echo "        counts"
  echo "        for each row the number of columns that contain data."
  echo ""
  echo "    --output_obs_total_counts_vars"
  echo "        type: string"
  echo "        default: total_counts"
  echo "        Name of the column for .obs describing, for each observation (row),"
  echo "        the sum of the stored values in the columns."
  echo ""
  echo "Metrics added to .var:"
  echo "    --output_var_num_nonzero_obs"
  echo "        type: string"
  echo "        default: num_nonzero_obs"
  echo "        Name of column describing, for each feature, the number of stored values"
  echo "        (including explicit zeroes). In other words, the name of the column that"
  echo "        counts"
  echo "        for each column the number of rows that contain data."
  echo ""
  echo "    --output_var_total_counts_obs"
  echo "        type: string"
  echo "        default: total_counts"
  echo "        Name of the column in .var describing, for each feature (column),"
  echo "        the sum of the stored values in the rows."
  echo ""
  echo "    --output_var_obs_mean"
  echo "        type: string"
  echo "        default: obs_mean"
  echo "        Name of the column in .obs providing the mean of the values in each row."
  echo ""
  echo "    --output_var_pct_dropout"
  echo "        type: string"
  echo "        default: pct_dropout"
  echo "        Name of the column in .obs providing for each feature the percentage of"
  echo "        observations the feature does not appear on (i.e. is missing). Same as"
  echo "        \`--num_nonzero_obs\`"
  echo "        but percentage based."
  echo ""
  echo "Outputs:"
  echo "    --output"
  echo "        type: file, output, file must exist"
  echo "        example: output.h5mu"
  echo "        Output data. Can be encoded using H5 or Zarr v3, matching the format of"
  echo "        the input data."
  echo ""
  echo "    --output_compression"
  echo "        type: string"
  echo "        example: gzip"
  echo "        choices: [ gzip, lzf ]"
  echo "        Compression format to use for the output AnnData and/or Mudata H5 files."
  echo "        By default no compression is applied."
  echo ""
  echo "Viash built in Computational Requirements:"
  echo "    ---cpus=INT"
  echo "        Number of CPUs to use"
  echo "    ---memory=STRING"
  echo "        Amount of memory to use. Examples: 4GB, 3MiB."
  echo ""
  echo "Viash built in Docker:"
  echo "    ---setup=STRATEGY"
  echo "        Setup the docker container. Options are: alwaysbuild, alwayscachedbuild, ifneedbebuild, ifneedbecachedbuild, alwayspull, alwayspullelsebuild, alwayspullelsecachedbuild, ifneedbepull, ifneedbepullelsebuild, ifneedbepullelsecachedbuild, push, pushifnotpresent, donothing."
  echo "        Default: ifneedbepullelsecachedbuild"
  echo "    ---dockerfile"
  echo "        Print the dockerfile to stdout."
  echo "    ---docker_run_args=ARG"
  echo "        Provide runtime arguments to Docker. See the documentation on \`docker run\` for more information."
  echo "    ---docker_image_id"
  echo "        Print the docker image id to stdout."
  echo "    ---debug"
  echo "        Enter the docker container for debugging purposes."
  echo ""
  echo "Viash built in Engines:"
  echo "    ---engine=ENGINE_ID"
  echo "        Specify the engine to use. Options are: docker, native."
  echo "        Default: docker"
}

# initialise array
VIASH_POSITIONAL_ARGS=''

while [[ $# -gt 0 ]]; do
    case "$1" in
        -h|--help)
            ViashHelp
            exit
            ;;
        ---v|---verbose)
            let "VIASH_VERBOSITY=VIASH_VERBOSITY+1"
            shift 1
            ;;
        ---verbosity)
            VIASH_VERBOSITY="$2"
            shift 2
            ;;
        ---verbosity=*)
            VIASH_VERBOSITY="$(ViashRemoveFlags "$1")"
            shift 1
            ;;
        --version)
            echo "calculate_qc_metrics v4.1.0"
            exit
            ;;
        --input)
            [ -n "$VIASH_PAR_INPUT" ] && ViashError Bad arguments for option \'--input\': \'$VIASH_PAR_INPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_INPUT="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --input. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --input=*)
            [ -n "$VIASH_PAR_INPUT" ] && ViashError Bad arguments for option \'--input=*\': \'$VIASH_PAR_INPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_INPUT=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --modality)
            [ -n "$VIASH_PAR_MODALITY" ] && ViashError Bad arguments for option \'--modality\': \'$VIASH_PAR_MODALITY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_MODALITY="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --modality. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --modality=*)
            [ -n "$VIASH_PAR_MODALITY" ] && ViashError Bad arguments for option \'--modality=*\': \'$VIASH_PAR_MODALITY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_MODALITY=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --layer)
            [ -n "$VIASH_PAR_LAYER" ] && ViashError Bad arguments for option \'--layer\': \'$VIASH_PAR_LAYER\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_LAYER="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --layer. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --layer=*)
            [ -n "$VIASH_PAR_LAYER" ] && ViashError Bad arguments for option \'--layer=*\': \'$VIASH_PAR_LAYER\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_LAYER=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --var_qc_metrics)
            if [ -z "$VIASH_PAR_VAR_QC_METRICS" ]; then
              VIASH_PAR_VAR_QC_METRICS="$2"
            else
              VIASH_PAR_VAR_QC_METRICS="$VIASH_PAR_VAR_QC_METRICS;""$2"
            fi
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --var_qc_metrics. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --var_qc_metrics=*)
            if [ -z "$VIASH_PAR_VAR_QC_METRICS" ]; then
              VIASH_PAR_VAR_QC_METRICS=$(ViashRemoveFlags "$1")
            else
              VIASH_PAR_VAR_QC_METRICS="$VIASH_PAR_VAR_QC_METRICS;"$(ViashRemoveFlags "$1")
            fi
            shift 1
            ;;
        --var_qc_metrics_fill_na_value)
            [ -n "$VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE" ] && ViashError Bad arguments for option \'--var_qc_metrics_fill_na_value\': \'$VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --var_qc_metrics_fill_na_value. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --var_qc_metrics_fill_na_value=*)
            [ -n "$VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE" ] && ViashError Bad arguments for option \'--var_qc_metrics_fill_na_value=*\': \'$VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --top_n_vars)
            if [ -z "$VIASH_PAR_TOP_N_VARS" ]; then
              VIASH_PAR_TOP_N_VARS="$2"
            else
              VIASH_PAR_TOP_N_VARS="$VIASH_PAR_TOP_N_VARS;""$2"
            fi
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --top_n_vars. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --top_n_vars=*)
            if [ -z "$VIASH_PAR_TOP_N_VARS" ]; then
              VIASH_PAR_TOP_N_VARS=$(ViashRemoveFlags "$1")
            else
              VIASH_PAR_TOP_N_VARS="$VIASH_PAR_TOP_N_VARS;"$(ViashRemoveFlags "$1")
            fi
            shift 1
            ;;
        --output_obs_num_nonzero_vars)
            [ -n "$VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS" ] && ViashError Bad arguments for option \'--output_obs_num_nonzero_vars\': \'$VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_obs_num_nonzero_vars. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --output_obs_num_nonzero_vars=*)
            [ -n "$VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS" ] && ViashError Bad arguments for option \'--output_obs_num_nonzero_vars=*\': \'$VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --output_obs_total_counts_vars)
            [ -n "$VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS" ] && ViashError Bad arguments for option \'--output_obs_total_counts_vars\': \'$VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_obs_total_counts_vars. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --output_obs_total_counts_vars=*)
            [ -n "$VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS" ] && ViashError Bad arguments for option \'--output_obs_total_counts_vars=*\': \'$VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --output_var_num_nonzero_obs)
            [ -n "$VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS" ] && ViashError Bad arguments for option \'--output_var_num_nonzero_obs\': \'$VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_var_num_nonzero_obs. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --output_var_num_nonzero_obs=*)
            [ -n "$VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS" ] && ViashError Bad arguments for option \'--output_var_num_nonzero_obs=*\': \'$VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --output_var_total_counts_obs)
            [ -n "$VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS" ] && ViashError Bad arguments for option \'--output_var_total_counts_obs\': \'$VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_var_total_counts_obs. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --output_var_total_counts_obs=*)
            [ -n "$VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS" ] && ViashError Bad arguments for option \'--output_var_total_counts_obs=*\': \'$VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --output_var_obs_mean)
            [ -n "$VIASH_PAR_OUTPUT_VAR_OBS_MEAN" ] && ViashError Bad arguments for option \'--output_var_obs_mean\': \'$VIASH_PAR_OUTPUT_VAR_OBS_MEAN\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_VAR_OBS_MEAN="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_var_obs_mean. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --output_var_obs_mean=*)
            [ -n "$VIASH_PAR_OUTPUT_VAR_OBS_MEAN" ] && ViashError Bad arguments for option \'--output_var_obs_mean=*\': \'$VIASH_PAR_OUTPUT_VAR_OBS_MEAN\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_VAR_OBS_MEAN=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --output_var_pct_dropout)
            [ -n "$VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT" ] && ViashError Bad arguments for option \'--output_var_pct_dropout\': \'$VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_var_pct_dropout. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --output_var_pct_dropout=*)
            [ -n "$VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT" ] && ViashError Bad arguments for option \'--output_var_pct_dropout=*\': \'$VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --output)
            [ -n "$VIASH_PAR_OUTPUT" ] && ViashError Bad arguments for option \'--output\': \'$VIASH_PAR_OUTPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --output=*)
            [ -n "$VIASH_PAR_OUTPUT" ] && ViashError Bad arguments for option \'--output=*\': \'$VIASH_PAR_OUTPUT\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        --output_compression)
            [ -n "$VIASH_PAR_OUTPUT_COMPRESSION" ] && ViashError Bad arguments for option \'--output_compression\': \'$VIASH_PAR_OUTPUT_COMPRESSION\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_COMPRESSION="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to --output_compression. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        --output_compression=*)
            [ -n "$VIASH_PAR_OUTPUT_COMPRESSION" ] && ViashError Bad arguments for option \'--output_compression=*\': \'$VIASH_PAR_OUTPUT_COMPRESSION\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_PAR_OUTPUT_COMPRESSION=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        ---engine)
            VIASH_ENGINE_ID="$2"
            shift 2
            ;;
        ---engine=*)
            VIASH_ENGINE_ID="$(ViashRemoveFlags "$1")"
            shift 1
            ;;
        ---setup)
            VIASH_MODE='setup'
            VIASH_SETUP_STRATEGY="$2"
            shift 2
            ;;
        ---setup=*)
            VIASH_MODE='setup'
            VIASH_SETUP_STRATEGY="$(ViashRemoveFlags "$1")"
            shift 1
            ;;
        ---dockerfile)
            VIASH_MODE='dockerfile'
            shift 1
            ;;
        ---docker_run_args)
            VIASH_DOCKER_RUN_ARGS+=("$2")
            shift 2
            ;;
        ---docker_run_args=*)
            VIASH_DOCKER_RUN_ARGS+=("$(ViashRemoveFlags "$1")")
            shift 1
            ;;
        ---docker_image_id)
            VIASH_MODE='docker_image_id'
            shift 1
            ;;
        ---debug)
            VIASH_MODE='debug'
            shift 1
            ;;
        ---cpus)
            [ -n "$VIASH_META_CPUS" ] && ViashError Bad arguments for option \'---cpus\': \'$VIASH_META_CPUS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_META_CPUS="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to ---cpus. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        ---cpus=*)
            [ -n "$VIASH_META_CPUS" ] && ViashError Bad arguments for option \'---cpus=*\': \'$VIASH_META_CPUS\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_META_CPUS=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        ---memory)
            [ -n "$VIASH_META_MEMORY" ] && ViashError Bad arguments for option \'---memory\': \'$VIASH_META_MEMORY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_META_MEMORY="$2"
            [ $# -lt 2 ] && ViashError Not enough arguments passed to ---memory. Use "--help" to get more information on the parameters. && exit 1
            shift 2
            ;;
        ---memory=*)
            [ -n "$VIASH_META_MEMORY" ] && ViashError Bad arguments for option \'---memory=*\': \'$VIASH_META_MEMORY\' \& \'$2\' - you should provide exactly one argument for this option. && exit 1
            VIASH_META_MEMORY=$(ViashRemoveFlags "$1")
            shift 1
            ;;
        *)  # positional arg or unknown option
            # since the positional args will be eval'd, can we always quote, instead of using ViashQuote
            VIASH_POSITIONAL_ARGS="$VIASH_POSITIONAL_ARGS '$1'"
            [[ $1 == -* ]] && ViashWarning $1 looks like a parameter but is not a defined parameter and will instead be treated as a positional argument. Use "--help" to get more information on the parameters.
            shift # past argument
            ;;
    esac
done

# parse positional parameters
eval set -- $VIASH_POSITIONAL_ARGS


if   [ "$VIASH_ENGINE_ID" == "native" ]  ; then
  VIASH_ENGINE_TYPE='native'
elif   [ "$VIASH_ENGINE_ID" == "docker" ]  ; then
  VIASH_ENGINE_TYPE='docker'
else
  ViashError "Engine '$VIASH_ENGINE_ID' is not recognized. Options are: docker, native."
  exit 1
fi

if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
  # check if docker is installed properly
  ViashDockerInstallationCheck

  # determine docker image id
  if [[ "$VIASH_ENGINE_ID" == 'docker' ]]; then
    VIASH_DOCKER_IMAGE_ID='images.viash-hub.com/vsh/openpipeline/qc/calculate_qc_metrics:v4.1.0'
  fi

  # print dockerfile
  if [ "$VIASH_MODE" == "dockerfile" ]; then
    ViashDockerfile "$VIASH_ENGINE_ID"
    exit 0

  elif [ "$VIASH_MODE" == "docker_image_id" ]; then
    echo "$VIASH_DOCKER_IMAGE_ID"
    exit 0
  
  # enter docker container
  elif [[ "$VIASH_MODE" == "debug" ]]; then
    VIASH_CMD="docker run --entrypoint=bash ${VIASH_DOCKER_RUN_ARGS[@]} -v '$(pwd)':/pwd --workdir /pwd -t $VIASH_DOCKER_IMAGE_ID"
    ViashNotice "+ $VIASH_CMD"
    eval $VIASH_CMD
    exit 

  # build docker image
  elif [ "$VIASH_MODE" == "setup" ]; then
    ViashDockerSetup "$VIASH_DOCKER_IMAGE_ID" "$VIASH_SETUP_STRATEGY"
    ViashDockerCheckCommands "$VIASH_DOCKER_IMAGE_ID" 'bash'
    exit 0
  fi

  # check if docker image exists
  ViashDockerSetup "$VIASH_DOCKER_IMAGE_ID" ifneedbepullelsecachedbuild
  ViashDockerCheckCommands "$VIASH_DOCKER_IMAGE_ID" 'bash'
fi

# setting computational defaults

# helper function for parsing memory strings
function ViashMemoryAsBytes {
  local memory=`echo "$1" | tr '[:upper:]' '[:lower:]' | tr -d '[:space:]'`
  local memory_regex='^([0-9]+)([kmgtp]i?b?|b)$'
  if [[ $memory =~ $memory_regex ]]; then
    local number=${memory/[^0-9]*/}
    local symbol=${memory/*[0-9]/}
    
    case $symbol in
      b)      memory_b=$number ;;
      kb|k)   memory_b=$(( $number * 1000 )) ;;
      mb|m)   memory_b=$(( $number * 1000 * 1000 )) ;;
      gb|g)   memory_b=$(( $number * 1000 * 1000 * 1000 )) ;;
      tb|t)   memory_b=$(( $number * 1000 * 1000 * 1000 * 1000 )) ;;
      pb|p)   memory_b=$(( $number * 1000 * 1000 * 1000 * 1000 * 1000 )) ;;
      kib|ki)   memory_b=$(( $number * 1024 )) ;;
      mib|mi)   memory_b=$(( $number * 1024 * 1024 )) ;;
      gib|gi)   memory_b=$(( $number * 1024 * 1024 * 1024 )) ;;
      tib|ti)   memory_b=$(( $number * 1024 * 1024 * 1024 * 1024 )) ;;
      pib|pi)   memory_b=$(( $number * 1024 * 1024 * 1024 * 1024 * 1024 )) ;;
    esac
    echo "$memory_b"
  fi
}
# compute memory in different units
if [ ! -z ${VIASH_META_MEMORY+x} ]; then
  VIASH_META_MEMORY_B=`ViashMemoryAsBytes $VIASH_META_MEMORY`
  # do not define other variables if memory_b is an empty string
  if [ ! -z "$VIASH_META_MEMORY_B" ]; then
    VIASH_META_MEMORY_KB=$(( ($VIASH_META_MEMORY_B+999) / 1000 ))
    VIASH_META_MEMORY_MB=$(( ($VIASH_META_MEMORY_KB+999) / 1000 ))
    VIASH_META_MEMORY_GB=$(( ($VIASH_META_MEMORY_MB+999) / 1000 ))
    VIASH_META_MEMORY_TB=$(( ($VIASH_META_MEMORY_GB+999) / 1000 ))
    VIASH_META_MEMORY_PB=$(( ($VIASH_META_MEMORY_TB+999) / 1000 ))
    VIASH_META_MEMORY_KIB=$(( ($VIASH_META_MEMORY_B+1023) / 1024 ))
    VIASH_META_MEMORY_MIB=$(( ($VIASH_META_MEMORY_KIB+1023) / 1024 ))
    VIASH_META_MEMORY_GIB=$(( ($VIASH_META_MEMORY_MIB+1023) / 1024 ))
    VIASH_META_MEMORY_TIB=$(( ($VIASH_META_MEMORY_GIB+1023) / 1024 ))
    VIASH_META_MEMORY_PIB=$(( ($VIASH_META_MEMORY_TIB+1023) / 1024 ))
  else
    # unset memory if string is empty
    unset $VIASH_META_MEMORY_B
  fi
fi
# unset nproc if string is empty
if [ -z "$VIASH_META_CPUS" ]; then
  unset $VIASH_META_CPUS
fi


# check whether required parameters exist
if [ -z ${VIASH_PAR_INPUT+x} ]; then
  ViashError '--input' is a required argument. Use "--help" to get more information on the parameters.
  exit 1
fi
if [ -z ${VIASH_META_NAME+x} ]; then
  ViashError 'name' is a required argument. Use "--help" to get more information on the parameters.
  exit 1
fi
if [ -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then
  ViashError 'functionality_name' is a required argument. Use "--help" to get more information on the parameters.
  exit 1
fi
if [ -z ${VIASH_META_RESOURCES_DIR+x} ]; then
  ViashError 'resources_dir' is a required argument. Use "--help" to get more information on the parameters.
  exit 1
fi
if [ -z ${VIASH_META_EXECUTABLE+x} ]; then
  ViashError 'executable' is a required argument. Use "--help" to get more information on the parameters.
  exit 1
fi
if [ -z ${VIASH_META_CONFIG+x} ]; then
  ViashError 'config' is a required argument. Use "--help" to get more information on the parameters.
  exit 1
fi
if [ -z ${VIASH_META_TEMP_DIR+x} ]; then
  ViashError 'temp_dir' is a required argument. Use "--help" to get more information on the parameters.
  exit 1
fi

# filling in defaults
if [ -z ${VIASH_PAR_MODALITY+x} ]; then
  VIASH_PAR_MODALITY="rna"
fi
if [ -z ${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS+x} ]; then
  VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS="num_nonzero_vars"
fi
if [ -z ${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS+x} ]; then
  VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS="total_counts"
fi
if [ -z ${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS+x} ]; then
  VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS="num_nonzero_obs"
fi
if [ -z ${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS+x} ]; then
  VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS="total_counts"
fi
if [ -z ${VIASH_PAR_OUTPUT_VAR_OBS_MEAN+x} ]; then
  VIASH_PAR_OUTPUT_VAR_OBS_MEAN="obs_mean"
fi
if [ -z ${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT+x} ]; then
  VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT="pct_dropout"
fi

# check whether required files exist
if [ ! -z "$VIASH_PAR_INPUT" ] && [ ! -e "$VIASH_PAR_INPUT" ]; then
  ViashError "Input file '$VIASH_PAR_INPUT' does not exist."
  exit 1
fi

# check whether parameters values are of the right type
if [[ -n "$VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE" ]]; then
  if ! [[ "$VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE" =~ ^(true|True|TRUE|false|False|FALSE|yes|Yes|YES|no|No|NO)$ ]]; then
    ViashError '--var_qc_metrics_fill_na_value' has to be a boolean. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [ -n "$VIASH_PAR_TOP_N_VARS" ]; then
  IFS=';'
  set -f
  for val in $VIASH_PAR_TOP_N_VARS; do
    if ! [[ "${val}" =~ ^[-+]?[0-9]+$ ]]; then
      ViashError '--top_n_vars' has to be an integer. Use "--help" to get more information on the parameters.
      exit 1
    fi
  done
  set +f
  unset IFS
fi

if [[ -n "$VIASH_META_CPUS" ]]; then
  if ! [[ "$VIASH_META_CPUS" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'cpus' has to be an integer. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_B" ]]; then
  if ! [[ "$VIASH_META_MEMORY_B" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_b' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_KB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_KB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_kb' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_MB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_MB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_mb' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_GB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_GB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_gb' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_TB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_TB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_tb' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_PB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_PB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_pb' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_KIB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_KIB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_kib' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_MIB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_MIB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_mib' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_GIB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_GIB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_gib' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_TIB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_TIB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_tib' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi
if [[ -n "$VIASH_META_MEMORY_PIB" ]]; then
  if ! [[ "$VIASH_META_MEMORY_PIB" =~ ^[-+]?[0-9]+$ ]]; then
    ViashError 'memory_pib' has to be a long. Use "--help" to get more information on the parameters.
    exit 1
  fi
fi

# check whether value is belongs to a set of choices
if [ ! -z "$VIASH_PAR_OUTPUT_COMPRESSION" ]; then
  VIASH_PAR_OUTPUT_COMPRESSION_CHOICES=("gzip;lzf")
  IFS=';'
  set -f
  if ! [[ ";${VIASH_PAR_OUTPUT_COMPRESSION_CHOICES[*]};" =~ ";$VIASH_PAR_OUTPUT_COMPRESSION;" ]]; then
    ViashError '--output_compression' specified value of \'$VIASH_PAR_OUTPUT_COMPRESSION\' is not in the list of allowed values. Use "--help" to get more information on the parameters.
    exit 1
  fi
  set +f
  unset IFS
fi

# create parent directories of output files, if so desired
if [ ! -z "$VIASH_PAR_OUTPUT" ] && [ ! -d "$(dirname "$VIASH_PAR_OUTPUT")" ]; then
  mkdir -p "$(dirname "$VIASH_PAR_OUTPUT")"
fi

if  [ "$VIASH_ENGINE_ID" == "native" ]  ; then
  if [ "$VIASH_MODE" == "run" ]; then
    VIASH_CMD="bash"
  else
    ViashError "Engine '$VIASH_ENGINE_ID' does not support mode '$VIASH_MODE'."
    exit 1
  fi
fi

if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
  # detect volumes from file arguments
  VIASH_CHOWN_VARS=()
if [ ! -z "$VIASH_PAR_INPUT" ]; then
  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_INPUT")" )
  VIASH_PAR_INPUT=$(ViashDockerAutodetectMount "$VIASH_PAR_INPUT")
fi
if [ ! -z "$VIASH_PAR_OUTPUT" ]; then
  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_PAR_OUTPUT")" )
  VIASH_PAR_OUTPUT=$(ViashDockerAutodetectMount "$VIASH_PAR_OUTPUT")
  VIASH_CHOWN_VARS+=( "$VIASH_PAR_OUTPUT" )
fi
if [ ! -z "$VIASH_META_RESOURCES_DIR" ]; then
  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_RESOURCES_DIR")" )
  VIASH_META_RESOURCES_DIR=$(ViashDockerAutodetectMount "$VIASH_META_RESOURCES_DIR")
fi
if [ ! -z "$VIASH_META_EXECUTABLE" ]; then
  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_EXECUTABLE")" )
  VIASH_META_EXECUTABLE=$(ViashDockerAutodetectMount "$VIASH_META_EXECUTABLE")
fi
if [ ! -z "$VIASH_META_CONFIG" ]; then
  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_CONFIG")" )
  VIASH_META_CONFIG=$(ViashDockerAutodetectMount "$VIASH_META_CONFIG")
fi
if [ ! -z "$VIASH_META_TEMP_DIR" ]; then
  VIASH_DIRECTORY_MOUNTS+=( "$(ViashDockerAutodetectMountArg "$VIASH_META_TEMP_DIR")" )
  VIASH_META_TEMP_DIR=$(ViashDockerAutodetectMount "$VIASH_META_TEMP_DIR")
fi
  
  # get unique mounts
  VIASH_UNIQUE_MOUNTS=($(for val in "${VIASH_DIRECTORY_MOUNTS[@]}"; do echo "$val"; done | sort -u))
fi

if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
  # change file ownership
  function ViashPerformChown {
    if (( ${#VIASH_CHOWN_VARS[@]} )); then
      set +e
      VIASH_CMD="docker run --entrypoint=bash --rm ${VIASH_UNIQUE_MOUNTS[@]} $VIASH_DOCKER_IMAGE_ID -c 'chown $(id -u):$(id -g) --silent --recursive ${VIASH_CHOWN_VARS[@]}'"
      ViashDebug "+ $VIASH_CMD"
      eval $VIASH_CMD
      set -e
    fi
  }
  trap ViashPerformChown EXIT
fi

if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
  # helper function for filling in extra docker args
  if [ ! -z "$VIASH_META_MEMORY_B" ]; then
    VIASH_DOCKER_RUN_ARGS+=("--memory=${VIASH_META_MEMORY_B}")
  fi
  if [ ! -z "$VIASH_META_CPUS" ]; then
    VIASH_DOCKER_RUN_ARGS+=("--cpus=${VIASH_META_CPUS}")
  fi
fi

if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
  VIASH_CMD="docker run --entrypoint=bash ${VIASH_DOCKER_RUN_ARGS[@]} ${VIASH_UNIQUE_MOUNTS[@]} $VIASH_DOCKER_IMAGE_ID"
fi


# set dependency paths


ViashDebug "Running command: $(echo $VIASH_CMD)"
cat << VIASHEOF | eval $VIASH_CMD
set -e
tempscript=\$(mktemp "$VIASH_META_TEMP_DIR/viash-run-calculate_qc_metrics-XXXXXX").py
function clean_up {
  rm "\$tempscript"
}
function interrupt {
  echo -e "\nCTRL-C Pressed..."
  exit 1
}
trap clean_up EXIT
trap interrupt INT SIGINT
cat > "\$tempscript" << 'VIASHMAIN'
import sys
import h5py
from anndata.io import read_elem, write_elem
from anndata import AnnData, settings
from scipy.sparse import csr_array
import numpy as np
from contextlib import contextmanager, closing, nullcontext
from shutil import copytree, copyfile
from functools import partial
import zarr

settings.zarr_write_format = 3

## VIASH START
# The following code has been auto-generated by Viash.
par = {
  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'var_qc_metrics': $( if [ ! -z ${VIASH_PAR_VAR_QC_METRICS+x} ]; then echo "r'${VIASH_PAR_VAR_QC_METRICS//\'/\'\"\'\"r\'}'.split(';')"; else echo None; fi ),
  'var_qc_metrics_fill_na_value': $( if [ ! -z ${VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE+x} ]; then echo "r'${VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE//\'/\'\"\'\"r\'}'.lower() == 'true'"; else echo None; fi ),
  'top_n_vars': $( if [ ! -z ${VIASH_PAR_TOP_N_VARS+x} ]; then echo "list(map(int, r'${VIASH_PAR_TOP_N_VARS//\'/\'\"\'\"r\'}'.split(';')))"; else echo None; fi ),
  'output_obs_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'output_obs_total_counts_vars': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'output_var_num_nonzero_obs': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'output_var_total_counts_obs': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'output_var_obs_mean': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_OBS_MEAN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_OBS_MEAN//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'output_var_pct_dropout': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\'/\'\"\'\"r\'}'"; else echo None; fi )
}
meta = {
  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\'/\'\"\'\"r\'}'"; else echo None; fi ),
  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\'/\'\"\'\"r\'}')"; else echo None; fi ),
  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\'/\'\"\'\"r\'}')"; else echo None; fi )
}
dep = {
  
}

## VIASH END

sys.path.append(meta["resources_dir"])
from setup_logger import setup_logger

logger = setup_logger()


@contextmanager
def mudata_opener(file_loc, mode=None):
    open_mudata = None
    input_is_zarr = False
    try:
        open_mudata = zarr.open(file_loc, zarr_format=3, mode=mode)
        input_is_zarr = True
        yield open_mudata, input_is_zarr
    except (zarr.errors.GroupNotFoundError, NotADirectoryError):
        try:
            open_mudata = h5py.File(file_loc, mode=mode)
            yield open_mudata, input_is_zarr
        except (FileNotFoundError, IsADirectoryError, KeyError) as e:
            e.add_note(f"Could not open file {file_loc}.")
            raise e
        finally:
            try:
                if open_mudata:
                    open_mudata.close()
                    del open_mudata
            except (AttributeError, UnboundLocalError):
                pass


def calculate_var_statistics(layer):
    logger.info("Calculating statistics to store in .var")
    var_columns_to_add = {}
    if par["output_var_obs_mean"]:
        logger.info(
            "(var) Calculating mean per observation, which will be stored at %s.",
            par["output_var_obs_mean"],
        )
        obs_mean = np.ravel(mean_csr_array(layer, axis=0))
        var_columns_to_add[par["output_var_obs_mean"]] = obs_mean
    if par["output_var_total_counts_obs"]:
        logger.info(
            "(var) Calculating total counts for each observation, to be stored at %s.",
            par["output_var_total_counts_obs"],
        )
        total_counts_obs = np.ravel(layer.sum(axis=0))
        var_columns_to_add[par["output_var_total_counts_obs"]] = total_counts_obs

    # This is the same as the old .nnz(axis=0), but this new implementation only works for csr_arrays!
    # See https://github.com/scipy/scipy/issues/19405#issuecomment-1773553180
    num_nonzero_obs = np.bincount(layer.indices, minlength=layer.shape[1])
    if par["output_var_num_nonzero_obs"]:
        logger.info(
            "(var) Retreiving the number of non-zero elements for each row for result column %s",
            par["output_var_num_nonzero_obs"],
        )
        var_columns_to_add[par["output_var_num_nonzero_obs"]] = num_nonzero_obs
    if par["output_var_pct_dropout"]:
        logger.info(
            "(var) Fetching for each feature the percentage of observations missing that feature (column %s)",
            par["output_var_pct_dropout"],
        )
        var_columns_to_add[par["output_var_pct_dropout"]] = (
            1 - num_nonzero_obs / layer.shape[0]
        ) * 100
    logger.info("Calculating .var statistics finished.")
    return var_columns_to_add


def calculate_obs_statistics(layer, var):
    logger.info("Calculating statistics to store in .obs")
    obs_columns_to_add = {}
    total_counts_var = np.ravel(layer.sum(axis=1))

    if par["output_obs_num_nonzero_vars"]:
        logger.info(
            "(obs) Retreiving the number of non-zero elements for each feature to be stored in column %s",
            par["output_var_num_nonzero_obs"],
        )
        # This is the same as the old .nnz(axis=1), but this new implementation only works for csr_arrays!
        # See https://github.com/scipy/scipy/issues/19405#issuecomment-1773553180
        num_nonzero_vars = np.diff(layer.indptr)
        obs_columns_to_add[par["output_obs_num_nonzero_vars"]] = num_nonzero_vars

    if par["output_obs_total_counts_vars"]:
        logger.info(
            "(obs) Calculating total counts for each feature, to be stored at %s.",
            par["output_obs_total_counts_vars"],
        )
        obs_columns_to_add[par["output_obs_total_counts_vars"]] = total_counts_var

    top_metrics = {}
    if par["top_n_vars"]:
        logger.info(
            "(obs) Calculating the cumulative proportions to the %s most expressed vars.",
            ", ".join([f"{top_n_var}th" for top_n_var in par["top_n_vars"]]),
        )
        par["top_n_vars"] = sorted(par["top_n_vars"])
        distributions = get_top_from_csr_matrix(layer, par["top_n_vars"])
        top_metrics = {
            distribution_size: distribution * 100
            for distribution_size, distribution in zip(
                par["top_n_vars"], distributions.T
            )
        }
        obs_columns_to_add |= {
            f"pct_of_counts_in_top_{n_top}_vars": col
            for n_top, col in top_metrics.items()
        }
    for qc_metric in par.get("var_qc_metrics", []) or []:
        logger.info(
            "(obs) Retreiving the proportion of total 'True' values in column %s",
            qc_metric,
        )
        if qc_metric not in var:
            raise ValueError(
                f"Value for --var_qc_metrics, '{qc_metric}' "
                f"not found in .var for modality {par['modality']}"
            )
        qc_column = var[qc_metric]
        if qc_column.isna().any():
            if par["var_qc_metrics_fill_na_value"] is None:
                raise ValueError(
                    f"The .var column '{qc_metric}', selected by '--var_qc_metrics', contains NA values. "
                    "It is ambiguous whether or not to include these values in the static calulation. "
                    "You can explicitly map the NA values to 'False' or 'True using '--var_qc_metrics_fill_na_value'"
                )
            else:
                qc_column = qc_column.fillna(
                    par["var_qc_metrics_fill_na_value"], inplace=False
                )
        qc_column = qc_column.to_list()
        if set(np.unique(qc_column)) - {True, False}:
            raise ValueError(
                f"Column {qc_metric} in .var for modality {par['modality']} "
                f"must only contain boolean values"
            )
        total_counts_qc_metric = np.ravel(layer[:, qc_column].sum(axis=1))
        obs_columns_to_add |= {
            f"total_counts_{qc_metric}": total_counts_qc_metric,
            f"pct_{qc_metric}": total_counts_qc_metric / total_counts_var * 100,
        }
    logger.info("Finised calculating obs statistics")
    return obs_columns_to_add


def cast_layer_dtype(layer):
    # from the np.sum documentation:
    # Especially when summing a large number of lower precision floating point numbers,
    # such as float32, numerical errors can become significant. In such cases it can
    # be advisable to use dtype="float64" to use a higher precision for the output.

    # However, the 'dtype' from SciPy's implementation of sum cannot be used for this
    # as it does not use an internal accumulator but matrix multiplication for calculating
    # the sum. So here we cast explicitly to a higher precision before doing the calcualtions.
    # The downside is that this is inefficient.
    # See https://github.com/scipy/scipy/issues/23768#issuecomment-3909317463
    original_dtype = layer.dtype
    target_dtype = original_dtype
    if np.issubdtype(original_dtype, np.floating) and np.can_cast(
        original_dtype, np.float64, casting="safe"
    ):
        # use promote_types in orde to make suresure not to cast np.float128
        # or anything else to a lower precision dtype
        target_dtype = np.promote_types(np.float64, original_dtype)
        logger.info(
            "Using target dtype %s for layer. Casting may be required for higher precision.",
            target_dtype,
        )
    result = csr_array(layer, dtype=target_dtype, copy=False)
    logger.info("Constructed CSR of shape %s and dtype %s", result.shape, result.dtype)
    return result


def mean_csr_array(input_csr_array, axis):
    # TODO: replace this function with the native SciPy version
    # when anndata supports SciPy >= 1.17.0
    # We use this version to avoid creating duplicates of the data in memory
    # when the \`astype\` function is called.
    # See https://github.com/scipy/scipy/pull/23797
    sparse_array_dtype = input_csr_array.dtype
    integral = np.issubdtype(sparse_array_dtype, np.integer) or np.issubdtype(
        sparse_array_dtype, np.bool_
    )

    # intermediate dtype for summation
    inter_dtype = np.float64 if integral else sparse_array_dtype
    inter_cast = input_csr_array.astype(inter_dtype, copy=False)
    divided = inter_cast.data * (1.0 / input_csr_array.shape[axis])
    divided_csr = csr_array(
        (divided, input_csr_array.indices, input_csr_array.indptr),
        shape=input_csr_array.shape,
    )
    return divided_csr.sum(axis=axis, dtype=inter_dtype)


def get_top_from_csr_matrix(array, top_n_genes):
    # csr matrices stores a 3D matrix in a format such that data for individual cells
    # are stored in 1 array. Another array (indptr) here stores the ranges of indices
    # to select from the data-array (.e.g. data[indptr[0]:indptr[1]] for row 0) for each row.
    # Another array 'indices' maps each element of data to a column
    # (data and indices arrays have the same length)
    top_n_genes = np.array(top_n_genes).astype(np.int64)
    assert np.all(top_n_genes[:-1] <= top_n_genes[1:]), "top_n_genes must be sorted"
    row_indices, data = array.indptr, array.data
    number_of_rows, max_genes_to_parse = row_indices.size - 1, top_n_genes[-1]
    top_data = np.zeros((number_of_rows, max_genes_to_parse), dtype=data.dtype)
    # Loop over each row to create a dense matrix without the 0 counts,
    # but not for the whole matrix, only store the genes up until
    # the largest number of top n genes.
    for row_number in range(number_of_rows):
        row_start_index, row_end_index = (
            row_indices[row_number],
            row_indices[row_number + 1],
        )
        row_data = data[row_start_index:row_end_index]  # all non-zero counts for an row
        try:
            # There are less genes with counts in the row than the
            # maximum number of genes we would like to select
            # all these genes are in the top genes, just store them
            top_data[row_number, : row_end_index - row_start_index] = row_data
        except ValueError:
            # Store the counts for the top genes
            top_data[row_number, :] = np.partition(row_data, -max_genes_to_parse)[
                -max_genes_to_parse:
            ]

    # Partition works from smallest to largest, but we want largest
    # so do smallest to largest first (but with reversed indices)
    top_data = np.partition(top_data, max_genes_to_parse - top_n_genes)
    # And then switch the order around
    top_data = np.flip(top_data, axis=1)

    cumulative = top_data.cumsum(axis=1, dtype=np.float64)[:, top_n_genes - 1]
    return cumulative / np.expand_dims(array.sum(axis=1), 1)


def main():
    logger.info("Started %s.", meta["name"])
    mod_element_loc = f"mod/{par['modality']}"
    layer_element_name = (
        f"{mod_element_loc}/X"
        if not par["layer"]
        else f"{mod_element_loc}/layers/{par['layer']}"
    )
    # In order to match the format (zarr or h5) of the input to the output
    with mudata_opener(par["input"], mode="r") as (open_mudata, input_is_zarr):
        logger.info(
            "Openened %s in %s format.", par["input"], "zarr" if input_is_zarr else "h5"
        )
        mods = list(open_mudata["mod"].keys())
        logger.info("Found modalities: %s", ", ".join(mods))
        logger.info("Reading metadata frames for modality %s", par["modality"])
        var = read_elem(open_mudata[f"{mod_element_loc}/var"])
        logger.info(".var shape for %s is %s", par["modality"], var.shape)
        obs = read_elem(open_mudata[f"{mod_element_loc}/obs"])
        logger.info(".obs shape for %s is %s", par["modality"], obs.shape)
        minimal_anndata = AnnData(var=var, obs=obs)
        logger.info("Reading layer %s", "X" if not par["layer"] else par["layer"])
        layer = read_elem(open_mudata[layer_element_name])
        logger.info("Found layer with shape %s and dtype %s", layer.shape, layer.dtype)

    layer = cast_layer_dtype(layer)
    logger.info("Eliminating explicit zeros from sparse layer.")
    layer.eliminate_zeros()

    var_columns_to_add = calculate_var_statistics(layer)
    minimal_anndata.var = minimal_anndata.var.assign(**var_columns_to_add)

    # obs statistics
    obs_columns_to_add = calculate_obs_statistics(layer, minimal_anndata.var)
    minimal_anndata.obs = minimal_anndata.obs.assign(**obs_columns_to_add)

    logger.info("Writing to %s", par["output"])
    try:
        copytree(par["input"], par["output"], symlinks=True)
    except NotADirectoryError:
        copyfile(par["input"], par["output"], follow_symlinks=True)
        logger.info("Copied input file %s to %s", par["input"], par["output"])
    else:
        logger.info("Copied input directory %s", par["input"], par["output"])
    logger.info("Using a %s writer", "zarr" if input_is_zarr else "H5")
    write_opener = (
        partial(zarr.open, zarr_format=3, use_consolidated=False)
        if input_is_zarr
        else h5py.File
    )
    context = (
        nullcontext if input_is_zarr else closing
    )  # zarr format does not need to be closed
    logger.info("Overwriting slots.")
    with context(write_opener(par["output"], mode="a")) as open_output:
        write_elem(open_output[mod_element_loc], "obs", minimal_anndata.obs)
        write_elem(open_output[mod_element_loc], "var", minimal_anndata.var)
    if input_is_zarr:
        zarr.consolidate_metadata(open_output.store)
    logger.info("Finished!")


if __name__ == "__main__":
    main()
VIASHMAIN
python -B "\$tempscript" &
wait "\$!"

VIASHEOF


if [[ "$VIASH_ENGINE_TYPE" == "docker" ]]; then
  # strip viash automount from file paths
  
  if [ ! -z "$VIASH_PAR_INPUT" ]; then
    VIASH_PAR_INPUT=$(ViashDockerStripAutomount "$VIASH_PAR_INPUT")
  fi
  if [ ! -z "$VIASH_PAR_OUTPUT" ]; then
    VIASH_PAR_OUTPUT=$(ViashDockerStripAutomount "$VIASH_PAR_OUTPUT")
  fi
  if [ ! -z "$VIASH_META_RESOURCES_DIR" ]; then
    VIASH_META_RESOURCES_DIR=$(ViashDockerStripAutomount "$VIASH_META_RESOURCES_DIR")
  fi
  if [ ! -z "$VIASH_META_EXECUTABLE" ]; then
    VIASH_META_EXECUTABLE=$(ViashDockerStripAutomount "$VIASH_META_EXECUTABLE")
  fi
  if [ ! -z "$VIASH_META_CONFIG" ]; then
    VIASH_META_CONFIG=$(ViashDockerStripAutomount "$VIASH_META_CONFIG")
  fi
  if [ ! -z "$VIASH_META_TEMP_DIR" ]; then
    VIASH_META_TEMP_DIR=$(ViashDockerStripAutomount "$VIASH_META_TEMP_DIR")
  fi
fi


# check whether required files exist
if [ ! -z "$VIASH_PAR_OUTPUT" ] && [ ! -e "$VIASH_PAR_OUTPUT" ]; then
  ViashError "Output file '$VIASH_PAR_OUTPUT' does not exist."
  exit 1
fi


exit 0
