Testing

gistfile1.txt

#!/usr/bin/env bash
#
# csv_to_pki_json.sh
#
# Pure POSIX shell + awk implementation (NO python, NO jq, NO extra
# installs) - works in plain Git Bash / MINGW / any minimal shell that
# ships awk, sed, grep (i.e. basically everywhere).
#
# IMPORTANT: True .xlsx files are zip archives containing XML - a shell
# without unzip/python literally cannot decode that binary format. So
# this version takes a CSV export of your Excel sheet instead:
#   Excel -> File -> Save As -> CSV (Comma delimited) (*.csv)
# Any spreadsheet (Excel, LibreOffice, Numbers, Google Sheets) can export
# CSV without installing anything extra.
#
# Usage:
#   ./csv_to_pki_json.sh -i devices.csv -c "Common Name" \
#       -s YOUR_STRUCT_KEY_ID -e your.group.mailbox@volkswagen.de \
#       [-t vwpki/802x-client/p12] [-v 1y] [-d false] \
#       [-o out_dir] [-n prefix] [--chunk-size 250] [--delim ,]
#
# -c accepts either the header name (case-insensitive, matched against
# row 1) or a 1-based column index (e.g. -c 3).
#
# Output:
#   out_dir/prefix_001.json
#   out_dir/prefix_002.json
#   ...

set -eu

CERT_TYPE="vwpki/802x-client/p12"
VALIDITY="1y"
DRY_RUN="false"
CHUNK_SIZE=250
OUT_DIR="./output"
PREFIX="pki_request"
DELIM=","
INPUT_FILE=""
COLUMN=""
STRUCT_KEY_ID=""
CONTACT_EMAIL=""

usage() {
  sed -n '2,/^$/p' "$0" | sed 's/^# \{0,1\}//'
  exit 1
}

while [ $# -gt 0 ]; do
  case "$1" in
    -i|--input)          INPUT_FILE="$2"; shift 2 ;;
    -c|--column)         COLUMN="$2"; shift 2 ;;
    -s|--struct-key-id)  STRUCT_KEY_ID="$2"; shift 2 ;;
    -e|--email)          CONTACT_EMAIL="$2"; shift 2 ;;
    -t|--cert-type)      CERT_TYPE="$2"; shift 2 ;;
    -v|--validity)       VALIDITY="$2"; shift 2 ;;
    -d|--dry-run)        DRY_RUN="$2"; shift 2 ;;
    -o|--out-dir)        OUT_DIR="$2"; shift 2 ;;
    -n|--prefix)         PREFIX="$2"; shift 2 ;;
    --chunk-size)        CHUNK_SIZE="$2"; shift 2 ;;
    --delim)             DELIM="$2"; shift 2 ;;
    -h|--help)           usage ;;
    *) echo "Unknown argument: $1" >&2; usage ;;
  esac
done

[ -z "$INPUT_FILE" ] && { echo "Error: -i/--input (CSV file) is required" >&2; exit 1; }
[ -z "$COLUMN" ] && { echo "Error: -c/--column (header name or index) is required" >&2; exit 1; }
[ -z "$STRUCT_KEY_ID" ] && { echo "Error: -s/--struct-key-id is required" >&2; exit 1; }
[ -z "$CONTACT_EMAIL" ] && { echo "Error: -e/--email is required" >&2; exit 1; }
[ ! -f "$INPUT_FILE" ] && { echo "Error: input file '$INPUT_FILE' not found" >&2; exit 1; }

case "$CHUNK_SIZE" in
  ''|*[!0-9]*) echo "Error: --chunk-size must be a positive integer" >&2; exit 1 ;;
esac
if [ "$CHUNK_SIZE" -lt 1 ] || [ "$CHUNK_SIZE" -gt 250 ]; then
  echo "Error: --chunk-size must be between 1 and 250" >&2
  exit 1
fi

case "$DRY_RUN" in
  true|false) : ;;
  *) echo "Error: -d/--dry-run must be true or false" >&2; exit 1 ;;
esac

mkdir -p "$OUT_DIR"

awk -v col="$COLUMN" \
    -v delim="$DELIM" \
    -v struct_key_id="$STRUCT_KEY_ID" \
    -v contact_email="$CONTACT_EMAIL" \
    -v cert_type="$CERT_TYPE" \
    -v validity="$VALIDITY" \
    -v dry_run="$DRY_RUN" \
    -v chunk_size="$CHUNK_SIZE" \
    -v out_dir="$OUT_DIR" \
    -v prefix="$PREFIX" \
'
function jesc(s) {
  gsub(/\\/, "\\\\", s)
  gsub(/"/, "\\\"", s)
  gsub(/\t/, "\\t", s)
  gsub(/\r/, "", s)
  return s
}

function write_header(fh) {
  print "{" > fh
  print "  \"business_context\": {" > fh
  print "    \"struct_key_id\": \"" jesc(struct_key_id) "\"," > fh
  print "    \"contact_email\": \"" jesc(contact_email) "\"" > fh
  print "  }," > fh
  print "  \"certificate_type\": \"" jesc(cert_type) "\"," > fh
  print "  \"validity\": \"" jesc(validity) "\"," > fh
  print "  \"dry_run\": " dry_run "," > fh
  print "  \"subjects\": [" > fh
}

function write_footer(fh) {
  print "  ]" > fh
  print "}" > fh
  close(fh)
}

BEGIN {
  FS = delim
  col_idx = 0
  if (col ~ /^[0-9]+$/) col_idx = col + 0
  count = 0
  file_index = 0
  in_batch = 0
}

NR == 1 {
  if (col_idx == 0) {
    for (i = 1; i <= NF; i++) {
      h = $i
      gsub(/^[ \t\r"]+|[ \t\r"]+$/, "", h)
      if (tolower(h) == tolower(col)) { col_idx = i; break }
    }
    if (col_idx == 0) {
      print "Error: column \x27" col "\x27 not found in header" > "/dev/stderr"
      exit 1
    }
  }
  next
}

{
  if (col_idx > NF) next
  val = $col_idx
  gsub(/^[ \t\r"]+|[ \t\r"]+$/, "", val)
  if (val == "") next

  if (count % chunk_size == 0) {
    if (in_batch) {
      print "    }" > cur_file
      write_footer(cur_file)
      print "Wrote " cur_file " (" batch_n " subjects)" > "/dev/stderr"
    }
    file_index++
    cur_file = sprintf("%s/%s_%03d.json", out_dir, prefix, file_index)
    write_header(cur_file)
    in_batch = 1
    batch_n = 0
  }

  if (batch_n > 0) print "    }," > cur_file
  print "    {" > cur_file
  print "      \"common_name\": \"" jesc(val) "\"" > cur_file
  batch_n++
  count++
}

END {
  if (in_batch) {
    print "    }" > cur_file
    write_footer(cur_file)
    print "Wrote " cur_file " (" batch_n " subjects)" > "/dev/stderr"
  }
  if (count == 0) {
    print "Error: no values found in column \x27" col "\x27" > "/dev/stderr"
    exit 1
  }
  print "Done. Generated " file_index " file(s) in \x27" out_dir "\x27." > "/dev/stderr"
}
' "$INPUT_FILE"