wcx_script/wunf_upd_prefix.sh

#!/usr/bin/env bash
set -euo pipefail

CSV_DEFAULT="wunf_scenes.csv"
DRY_RUN=0

usage() {
  echo "Användning:"
  echo "  $0 [--dry-run] <filnamn> [csv-fil]"
  echo "  $0 [--dry-run] --all <katalog> [csv-fil]"
  echo "  $0 [--dry-run] --list <textfil> [csv-fil]   # en väg per rad"
  exit 1
}

rename_file() {
  local FILE="$1"
  local CSV="$2"

  if [[ ! -f "$FILE" ]]; then
    echo "Fel: Filen finns inte: $FILE" >&2
    return 1
  fi

  local base dir rest name_segment code newbase newpath

  base="$(basename -- "$FILE")"
  dir="$(dirname -- "$FILE")"

  # Hoppa över om den redan har index, t.ex. [WUNF_427]...
  if [[ "${base:0:7}" == "[WUNF_" ]]; then
    echo "Hoppar över (har redan index): $base" >&2
    return 0
  fi

  # Kräver exakt prefixet "[WUNF]"
  if [[ "${base:0:6}" != "[WUNF]" ]]; then
    echo "Hoppar över (ej [WUNF]-prefix): $base" >&2
    return 0
  fi

  # Ta bort prefixet "[WUNF]"
  rest="${base:6}"

  # Ta bort ett eller flera ledande underscores efter prefixet
  while [[ "${rest:0:1}" == "_" ]]; do
    rest="${rest:1}"
  done

  # Ta bort extension (.mp4, .mkv osv)
  name_segment="${rest%.*}"

  if [[ -z "$name_segment" ]]; then
    echo "Ingen namn-del efter prefix i: $base" >&2
    return 0
  fi

  # Slå upp kod i CSV via tokeniserad matchning
  code="$(
    awk -F';' -v fstring="$name_segment" '
      BEGIN {
        # 1) Tokenisera filnamnets namn-del
        s = fstring
        gsub(/_+/, " ", s)                     # "_" -> mellanslag
        gsub(/[[:space:]]+/, " ", s)           # normalisera whitespace
        sub(/^[[:space:]]+/, "", s)
        sub(/[[:space:]]+$/, "", s)
        s = tolower(s)

        if (length(s) == 0) { exit 0 }

        n_ft = split(s, ft, " ")               # ft[] = filtokens

        bestScore = 0
        bestCode  = ""
        ambiguous = 0
      }
      {
        # 2) Tokenisera namn-kolumnen i CSV
        name = $1
        gsub(/,/, " ", name)                   # kommatecken ignoreras
        gsub(/[[:space:]]+/, " ", name)
        sub(/^[[:space:]]+/, "", name)
        sub(/[[:space:]]+$/, "", name)
        lname = tolower(name)
        if (length(lname) == 0) next

        n_rt = split(lname, rt, " ")           # rt[] = CSV-tokens

        # kan inte vara prefix om den är längre än filens tokenlista
        if (n_rt > n_ft) next

        # 3) Kolla om CSV-tokenlistan är prefix till filnamnets tokens
        matchOK = 1
        for (i = 1; i <= n_rt; i++) {
          if (rt[i] != ft[i]) { matchOK = 0; break }
        }
        if (!matchOK) next

        matchLen = n_rt

        if (matchLen > 0) {
          if (matchLen > bestScore) {
            bestScore = matchLen
            bestCode  = $2
            ambiguous = 0
          } else if (matchLen == bestScore) {
            # Fler rader med samma bästa längd => tvetydigt
            ambiguous = 1
          }
        }
      }
      END {
        if (bestScore > 0 && !ambiguous && length(bestCode) > 0) {
          c = bestCode
          gsub(/^[ \t]+|[ \t]+$/, "", c)
          print c
        }
      }
    ' "$CSV"
  )"

  if [[ -z "$code" ]]; then
    echo "Ingen unik match för filen: $base" >&2
    return 0
  fi

  # Bygg prefix med paddad index:
  # förväntat format: t.ex. "WUNF 1", "WUNF 88", "WUNF 427"
  local trimmed_code prefix number number_padded code_us
  trimmed_code="$(printf '%s' "$code" | sed -E 's/^[[:space:]]+|[[:space:]]+$//g')"

  if [[ "$trimmed_code" =~ ^([A-Za-z]+)[[:space:]]*([0-9]+)$ ]]; then
    prefix="${BASH_REMATCH[1]}"
    number="${BASH_REMATCH[2]}"
    printf -v number_padded "%03d" "$number"
    code_us="${prefix}_${number_padded}"
  else
    # Fallback om formatet skulle avvika – använd gammal logik
    code_us="${trimmed_code// /_}"
  fi

  # Ersätt bara själva prefixet [WUNF] i början
  newbase="$(printf '%s' "$base" | sed -E "s/^\[WUNF\]/[$code_us]/")"
  newpath="$dir/$newbase"

  if [[ "$base" == "$newbase" ]]; then
    echo "Hoppar över (redan rätt prefix): $base" >&2
    return 0
  fi

  if [[ -e "$newpath" && $DRY_RUN -eq 0 ]]; then
    echo "Fel: Målfilen finns redan: $newpath" >&2
    return 1
  fi

  if [[ $DRY_RUN -eq 1 ]]; then
    echo "DRY-RUN: skulle byta namn:"
    echo "  Från: $base"
    echo "  Till:  $newbase"
  else
    mv -- "$dir/$base" "$newpath"
    echo "Bytt namn:"
    echo "  Från: $base"
    echo "  Till:  $newbase"
  fi
}

### Huvudprogram

if [[ $# -lt 1 ]]; then
  usage
fi

# Global flagga --dry-run
if [[ "${1:-}" == "--dry-run" ]]; then
  DRY_RUN=1
  shift
fi

if [[ $# -lt 1 ]]; then
  usage
fi

mode="$1"

case "$mode" in
  --all)
    if [[ $# -lt 2 ]]; then usage; fi
    DIR="$2"
    CSV="${3:-$CSV_DEFAULT}"

    if [[ ! -d "$DIR" ]]; then
      echo "Fel: Katalog finns inte: $DIR" >&2
      exit 1
    fi
    if [[ ! -f "$CSV" ]]; then
      echo "Fel: CSV-filen finns inte: $CSV" >&2
      exit 1
    fi

    shopt -s nullglob
    for f in "$DIR"/\[WUNF\]*.mp4; do
      rename_file "$f" "$CSV" || true
    done
    ;;

  --list)
    if [[ $# -lt 2 ]]; then usage; fi
    LIST="$2"
    CSV="${3:-$CSV_DEFAULT}"

    if [[ ! -f "$LIST" ]]; then
      echo "Fel: Textfil finns inte: $LIST" >&2
      exit 1
    fi
    if [[ ! -f "$CSV" ]]; then
      echo "Fel: CSV-filen finns inte: $CSV" >&2
      exit 1
    fi

    while IFS= read -r line || [[ -n "$line" ]]; do
      line="${line%$'\r'}"
      trimmed="$(printf '%s' "$line" | sed -E 's/^[[:space:]]+|[[:space:]]+$//g')"
      [[ -z "$trimmed" || "$trimmed" =~ ^# ]] && continue
      rename_file "$trimmed" "$CSV" || true
    done < "$LIST"
    ;;

  *)
    FILE="$mode"
    CSV="${2:-$CSV_DEFAULT}"

    if [[ ! -f "$CSV" ]]; then
      echo "Fel: CSV-filen finns inte: $CSV" >&2
      exit 1
    fi

    rename_file "$FILE" "$CSV"
    ;;
esac