renam and filter

rename_and_filter.sh

@@ -2,49 +2,49 @@
 # =============================================================================
 #  rename_and_filter.sh
 #
-#  Version: 1.3.1
+#  Version: 1.9.0
-#  Last updated: 2025-10-04
+#  Last updated: 2025-11-13 (camelCase- & siffersplit borttagen)
 #
 #  Summary
 #  -------
-#  Given a directory, a comma-separated file of reserved words, and an optional
+#  Döper om filer i en katalog baserat på en kommaseparerad lista med spärrade
-#  file extension filter:
+#  ord (reserved words). Matchning sker mot _tokens_ i basnamnet (skiljetecken
-#    1) Removes any filename tokens (separated by "_") that match reserved words
+#  "_"), med **exakt token-matchning** (case-insensitivt). Inga substring-
-#       (case-insensitive, exact token matches).
+#  regler och ingen automatisk camelCase/siffersplit.
-#       - EXTRA: If a reserved word appears as the last token (e.g., "..._good"),
-#         it is removed even if it's only delimited on the left side.
-#    2) Replaces one or more spaces with a single underscore "_".
-#    3) Collapses multiple "_" into a single "_", and trims leading/trailing "_".
 #
-#  Notes:
+#  Regler (AUKTORITATIV ORDNING)
-#  - No files are deleted; only renaming occurs.
+#  -----------------------------
-#  - Processes files only in the given directory (non-recursive).
+#  OBS: Numreringen återspeglar **exakt tillämpningsordning** i koden. Om en
-#  - Collision handling: if target name exists, suffix "_N" before the extension.
+#  ny regel införs mellan t.ex. 3 och 4 ska **alla efterföljande regler
-#  - Reserved words match only against the basename tokens (extension excluded).
+#  renumreras**. Referera alltid till regler som "Regel N".
 #
-#  Extension filter:
+#  Regler:
-#  - Default: mp4
+#    1) [strip_reserved] Ta bort tokens som exakt matchar ett spärrat ord
-#  - "*"    : all files
+#       (case-insensitivt).
-#  - "jpg"  : only .jpg (with or without leading dot accepted, e.g. "jpg" or ".jpg")
+#    2) [strip_reserved] Ta bort tokens som består av enbart siffror, oavsett
+#       längd (t.ex. 2, 22, 123).
+#    3) [strip_reserved] Normalisera basnamnet: ersätt "-" → "_", whitespace → "_",
+#       kollapsa multipla "_" och dela på "_".
+#    4) [normalize_name] Finjustera: kollapsa ev. kvarvarande multipla "_", trimma
+#       ledande/efterföljande "_" i basnamnet. Extension bibehålls.
+#    5) [candidate_name] Krockhantering: om målnamn finns, suffix "_N" före
+#       extension.
 #
-#  Logging:
+#  Noter:
-#    RENAME: "old_name.ext" => "new_name.ext"
+#  - Ingen fil raderas; endast rename.
-#    NOCHANGE (dry-run): "file.ext" (nothing to do)
+#  - Icke-rekursiv som standard; använd --recursive för att gå ned i underkataloger.
+#  - Extensionfilter: en, flera (kommaseparerade) eller "*" för alla filer.
 #
-#  Usage
+#  Exempel
-#  -----
+#  -------
-#    ./rename_and_filter.sh [--dry-run] <directory> <reserved_words.csv> [extension]
+#    # Torrkörning (default mp4)
+#    ./rename_and_filter.sh --dry-run ./videos reserved_words.csv
 #
-#  Examples
+#    # Alla filer rekursivt
-#  --------
+#    ./rename_and_filter.sh -n -r ./media reserved_words.csv "*"
-#    # Dry-run on mp4 files (default):
-#    ./rename_and_filter.sh --dry-run ./videos ./reserved_words.csv
 #
-#    # All files:
+#    # jpg & png
-#    ./rename_and_filter.sh ./media ./reserved_words.csv "*"
+#    ./rename_and_filter.sh -n ./pics reserved_words.csv jpg,png
-#
-#    # Only jpg:
-#    ./rename_and_filter.sh ./pictures ./reserved_words.csv jpg
 #
 #  Exit codes
 #  ----------
@@ -53,87 +53,87 @@
 #    2  Requires bash >= 4
 # =============================================================================
 
-set -euo pipefail
+# Viktigt: vi kör INTE set -e, för att undvika aborter mitt i loopen.
+# Vi behåller -u och pipefail för rimlig säkerhet.
+set -uo pipefail
+SCRIPT_VERSION="1.9.0"
 
-# Requires bash 4 for associative arrays
+# Bash 4 krävs för assoc. arrayer & case-conversion
 if [ "${BASH_VERSINFO:-0}" -lt 4 ]; then
   echo "This script requires bash >= 4." >&2
   exit 2
 fi
 
 DRY_RUN=0
+RECURSIVE=0
+MIN_LEN=0
+VERBOSE=0
 
 print_help() {
-  sed -n '2,160p' "$0" | sed 's/^# \{0,1\}//'
+  sed -n '2,200p' "$0" | sed 's/^# \{0,1\}//'
+}
+
+print_rules() {
+  cat <<'RULES'
+Regler (auktoritativ ordning)
+1) [strip_reserved] Ta bort tokens som exakt matchar ett spärrat ord (case-insensitivt).
+2) [strip_reserved] Ta bort tokens som består av enbart siffror, oavsett längd (t.ex. 2, 22, 123).
+3) [strip_reserved] Normalisera basnamnet: ersätt "-" → "_", whitespace → "_", kollapsa multipla "_" och dela på "_".
+4) [normalize_name] Finjustera: kollapsa ev. kvarvarande multipla "_", trimma ledande/efterföljande "_" i basnamnet. Extension bibehålls.
+5) [candidate_name] Krockhantering: om målnamn finns, suffix "_N" före extension.
+RULES
 }
 
 log() {
   printf '[%s] %s\n' "$(date '+%Y-%m-%d %H:%M:%S')" "$*"
 }
 
-rename_file() {
-  local src="$1"
-  local dst="$2"
-
-  # Guard: identical path, nothing to do
-  if [[ "$src" == "$dst" ]]; then
-    if (( DRY_RUN )); then
-      # CHANGE: add `--` so basename doesn't treat dash-prefixed names as options
-      log "NOCHANGE (dry-run): \"$(basename -- "$src")\" (nothing to do)"
-    fi
-    return 0
-  fi
-
-  local base ext candidate n
-  base="${dst##*/}"
-  ext=""
-  if [[ "$base" == *.* ]]; then
-    ext=".${base##*.}"
-    base="${base%.*}"
-  fi
-  candidate="$dst"
-  n=1
-  # Avoid collisions (unless it's the same file, already handled above)
-  while [[ -e "$candidate" && "$src" != "$candidate" ]]; do
-    candidate="$(dirname "$dst")/${base}_$n$ext"
-    ((n++))
-  done
-
-  if (( DRY_RUN )); then
-    # CHANGE: add `--` for safe logging
-    log "RENAME: \"$(basename -- "$src")\" => \"$(basename -- "$candidate")\""
-  else
-    mv -v -- "$src" "$candidate" >/dev/null
-    # CHANGE: add `--` for safe logging
-    log "RENAME: \"$(basename -- "$src")\" => \"$(basename -- "$candidate")\""
-  fi
-}
-
 # --- Argument parsing ---
-if [[ "${1:-}" == "-h" || "${1:-}" == "--help" ]]; then
+ARGS=()
-  print_help
+while (("$#")); do
-  exit 0
+  case "${1}" in
-fi
+    -h|--help)      print_help; exit 0;;
+    --print-rules)  print_rules; exit 0;;
+    -n|--dry-run)   DRY_RUN=1; shift;;
+    -r|--recursive) RECURSIVE=1; shift;;
+    -v|--verbose)   VERBOSE=1; shift;;
+    --min-len)      MIN_LEN=${2:-0}; shift 2;;
+    --)             shift; break;;
+    -* )            echo "Unknown option: $1" >&2; exit 1;;
+    *  )            ARGS+=("$1"); shift;;
+  esac
+done
 
-if [[ "${1:-}" == "-n" || "${1:-}" == "--dry-run" ]]; then
+# Positional args
-  DRY_RUN=1
+if (( ${#ARGS[@]} < 2 || ${#ARGS[@]} > 3 )); then
-  shift
+  echo "Error: Provide <directory> <reserved_words.csv> [extension|list|*]. Use --help for info." >&2
-fi
-
-if [[ $# -lt 2 || $# -gt 3 ]]; then
-  echo "Error: Provide <directory> <reserved_words.csv> [extension]. Use --help for info." >&2
   exit 1
 fi
 
-DIR="$1"
+DIR="${ARGS[0]}"
-WORD_FILE="$2"
+WORD_FILE="${ARGS[1]}"
-EXT="${3:-mp4}"
+EXT_ARG="${ARGS[2]:-mp4}"
 
-# Normalize EXT: strip leading dot if present
+# Normalize extensions list into a csv string (lowercase, strip leading dots)
-if [[ "$EXT" == .* ]]; then
+normalize_ext_list() {
-  EXT="${EXT#.}"
+  local arg="$1"
-fi
+  local -a out=()
+  if [[ "$arg" == "*" ]]; then
+    echo "*"
+    return 0
+  fi
+  IFS=',' read -r -a tmp <<< "$arg"
+  for e in "${tmp[@]}"; do
+    [[ -z "$e" ]] && continue
+    e="${e#.}"
+    out+=("${e,,}")
+  done
+  (IFS=','; echo "${out[*]}")
+}
 
+EXT_LIST_STR="$(normalize_ext_list "$EXT_ARG")"
+
+# Validate inputs
 if [[ ! -d "$DIR" ]]; then
   echo "Error: Directory does not exist: $DIR" >&2
   exit 1
@@ -143,67 +143,102 @@ if [[ ! -f "$WORD_FILE" ]]; then
   exit 1
 fi
 
-# --- Load reserved words into an associative set ---
+# --- Load reserved words (to associative set) ---
 declare -A RESERVED=()
 # Allow comma and newlines as separators
-# shellcheck disable=SC2002
+mapfile -t _tokens < <(tr ',\r\n' '\n' < "$WORD_FILE")
-mapfile -t _tokens < <(cat "$WORD_FILE" | tr ',\r\n' '\n')
 for raw in "${_tokens[@]}"; do
   # trim whitespace
-  w="${raw#"${raw%%[![:space:]]*}"}"   # ltrim
+  w="${raw#"${raw%%[![:space:]]*}"}"
-  w="${w%"${w##*[![:space:]]}"}"       # rtrim
+  w="${w%"${w##*[![:space:]]}"}"
   [[ -z "$w" ]] && continue
-  RESERVED["${w,,}"]=1
+  lw="${w,,}"        # case-insensitive
+  lw="${lw//-/_}"    # normalisera även ordlista: '-' → '_'
+  RESERVED["$lw"]=1
 done
 
-log "Processing directory: $DIR (dry-run=$DRY_RUN, extension=${EXT})"
+# --- Helpers ---
-log "Reserved words loaded: ${#RESERVED[@]}"
+matches_extension() {
+  local file="$1" ext_list="$2" base ext lc
+  [[ "$ext_list" == "*" ]] && return 0
+  base="$(basename "$file")"
+  [[ "$base" != *.* ]] && return 1
+  ext="${base##*.}"; lc="${ext,,}"
+  IFS=',' read -r -a arr <<< "$ext_list"
+  for e in "${arr[@]}"; do
+    if [[ "$lc" == "$e" ]]; then return 0; fi
+  done
+  return 1
+}
 
-# --- Normalize (rules 2 & 3) ---
 normalize_name() {
-  local name="$1"
+  local name="$1" base ext
-  # Replace one-or-more spaces with single underscore
+  # strip_reserved har redan gjort whitespace → "_" och en första kollaps
-  name="$(printf '%s' "$name" | sed -E 's/[[:space:]]+/_/g')"
-  # Collapse multiple underscores
   name="$(printf '%s' "$name" | sed -E 's/_+/_/g')"
-  # Trim leading/trailing underscores from the basename (keep extension)
+  base="$name"; ext=""
-  local base ext
-  base="$name"
-  ext=""
   if [[ "$base" == *.* ]]; then
     ext=".${base##*.}"
     base="${base%.*}"
   fi
-  base="${base##_}"
+  base="${base#_}"
-  base="${base%%_}"
+  base="${base%_}"
   printf '%s%s' "$base" "$ext"
 }
 
-# --- Remove reserved tokens from basename (plus end-token rule) ---
+candidate_name() {
+  local dst="$1" src="$2" base ext candidate n
+  base="${dst##*/}"; ext=""
+  if [[ "$base" == *.* ]]; then
+    ext=".${base##*.}"
+    base="${base%.*}"
+  fi
+  candidate="$dst"; n=1
+  while [[ -e "$candidate" && "$src" != "$candidate" ]]; do
+    candidate="$(dirname "$dst")/${base}_$n$ext"
+    ((n++))
+  done
+  printf '%s' "$candidate"
+}
+
 strip_reserved() {
-  local name="$1"
+  local name="$1" base ext token lw
-  local base ext token lw
   local -a parts new_parts=()
 
-  # CHANGE: pass `--` to basename for dash-prefixed names
+  base="$(basename "$name")"; ext=""
-  base="$(basename -- "$name")"
-  ext=""
   if [[ "$base" == *.* ]]; then
     ext=".${base##*.}"
     base="${base%.*}"
   fi
 
-  # Pre-normalize for tokenization
+  # Pre-normalisering & tokenisering:
+  #  - ersätt '-' → '_'
+  #  - ersätt all whitespace → '_'
+  #  - kollapsa multipla '_'
+  #  - dela på '_'
   local norm_base
-  norm_base="$(printf '%s' "$base" | sed -E 's/[[:space:]]+/_/g' | sed -E 's/_+/_/g')"
+  norm_base="$(printf '%s' "$base" \
+    | sed -E 's/-/_/g' \
+    | sed -E 's/[[:space:]]+/_/g' \
+    | sed -E 's/_+/_/g')"
 
   IFS='_' read -r -a parts <<< "$norm_base"
+
   for token in "${parts[@]}"; do
     [[ -z "$token" ]] && continue
     lw="${token,,}"
-    if [[ -n "${RESERVED[$lw]:-}" ]]; then
+
+    # (Regel 2) slopa rena siffertokens
+    if [[ "$token" =~ ^[[:digit:]]+$ ]]; then
+      (( VERBOSE )) && log "DEBUG: drop numeric token '$token'"
       continue
     fi
+
+    # (Regel 1) exakt token = reserverat ord (case-insensitivt)
+    if [[ -n "${RESERVED[$lw]:-}" ]]; then
+      (( VERBOSE )) && log "DEBUG: drop reserved token '$token'"
+      continue
+    fi
+
     new_parts+=("$token")
   done
 
@@ -214,63 +249,94 @@ strip_reserved() {
     new_base="$(IFS=_; echo "${new_parts[*]}")"
   fi
 
-  # Extra end-token rule (defensive): if last token is reserved, drop it
+  [[ -z "$new_base" ]] && new_base="untitled"
-  if [[ -n "$new_base" ]]; then
-    local last last_lc
-    last="${new_base##*_}"
-    last_lc="${last,,}"
-    if [[ -n "${RESERVED[$last_lc]:-}" ]]; then
-      if [[ "$new_base" == *_* ]]; then
-        new_base="${new_base%_*}"
-      else
-        new_base=""
-      fi
-    fi
-  fi
-
-  # Fallback if everything disappeared
-  if [[ -z "$new_base" ]]; then
-    new_base="untitled"
-  fi
-
   printf '%s%s' "$new_base" "$ext"
 }
 
-# --- File pattern based on EXT ---
+# --- Walk files ---
 shopt -s nullglob
-pattern="*"
-if [[ "$EXT" != "*" ]]; then
-  pattern="*.$EXT"
-fi
 
-# --- Main loop ---
+VISITED=0
-for path in "$DIR"/$pattern; do
+RENAMED=0
-  [[ -f "$path" ]] || continue
+UNCHANGED=0
+COLLISIONED=0
 
-  orig_basename="$(basename -- "$path")"
+log "rename_and_filter.sh version $SCRIPT_VERSION"
+log "Processing directory: $DIR (dry-run=$DRY_RUN, recursive=$RECURSIVE, min-len=$MIN_LEN, extensions=$EXT_LIST_STR, verbose=$VERBOSE)"
+log "Reserved words loaded: ${#RESERVED[@]}"
 
-  # NEW RULE (minimal): remove ALL leading dashes before processing
+process_one() {
-  clean_basename="$orig_basename"
+  local path="$1"
-  while [[ "$clean_basename" == -* ]]; do
+  [[ -f "$path" ]] || return 0
-    clean_basename="${clean_basename#-}"
-  done
 
-  # 1) remove reserved tokens from basename (on the cleaned name)
+  if ! matches_extension "$path" "$EXT_LIST_STR"; then
-  stripped="$(strip_reserved "$clean_basename")"
+    if (( DRY_RUN && VERBOSE )); then
+      log "SKIP (ext): $path"
+    fi
+    return 0
+  fi
 
-  # 2 & 3) normalize (spaces -> "_", collapse "_", trim)
+  ((VISITED++))
+  if (( DRY_RUN || VERBOSE )); then
+    log "Checking file: $path"
+  fi
+
+  local orig_basename stripped new_name bn target cand
+  orig_basename="$(basename "$path")"
+  stripped="$(strip_reserved "$orig_basename")"
   new_name="$(normalize_name "$stripped")"
 
-  # If nothing changed, do not attempt to rename at all
+  if (( VERBOSE )); then
-  if [[ "$new_name" == "$orig_basename" ]]; then
+    log "DEBUG: orig='${orig_basename}' stripped='${stripped}' normalized='${new_name}'"
-    if (( DRY_RUN )); then
-      log "NOCHANGE (dry-run): \"${orig_basename}\" (nothing to do)"
   fi
-    continue
+
+  bn="$new_name"
+  [[ "$bn" == *.* ]] && bn="${bn%.*}"
+  if (( MIN_LEN > 0 )) && (( ${#bn} < MIN_LEN )); then
+    if (( DRY_RUN || VERBOSE )); then
+      log "NOCHANGE: \"${orig_basename}\" (too short after normalize)"
+    fi
+    ((UNCHANGED++))
+    return 0
+  fi
+
+  if [[ "$new_name" == "$orig_basename" ]]; then
+    if (( DRY_RUN || VERBOSE )); then
+      log "NOCHANGE: \"${orig_basename}\" (nothing to do)"
+    fi
+    ((UNCHANGED++))
+    return 0
   fi
 
   target="$(dirname "$path")/$new_name"
-  rename_file "$path" "$target"
+  cand="$(candidate_name "$target" "$path")"
-done
+  if [[ "$cand" != "$target" ]]; then
+    ((COLLISIONED++))
+  fi
 
+  if (( DRY_RUN )); then
+    log "RENAME: \"$(basename "$path")\" => \"$(basename "$cand")\""
+  else
+    mv -- "$path" "$cand"
+    log "RENAME: \"$(basename "$path")\" => \"$(basename "$cand")\""
+  fi
+  ((RENAMED++))
+}
+
+# --- Huvudloop ---
+if (( RECURSIVE )); then
+  log "[DEBUG] Enter recursive loop (find '$DIR')"
+  while IFS= read -r -d '' path; do
+    log "[DEBUG] raw candidate (recursive): $path"
+    process_one "$path"
+  done < <(find "$DIR" -type f -print0)
+else
+  log "[DEBUG] Enter non-recursive loop over '$DIR'/*"
+  for path in "$DIR"/*; do
+    log "[DEBUG] raw candidate: $path"
+    process_one "$path"
+  done
+fi
+
+log "Summary: visited=$VISITED, renamed=$RENAMED, unchanged=$UNCHANGED, collisions_adjusted=$COLLISIONED"
 log "Done."