charsets

README.md

@@ -95,6 +95,11 @@ TEMPLATES=(
 | `{num:N}` | N digits | ~3.3 bits/char |
 | `{special:N}` | N from `!?@#%~=+` | ~3.0 bits/char |
 | `{NAME:N-M}` | N to M chars from charset NAME | log2(Σ size^L) |
+| `{hex:N}` | `3f9a` | 4.0 bits/char |
+| `{b58:N}` | `Vk2x` (Base58, no 0OIl) | ~5.9 bits/char |
+| `{b64:N}` | `hG_4` (URL-safe base64) | 6.0 bits/char |
+| `{alnum:N}` | `xR4x` | ~6.0 bits/char |
+| `{ascii:N}` | `p/(!` (all printable, no space) | ~6.5 bits/char |
 
 \* Word entropy depends on pool size. Values shown assume all built-in lists enabled (~14k words, ~13.8 bits/word).
 
@@ -109,7 +114,7 @@ CHARSETS=(
 )
 ```
 
-Names must match `[a-z][a-z0-9_]*` and override the built-in four (alpha, safe, num, special). Avoid duplicate characters; himi warns if it sees them, since they skew the distribution and overstate entropy.
+Names must match `[a-z][a-z0-9_]*` and override any built-in of the same name (alpha, safe, num, special, hex, b58, b64, alnum, ascii). Avoid duplicate characters; himi warns if it sees them, since they skew the distribution and overstate entropy.
 
 Range tokens pick the length weighted by `size^L`, so every possible output is equally likely. `{num:1-2}` covers 0-99 plus 0-9 (110 outcomes, ~6.8 bits), not a 50/50 coin flip on length, which would tank min-entropy.
 

config.sh

@@ -42,7 +42,8 @@ WORD_MAX_LENGTH=9
 #   {Word}       — random word (Title Case)
 #   {word!}      — random lower or UPPER (+1 bit entropy)
 #   {Word!}      — random lower/Title/UPPER (+1.58 bits entropy)
-#   {NAME:N}     — N chars from charset NAME (alpha/safe/num/special/custom)
+#   {NAME:N}     — N chars from charset NAME
+#                  built-in: alpha safe num special hex b58 b64 alnum ascii
 #   {NAME:N-M}   — N to M chars, weighted so every output is equally likely
 #                  (entropy = log2 of total outcomes, e.g. {num:1-2} ≈ 6.8 bits)
 # ------------------------------------------------------------------------------

himi

@@ -28,6 +28,9 @@ readonly CONFIG_SCHEMA_VERSION=2
 
 # Generic charset token: {name:N} or {name:N-M}
 readonly TOKEN_RE='\{([a-z][a-z0-9_]*):([0-9]+)(-([0-9]+))?\}'
+# Anything that still looks like a token after expansion (typo'd or unknown).
+# Shape-based so charsets containing { } (e.g. ascii) don't false-positive.
+readonly LEFTOVER_RE='\{[A-Za-z][A-Za-z0-9_]*(!|:[0-9]+(-[0-9]+)?)?\}'
 
 
 # ------------------------------------------------------------------------------
@@ -37,6 +40,11 @@ ALPHA_CHARSET='0123456789ABCDEFGHJKMNPQRSTVWXYZ'  # Crockford Base32 (32 chars,
 SAFE_CHARSET='ACDEFGHJKMNPQRTWXY234679'           # Ultra-safe (24 chars, 4.6 bits) - no B/8, 0/O, 1/I/L, 5/S
 NUM_CHARSET='0123456789'
 SPECIAL_CHARSET='!?@#%~=+'
+HEX_CHARSET='0123456789abcdef'                                          # 16 chars, 4.0 bits
+B58_CHARSET='123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz' # Base58: no 0OIl (58 chars, 5.86 bits)
+B64_CHARSET='ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_' # URL-safe base64 (64 chars, 6.0 bits)
+ALNUM_CHARSET='ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789' # 62 chars, 5.95 bits
+ASCII_CHARSET='!"#$%&'\''()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~' # all printable, no space (94 chars, 6.55 bits)
 WORD_MIN_LENGTH=4
 WORD_MAX_LENGTH=8
 
@@ -200,6 +208,11 @@ get_charset() {
     safe)    printf '%s' "$SAFE_CHARSET" ;;
     num)     printf '%s' "$NUM_CHARSET" ;;
     special) printf '%s' "$SPECIAL_CHARSET" ;;
+    hex)     printf '%s' "$HEX_CHARSET" ;;
+    b58)     printf '%s' "$B58_CHARSET" ;;
+    b64)     printf '%s' "$B64_CHARSET" ;;
+    alnum)   printf '%s' "$ALNUM_CHARSET" ;;
+    ascii)   printf '%s' "$ASCII_CHARSET" ;;
     *)       return 1 ;;
   esac
 }
@@ -521,6 +534,9 @@ cmd_list() {
 # Uninstall: remove himi and all data
 # ------------------------------------------------------------------------------
 cmd_uninstall() {
+  local purge="false"
+  [[ "${1:-}" == "--purge" ]] && purge="true"
+
   read -p "Are you sure you want to remove himi and all custom wordlists? (y/N) " -n 1 -r
   echo
   if [[ ! $REPLY =~ ^[Yy]$ ]]; then
@@ -543,23 +559,43 @@ cmd_uninstall() {
           # Linux/GNU
           sed -i.bak "/$start_marker/,/$end_marker/d" "$HOME/$p"
         fi
-        ok "Cleaned up $p (backup created as $p.bak)"
+        if [[ "$purge" == "true" ]]; then
+          rm -f "$HOME/$p.bak"
+          ok "Cleaned up $p (no backup, --purge)"
+        else
+          ok "Cleaned up $p (backup created as $p.bak)"
+        fi
       fi
     fi
   done
 
-  # 2. Remove completions
+  # 2. Remove completions (and the .zfunc dir if we leave it empty)
   rm -f "${HOME}/.zfunc/_himi"
+  rmdir "${HOME}/.zfunc" 2>/dev/null || true
   rm -f "${XDG_DATA_HOME:-$HOME/.local/share}/bash-completion/completions/himi"
 
   # 3. Remove data and binary
   rm -rf "$CONFIG_DIR"
   ok "Removed configuration and wordlists."
 
-  # Final step: Binary deletes itself
+  # Final step: binary deletes itself. Prefer the file actually running
-  local self
+  # (works for ./himi too); fall back to PATH lookup without tripping set -e.
-  self=$(command -v himi)
+  local self="${BASH_SOURCE[0]}"
-  rm -f "$self"
+  if [[ ! -f "$self" ]]; then
+    self=$(command -v himi) || self=""
+  fi
+  [[ -n "$self" ]] && rm -f "$self"
+
+  # Flag any other copies still lurking in PATH
+  local leftover
+  if leftover=$(command -v himi 2>/dev/null) && [[ -n "$leftover" ]]; then
+    warn "another himi remains at: $leftover — remove it manually"
+  fi
+
+  if [[ "$purge" != "true" ]]; then
+    dim "Profile backups (*.bak) were kept; use 'himi uninstall --purge' for none."
+    printf "\n"
+  fi
 
   ok "himi has been uninstalled."
 }
@@ -953,7 +989,7 @@ cmd_test() {
   local example leftovers=""
   for ((i = 0; i < 5; i++)); do
     example=$(expand_template "$pattern")
-    [[ "$example" == *'{'* && "$example" == *'}'* ]] && leftovers="true"
+    [[ "$example" =~ $LEFTOVER_RE ]] && leftovers="true"
     printf "    %s\n" "$example"
   done
   printf "\n"
@@ -1012,7 +1048,7 @@ cmd_generate() {
     secrets+=$(expand_template "$pattern")
   done
 
-  if [[ "$secrets" == *'{'* && "$secrets" == *'}'* ]]; then
+  if [[ "$secrets" =~ $LEFTOVER_RE ]]; then
     warn "output contains unexpanded {token}s — check the template ('himi test' helps)"
   fi
 
@@ -1054,7 +1090,7 @@ Usage:
   himi entropy [name]       Show entropy for template(s)
   himi test '{pattern}'     Test a custom pattern
   himi config               Edit config in \$EDITOR
-  himi uninstall            Remove himi and all data
+  himi uninstall            Remove himi and all data (--purge: no profile backups)
   himi -h, --help           Show this help
 
 Templates (configure in $CONFIG_FILE):
@@ -1082,6 +1118,11 @@ EOF
   printf "  %-12s %s\n" "safe" "$SAFE_CHARSET"
   printf "  %-12s %s\n" "num" "$NUM_CHARSET"
   printf "  %-12s %s\n" "special" "$SPECIAL_CHARSET"
+  printf "  %-12s %s\n" "hex" "$HEX_CHARSET"
+  printf "  %-12s %s\n" "b58" "$B58_CHARSET"
+  printf "  %-12s %s\n" "b64" "$B64_CHARSET"
+  printf "  %-12s %s\n" "alnum" "$ALNUM_CHARSET"
+  printf "  %-12s %s\n" "ascii" "$ASCII_CHARSET"
   for entry in ${CHARSETS[@]+"${CHARSETS[@]}"}; do
     IFS='|' read -r name pattern <<< "$entry"
     printf "  %-12s %s\n" "$name" "$pattern"