File: rtsv.sh
   1 #!/bin/sh
   2 
   3 # The MIT License (MIT)
   4 #
   5 # Copyright © 2025 pacman64
   6 #
   7 # Permission is hereby granted, free of charge, to any person obtaining a copy
   8 # of this software and associated documentation files (the “Software”), to deal
   9 # in the Software without restriction, including without limitation the rights
  10 # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  11 # copies of the Software, and to permit persons to whom the Software is
  12 # furnished to do so, subject to the following conditions:
  13 #
  14 # The above copyright notice and this permission notice shall be included in
  15 # all copies or substantial portions of the Software.
  16 #
  17 # THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  18 # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  19 # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  20 # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  21 # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  22 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  23 # SOFTWARE.
  24 
  25 
  26 # rtsv [options...] [filenames...]
  27 #
  28 # Realign Tab Separated-Values, by padding with spaces to match each column's
  29 # widest value, right-aligning all numbers.
  30 
  31 
  32 case "$1" in
  33     -h|--h|-help|--help)
  34         awk '/^# +rtsv /, /^$/ { gsub(/^# ?/, ""); print }' "$0"
  35         exit 0
  36     ;;
  37 esac
  38 
  39 [ "$1" = "--" ] && shift
  40 
  41 command='awk'
  42 if [ -e /usr/bin/gawk ]; then
  43     command='gawk'
  44 fi
  45 
  46 # show all non-existing files given
  47 failed=0
  48 for arg in "$@"; do
  49     if [ "${arg}" = "-" ]; then
  50         continue
  51     fi
  52     if [ ! -e "${arg}" ]; then
  53         printf "no file named \"%s\"\n" "${arg}" > /dev/stderr
  54         failed=1
  55     fi
  56 done
  57 
  58 if [ "${failed}" -gt 0 ]; then
  59     exit 2
  60 fi
  61 
  62 ${command} -F "\t" '
  63     function match_number(v) {
  64         return match(v, /^[+-]?[0-9]+(\.[0-9]+)?$/)
  65     }
  66 
  67     function match_dot_digits(v) {
  68         return match(v, /\.[0-9]+$/)
  69     }
  70 
  71     {
  72         gsub(/\r$/, "")
  73 
  74         for (i = 1; i <= NF; i++) {
  75             data[NR][i] = $i
  76 
  77             if (match_number($i)) {
  78                 if (match_dot_digits($i)) {
  79                     dd = RLENGTH
  80                     if (dot_decs[i] < dd) dot_decs[i] = dd
  81                     iw = RSTART - 1
  82                     if (int_widths[i] < iw) int_widths[i] = iw
  83                 } else {
  84                     w = length($i)
  85                     if (int_widths[i] < w) int_widths[i] = w
  86                 }
  87 
  88                 continue
  89             }
  90 
  91             w = length($i)
  92             if (widths[i] < w) widths[i] = w
  93         }
  94     }
  95 
  96     END {
  97         # fix column-widths using the number-padding info
  98         for (i = 1; i <= NF; i++) {
  99             w = int_widths[i] + dot_decs[i]
 100             if (widths[i] < w) widths[i] = w
 101         }
 102 
 103         for (i = 1; i <= NR; i++) {
 104             last = length(data[i])
 105 
 106             for (j = 1; j <= last; j++) {
 107                 v = data[i][j]
 108 
 109                 # put 2-space gaps between columns
 110                 if (j > 1 && j < last) printf "  "
 111                 else if (j == last && v != "") printf "  "
 112 
 113                 if (!match_number(v)) {
 114                     # avoid adding trailing spaces at the end of lines
 115                     printf "%*s", (j == last) ? 0 : -widths[j], v
 116                     continue
 117                 }
 118 
 119                 w = length(v)
 120                 if (match_dot_digits(v)) {
 121                     dd = RLENGTH
 122                     iw = RSTART - 1
 123                 } else {
 124                     dd = 0
 125                     iw = w
 126                 }
 127 
 128                 dpad = dot_decs[j] - dd
 129                 ipad = int_widths[j] - iw
 130                 if (ipad < 0) ipad = 0
 131                 lpad = widths[j] - (ipad + w + dpad)
 132                 if (lpad < 0) lpad = 0
 133 
 134                 # avoid adding trailing spaces at the end of lines
 135                 if (j == last) dpad = 0
 136 
 137                 printf "%*s%*s%s%*s", lpad, "", ipad, "", v, dpad, ""
 138             }
 139 
 140             printf "\n"
 141         }
 142     }
 143 ' "$@"