File: realign.sh
   1 #!/bin/sh
   2 
   3 # The MIT License (MIT)
   4 #
   5 # Copyright © 2024 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 # realign [options...] [filenames...]
  27 #
  28 # Realign all detected columns, right-aligning any detected numbers in any
  29 # column.
  30 #
  31 # To realign columns from TSV (tab-separated values) lines, use any among
  32 # these as the leading option: `-t`, `-tab` `-tabs`, or `-tsv`.
  33 #
  34 # Double-dashed variants of supported options are also supported.
  35 
  36 
  37 # handle leading options
  38 case "$1" in
  39     -h|--h|-help|--help)
  40         awk '/^# +realign/, /^$/ { gsub(/^# ?/, ""); print }' "$0"
  41         exit 0
  42     ;;
  43 esac
  44 
  45 
  46 awk '
  47 function match_number(v) {
  48     return match(v, /^[+-]?[0-9]+(\.[0-9]+)?$/)
  49 }
  50 
  51 function match_dot_digits(v) {
  52     return match(v, /\.[0-9]+$/)
  53 }
  54 
  55 { gsub(/\r$/, "") }
  56 
  57 NR == 1 && /\t/ { FS = "\t"; $0 = $0 }
  58 
  59 {
  60     for (i = 1; i <= NF; i++) {
  61         data[NR][i] = $i
  62 
  63         if (match_number($i)) {
  64             if (match_dot_digits($i)) {
  65                 dd = RLENGTH
  66                 if (dot_decs[i] < dd) dot_decs[i] = dd
  67                 iw = RSTART - 1
  68                 if (int_widths[i] < iw) int_widths[i] = iw
  69             } else {
  70                 w = length($i)
  71                 if (int_widths[i] < w) int_widths[i] = w
  72             }
  73 
  74             continue
  75         }
  76 
  77         w = length($i)
  78         if (widths[i] < w) widths[i] = w
  79     }
  80 }
  81 
  82 END {
  83     # fix column-widths using the number-padding info
  84     for (i = 1; i <= NF; i++) {
  85         w = int_widths[i] + dot_decs[i]
  86         if (widths[i] < w) widths[i] = w
  87     }
  88 
  89     for (i = 1; i <= NR; i++) {
  90         last = length(data[i])
  91 
  92         for (j = 1; j <= last; j++) {
  93             if (j > 1) printf "  " # put 2-space gaps between columns
  94 
  95             v = data[i][j]
  96 
  97             if (!match_number(v)) {
  98                 # avoid adding trailing spaces at the end of lines
  99                 printf "%*s", (j == last) ? 0 : -widths[j], v
 100                 continue
 101             }
 102 
 103             w = length(v)
 104             if (match_dot_digits(v)) {
 105                 dd = RLENGTH
 106                 iw = RSTART - 1
 107             } else {
 108                 dd = 0
 109                 iw = w
 110             }
 111 
 112             dpad = dot_decs[j] - dd
 113             ipad = int_widths[j] - iw
 114             if (ipad < 0) ipad = 0
 115             lpad = widths[j] - (ipad + w + dpad)
 116             if (lpad < 0) lpad = 0
 117 
 118             # avoid adding trailing spaces at the end of lines
 119             if (j == last) dpad = 0
 120 
 121             printf "%*s%*s%s%*s", lpad, "", ipad, "", v, dpad, ""
 122         }
 123 
 124         printf "\n"
 125     }
 126 }' "$@"