File: sawk.sh 1 #!/bin/sh 2 3 # The MIT License (MIT) 4 # 5 # Copyright (c) 2026 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 # sawk [options...] [awk expression...] [files...] 27 # 28 # 29 # Summarize via AWK calculates some numeric statistics from an `awk` 30 # expression. These include: 31 # 32 # - min, max 33 # - (numeric) count, positives, zeros, negatives 34 # - sum, mean 35 # - geometric mean, standard deviation 36 # 37 # When not given an expression, the default is to use whole lines/records. 38 # A `.` (no quotes) for the expression also enables the default behavior. 39 # 40 # The handy case-insensitive shortcut options may cause this tool to fail, 41 # if the main `awk` tool installed doesn't support the special `IGNORECASE` 42 # variable. 43 # 44 # The AWK options available only in single-dash versions are 45 # 46 # -F fs, -Ffs, -F=fs make `fs` the field separator 47 # 48 # The other options are, available both in single and double-dash versions 49 # 50 # -h, -help show this help message 51 # -i, -ins match regexes case-insensitively; may fail `awk` 52 # -tsv split fields using tabs, same as using -F "\t" 53 54 55 case "$1" in 56 -h|--h|-help|--help) 57 awk '/^# +sawk /, /^$/ { gsub(/^# ?/, ""); print }' "$0" 58 exit 0 59 ;; 60 esac 61 62 tsv=0 63 ins=0 64 cmd='awk' 65 66 while [ $# -gt 0 ]; do 67 case "$1" in 68 -F) 69 if [ $# -lt 2 ]; then 70 printf "%s: expected value after -F option\n" "$0" >&2 71 exit 1 72 fi 73 cmd="${cmd} -F $2"; shift 2; continue 74 ;; 75 76 -F*) cmd="${cmd} $1"; shift; continue ;; 77 78 -i|--i|-ins|--ins|-insensitive|--insensitive) ins=1; shift; continue ;; 79 80 -tsv|--tsv) tsv=1; shift; continue ;; 81 82 -v) 83 if [ $# -lt 2 ]; then 84 printf "%s: missing variable assignment after -v option\n" \ 85 "$0" >&2 86 exit 1 87 fi 88 cmd="${cmd} -v $1"; shift 2; continue 89 ;; 90 91 -) break ;; 92 93 --) shift; break ;; 94 95 -*) 96 printf "%s: unsupported option %s\n" "$0" "$1" >&2 97 exit 1 98 ;; 99 esac 100 101 break 102 done 103 104 code="${1:-\$0}" 105 [ $# -gt 0 ] && shift 106 if [ "${code}" = '.' ]; then 107 code='$0' 108 fi 109 110 # show all non-existing files given 111 failed=0 112 for arg in "$@"; do 113 [ "${arg}" = '-' ] && continue 114 [ -e "${arg}" ] && continue 115 printf "%s: no file named \"%s\"\n" "$0" "${arg}" >&2 116 failed=1 117 done 118 119 [ "${failed}" -gt 0 ] && exit 2 120 121 ci=' 122 BEGIN { 123 if (IGNORECASE == "") { 124 m = "your `awk` command lacks case-insensitive regex-matching" 125 print(m) > "/dev/stderr" 126 exit 125 127 } 128 IGNORECASE = 1 129 } 130 ' 131 if [ "${ins}" -eq 0 ]; then 132 ci='' 133 fi 134 135 src="${ci}"' 136 BEGIN { 137 NUMERIC = INTS = POS = ZERO = NEG = 0 138 139 SUM = 0 140 MEAN = 0 141 PROD = 1 142 } 143 144 { 145 V = ('"${code}"') 146 147 if (V !~ /^ *(0|[0-9]+|[0-9]*\.[0-9]+) *$/) next 148 V = V + 0 149 150 NUMERIC++ 151 INTS += V % 1 == 0 152 if (V > 0) POS++ 153 else if (V < 0) NEG++ 154 else if (V == 0) ZERO++ 155 156 if (NUMERIC == 1) MIN = MAX = V 157 MIN = MIN < V ? MIN : V 158 MAX = MAX > V ? MAX : V 159 SUM += V 160 PROD *= V 161 if (V > 0) LNSUM += log(V) 162 163 # advance welford`s algorithm 164 D1 = V - MEAN 165 MEAN += D1 / NUMERIC 166 D2 = V - MEAN 167 MEANSQ += D1 * D2 168 } 169 170 END { 171 SUM = MEAN * NUMERIC 172 173 # separate name-value pairs using tabs, and prepare a 174 # pipeable command which ignores all-zero decimals 175 OFS = "\t" 176 177 print "numeric", NUMERIC 178 if (NUMERIC > 0) { 179 print "min", sprintf("%f", MIN) 180 print "max", sprintf("%f", MAX) 181 print "sum", sprintf("%f", SUM) 182 print "mean", sprintf("%f", MEAN) 183 print "geomean", (ZERO == 0 && NEG == 0) ? 184 sprintf("%f", exp(LNSUM / NUMERIC)) : "" 185 print "sd", sprintf("%f", sqrt(MEANSQ / NUMERIC)) 186 fmt = ((PROD > 1e6) || (PROD < -1e6)) ? "%f" : "%g" 187 print "product", sprintf(fmt, PROD) 188 } else { 189 print "min", "" 190 print "max", "" 191 print "sum", "" 192 print "mean", "" 193 print "geomean", "" 194 print "sd", "" 195 print "product", "" 196 } 197 print "integer", INTS 198 print "positive", POS 199 print "zero", ZERO 200 print "negative", NEG 201 } 202 ' 203 204 if [ "${tsv}" -eq 1 ]; then 205 ${cmd} -F "\t" "${src}" "$@" 206 else 207 ${cmd} "${src}" "$@" 208 fi