在bash中對文本文件進行排序

Question

我有這個文本文件：

0_0_0_0_1_1_1_1_1  [
  0.01155712 0.5775286 0.01599521 0.383362 0.01155712 ]
0_1_1_0_0_1_1_1_232  [
  4.980576e-09 1.21296e-06 0.0001519765 0.9998468 4.980576e-09 ]
0_1_1_0_0_1_1_1_226  [
  0.009718912 0.5821248 0.013627 0.3848103 0.009718912 ]
0_1_1_0_0_1_1_1_227  [
  0.009751211 0.5818524 0.01370924 0.3849359 0.009751211 ]
0_0_0_0_1_1_1_1_0  [
  0.0101164 0.580949 0.01410898 0.3847092 0.0101164 ]
0_0_0_0_1_1_1_1_10  [
  2.829467e-13 0.9999987 1.257838e-06 1.43308e-11 2.829467e-13 ]
0_1_1_0_0_1_1_1_228  [
  1.050669e-09 3.34911e-06 0.0006787472 0.9993179 1.050669e-09 ]
0_0_0_0_1_1_1_1_5  [
  8.811786e-08 2.491895e-05 0.0001959623 0.9997789 8.811786e-08 ]
0_1_1_0_0_1_1_1_229  [
  4.031684e-08 0.0009897997 0.9982967 0.0007134909 4.031684e-08 ]
0_0_0_0_1_1_1_1_6  [
  0.00122948 0.1168619 0.005731953 0.8749472 0.001229479 ]
0_1_1_0_0_1_1_1_230  [
  1.156299e-06 0.9997224 4.28468e-05 0.0002323914 1.1563e-06 ]
0_1_1_0_0_1_1_1_231  [
  1.581783e-08 0.05209186 0.947908 5.398444e-08 1.581783e-08 ]
0_0_0_0_1_1_1_1_7  [
  0.001118228 0.8515728 0.005113816 0.141077 0.001118228 ]
0_1_1_0_0_1_1_1_233  [
  4.050781e-05 0.01894026 0.001323607 0.9796551 4.050781e-05 ]
0_0_0_0_1_1_1_1_8  [
  2.476718e-13 1.084459e-05 0.9999892 7.285134e-09 2.476718e-13 ]
0_1_1_0_0_1_1_1_234  [
  0.0002922548 0.8787904 0.002183357 0.1184417 0.0002922548 ]
0_1_1_0_0_1_1_1_235  [
  8.223566e-12 0.9999944 5.548976e-06 4.713996e-10 8.223566e-12 ]
0_1_1_0_0_1_1_1_236  [
  9.655272e-18 2.265286e-08 1 4.498827e-11 9.655272e-18 ]
0_1_1_0_0_1_1_1_237  [
  1.503673e-06 0.9467446 0.05282303 0.0004293863 1.503673e-06 ]
0_1_1_0_0_1_1_1_238  [
  4.519509e-12 1 1.623816e-08 3.29057e-08 4.519509e-12 ]
0_1_1_0_0_1_1_1_239  [
  0.01033381 0.580116 0.01444763 0.3847687 0.01033381 ]

格式為{name} _ {part} [{data points}]

舉個例子：

0_0_1_0_0_0_1_0 ：將是名字

_51 ：將是零件號

[1.651947e-12 5.555453e-05 0.9999444 3.284811e-09 1.651947e-12 ] ：將是數據點。

不知何故，需要以遞增的零件號對所有具有相同名稱的數據點進行排序和附加...如何以一種簡單的方式完成？ 我想用bash來做是有道理的，但是如何實際通過增加零件數進行附加還是一個謎。

所需的輸出：

0_1_1_0_0_1_1_1 [
  0.009718912 0.5821248 0.013627 0.3848103 0.009718912 
  0.009751211 0.5818524 0.01370924 0.3849359 0.009751211 
  1.050669e-09 3.34911e-06 0.0006787472 0.9993179 1.050669e-09 
  4.031684e-08 0.0009897997 0.9982967 0.0007134909 4.031684e-08 
  1.156299e-06 0.9997224 4.28468e-05 0.0002323914 1.1563e-06 
  1.581783e-08 0.05209186 0.947908 5.398444e-08 1.581783e-08 
  4.980576e-09 1.21296e-06 0.0001519765 0.9998468 4.980576e-09 
  4.050781e-05 0.01894026 0.001323607 0.9796551 4.050781e-05 
  0.0002922548 0.8787904 0.002183357 0.1184417 0.0002922548 
  8.223566e-12 0.9999944 5.548976e-06 4.713996e-10 8.223566e-12 
  9.655272e-18 2.265286e-08 1 4.498827e-11 9.655272e-18 
  1.503673e-06 0.9467446 0.05282303 0.0004293863 1.503673e-06 
  4.519509e-12 1 1.623816e-08 3.29057e-08 4.519509e-12 
  0.01033381 0.580116 0.01444763 0.3847687 0.01033381 ]

0_0_0_0_1_1_1_1 [
  0.0101164 0.580949 0.01410898 0.3847092 0.0101164 
  0.01155712 0.5775286 0.01599521 0.383362 0.01155712 
  8.811786e-08 2.491895e-05 0.0001959623 0.9997789 8.811786e-08 
  0.00122948 0.1168619 0.005731953 0.8749472 0.001229479 
  0.001118228 0.8515728 0.005113816 0.141077 0.001118228 
  2.476718e-13 1.084459e-05 0.9999892 7.285134e-09 2.476718e-13 
  2.829467e-13 0.9999987 1.257838e-06 1.43308e-11 2.829467e-13 ]

Answer 1

awk解決方案：

awk 'NF == 2{ match($1,/^[0-9]+(_[0-9]+){7}/); k = substr($1,RSTART,RLENGTH); next }
     { $NF=""; a[k]=a[k]"\n "$0 }
     END { for(i in a) printf "%s [%s ]\n\n",i,a[i] }' yourfile

---

• NF == 2處理具有2個字段的記錄（即“ name ”值）

• match($1,/^[0-9]+(_[0-9]+){7}/) -捕獲“ 名稱 ”序列

• k = substr($1,RSTART,RLENGTH) -將名稱視為數組鍵

• { $NF=""; a[k]=a[k]"\\n "$0 } { $NF=""; a[k]=a[k]"\\n "$0 } -將每行數據點 （不帶尾隨] ）附加到相關的名稱元素

• printf "%s [%s ]\\n\\n",i,a[i] -打印格式的輸出

輸出：

0_0_0_0_1_1_1_1 [
 0.01155712 0.5775286 0.01599521 0.383362 0.01155712 
 0.0101164 0.580949 0.01410898 0.3847092 0.0101164 
 2.829467e-13 0.9999987 1.257838e-06 1.43308e-11 2.829467e-13 
 8.811786e-08 2.491895e-05 0.0001959623 0.9997789 8.811786e-08 
 0.00122948 0.1168619 0.005731953 0.8749472 0.001229479 
 0.001118228 0.8515728 0.005113816 0.141077 0.001118228 
 2.476718e-13 1.084459e-05 0.9999892 7.285134e-09 2.476718e-13  ]

0_1_1_0_0_1_1_1 [
 4.980576e-09 1.21296e-06 0.0001519765 0.9998468 4.980576e-09 
 0.009718912 0.5821248 0.013627 0.3848103 0.009718912 
 0.009751211 0.5818524 0.01370924 0.3849359 0.009751211 
 1.050669e-09 3.34911e-06 0.0006787472 0.9993179 1.050669e-09 
 4.031684e-08 0.0009897997 0.9982967 0.0007134909 4.031684e-08 
 1.156299e-06 0.9997224 4.28468e-05 0.0002323914 1.1563e-06 
 1.581783e-08 0.05209186 0.947908 5.398444e-08 1.581783e-08 
 4.050781e-05 0.01894026 0.001323607 0.9796551 4.050781e-05 
 0.0002922548 0.8787904 0.002183357 0.1184417 0.0002922548 
 8.223566e-12 0.9999944 5.548976e-06 4.713996e-10 8.223566e-12 
 9.655272e-18 2.265286e-08 1 4.498827e-11 9.655272e-18 
 1.503673e-06 0.9467446 0.05282303 0.0004293863 1.503673e-06 
 4.519509e-12 1 1.623816e-08 3.29057e-08 4.519509e-12 
 0.01033381 0.580116 0.01444763 0.3847687 0.01033381  ]

Answer 2

perl -ne '
    if ($.%2==1) {
        # odd line numbers
        ($name,$part) = /(.*)_(\d+)/;
        # debug: print "name:$name part:$part\n";
    } else {
        # even line numbers
        ($datapoints) = /(.*)\]/;
        $hash{$name}{$part}=$datapoints;
    }
    END {
        while (($key,$v)=each %hash) {
            print "$key [\n", (
                map "${$v}{$_}\n", sort {$a<=>$b} keys %{$v}
            ), "]\n";
        }

        # or with debug
        # while (($key,$v)=each %hash) {
        #    print "$key [\n", (
        #        map ":$_:${$v}{$_}\n", sort {$a<=>$b} keys %{$v}
        #    ), "]\n";
        # }
    }
' input.txt

如果輸入不完全遵循給定格式，則另一個版本更安全

perl -n00e '
    while ( /([\d_]*)_(\d*) \s* \[ \s* (.*?) \s* \]/gmsx ) {
        ($name,$part,$datapoints) = ($1,$2,$3);
        $hash{$name}{$part}=$datapoints;
    }
    while (($key,$v)=each %hash) {
        print "$key [\n", (
            map "${$v}{$_}\n", sort {$a<=>$b} keys %{$v}
        ), "]\n";
    }
' input.txt