主页 / ubuntu / 问题 / 1342541
Accepted
Yunfei Chen
Asked: 2021-06-04 12:45:59 +0800 CST

基于文本文件的排序

  • 772

我正在尝试根据列中的第二个字段对文本文件进行排序，所以基本上我有一个像

"Year","District Code","District Name","Neighborhood Code","Neighborhood Name","Gender","Number"
"2017","1","Ciutat Vella","1","el Raval","Boys",283
"2017","1","Ciutat Vella","2","el Barri Gòtic","Boys",56
"2017","1","Ciutat Vella","3","la Barceloneta","Boys",51
"2017","1","Ciutat Vella","4","Sant Pere, Santa Caterina i la Ribera","Boys",90
"2017","2","Eixample","5","el Fort Pienc","Boys",117
"2017","2","Eixample","6","la Sagrada Família","Boys",207
"2017","2","Eixample","7","la Dreta de l'Eixample","Boys",185
"2017","2","Eixample","8","l'Antiga Esquerra de l'Eixample","Boys",176
"2017","2","Eixample","9","la Nova Esquerra de l'Eixample","Boys",218
"2017","2","Eixample","10","Sant Antoni","Boys",172
"2017","3","Sants-Montjuïc","11","el Poble Sec","Boys",156
"2017","3","Sants-Montjuïc","12","la Marina del Prat Vermell","Boys",4
"2017","3","Sants-Montjuïc","13","la Marina de Port","Boys",119
"2017","3","Sants-Montjuïc","14","la Font de la Guatlla","Boys",42
"2017","3","Sants-Montjuïc","15","Hostafrancs","Boys",78
"2017","3","Sants-Montjuïc","16","la Bordeta","Boys",81
"2017","3","Sants-Montjuïc","17","Sants - Badal","Boys",108
"2017","3","Sants-Montjuïc","18","Sants","Boys",155
"2017","4","Les Corts","19","les Corts","Boys",199
"2017","4","Les Corts","20","la Maternitat i Sant Ramon","Boys",101
"2017","4","Les Corts","21","Pedralbes","Boys",32
"2017","5","Sarrià-Sant Gervasi","22","Vallvidrera, el Tibidabo i les Planes","Boys",17
"2017","5","Sarrià-Sant Gervasi","23","Sarrià","Boys",122
"2017","5","Sarrià-Sant Gervasi","24","les Tres Torres","Boys",74
"2017","5","Sarrià-Sant Gervasi","25","Sant Gervasi - la Bonanova","Boys",104
"2017","5","Sarrià-Sant Gervasi","26","Sant Gervasi - Galvany","Boys",206
"2017","5","Sarrià-Sant Gervasi","27","el Putxet i el Farró","Boys",127
"2017","6","Gràcia","28","Vallcarca i els Penitents","Boys",66
"2017","6","Gràcia","29","el Coll","Boys",26
"2017","6","Gràcia","30","la Salut","Boys",62
"2017","6","Gràcia","31","la Vila de Gràcia","Boys",241
"2017","6","Gràcia","32","el Camp d'en Grassot i Gràcia Nova","Boys",149
"2017","7","Horta-Guinardó","33","el Baix Guinardó","Boys",97
"2017","7","Horta-Guinardó","34","Can Baró","Boys",41
"2017","7","Horta-Guinardó","35","el Guinardó","Boys",148
"2017","7","Horta-Guinardó","36","la Font d'en Fargues","Boys",26
"2017","7","Horta-Guinardó","37","el Carmel","Boys",123
"2017","7","Horta-Guinardó","38","la Teixonera","Boys",48
"2017","7","Horta-Guinardó","39","Sant Genís dels Agudells","Boys",28
"2017","7","Horta-Guinardó","40","Montbau","Boys",26
"2017","7","Horta-Guinardó","41","la Vall d'Hebron","Boys",12
"2017","7","Horta-Guinardó","42","la Clota","Boys",8
"2017","7","Horta-Guinardó","43","Horta","Boys",9

我想根据 District code 对整个文件进行排序，然后打印出结果，我只是在这里复制了几行，因为文件有 1000 多行所以。我尝试了几个命令：

cat $1 | sort -n -k 2
cat $1 | sort -t \" -n -k 2

但它似乎什么都不做？

还有一些行未排序，例如：

"2017","10","Sant Martí","64","el Camp de l'Arpa del Clot","Boys",148
"2017","10","Sant Martí","65","el Clot","Boys",125
"2017","10","Sant Martí","66","el Parc i la Llacuna del Poblenou","Boys",79
"2017","10","Sant Martí","67","la Vila Olímpica del Poblenou","Boys",34
"2017","10","Sant Martí","68","el Poblenou","Boys",160
"2017","10","Sant Martí","69","Diagonal Mar i el Front Marítim del Poblenou","Boys",103
"2017","10","Sant Martí","70","el Besòs i el Maresme","Boys",119
"2017","10","Sant Martí","71","Provençals del Poblenou","Boys",106
"2017","10","Sant Martí","72","Sant Martí de Provençals","Boys",100
"2017","10","Sant Martí","73","la Verneda i la Pau","Boys",122
"2017","1","Ciutat Vella","1","el Raval","Girls",219
"2017","1","Ciutat Vella","2","el Barri Gòtic","Girls",53
"2017","1","Ciutat Vella","3","la Barceloneta","Girls",50
"2017","1","Ciutat Vella","4","Sant Pere, Santa Caterina i la Ribera","Girls",90
"2017","2","Eixample","5","el Fort Pienc","Girls",114
"2017","2","Eixample","6","la Sagrada Família","Girls",183
"2017","2","Eixample","7","la Dreta de l'Eixample","Girls",159

所以以防万一你想知道。谢谢...

text-processing

2 个回答

  1. Best Answer

    试试这个代码：

    sort -t'"' -n -k4 "input.txt" > "output.txt"

    到底是怎么回事？

    它使用双引号"作为分隔符。这让我们可以将第二个引用的数字列作为“数字”而不是字符串。但它搞乱了列的计数。因为每一行都以"排序开头，所以在“年份”之前考虑了一个空列。同样，“年份”和“区号”之间的逗号分隔符也被视为一列。因此区代码是-k4，而不是-k2。

    如果我们使用逗号,作为分隔符，则地区代码用双引号引起来，并按排序视为字符串。

    input.txt是输入文件的名称。是output.txt输出文件的名称。

    参考：https ://stackoverflow.com/questions/24688548/bash-sort-quoted-csv-files-by-numeric-key

    希望这可以帮助

    • 4

  2. District Code问题是即使您使用正确的分隔符，引号也会使您的字段成为非数字字段,。

    对于 CSV 输入，我建议使用支持 CSV 的工具 - 除了处理引用之外，它还将处理标题（如果您使用 plain ，则需要将其删除并重新插入sort）。

    例如，您可以csvsort从基于 Python 的csvkit包中使用。不幸的是，默认情况下它会删除不必要的字段引用 - 但您可以csvformat从同一个包中将其放回去：

    csvsort -c "District Code" file.csv | csvformat -U1

    -I由于 csvsort 默认执行类型推断，您可以免费获得数字排序 - 使用或关闭推断--no-inference以查看它产生的差异。

    或者，您可以使用Miller（可从 Ubuntuuniverse存储库获得）：

    mlr --csv --quote-all sort -n "District Code" file.csv

    如果您的文件适合更简单的格式，您可以尝试--csvlite代替。--csv

    如果两者csvkit都不mlr可用，那么您可以使用 GNU 执行类似的操作awk。基本思想是创建两个数组，都由记录号索引，一个包含整行，另一个包含（数字转换的）第二列值。然后对后者进行数字排序并使用其排序索引来打印行：

    gawk -F, '
  {line[NR] = $0; gsub(/^"|"$/,"",$2); key[NR] = $2} 
  END {PROCINFO["sorted_in"] = "@val_num_asc"; for(k in key) print line[k]}
' file.csv

    或者（更优雅地）使用具有自定义比较功能的单个数组

    gawk -F, '
  function mycmp (i1, v1, i2, v2) {
    split(v1,a); gsub(/^"|"$/,"",a[2]); 
    split(v2,b); gsub(/^"|"$/,"",b[2]); 
    return a[2] - b[2]
  }
  {
    line[NR] = $0
  } 
  END {
    PROCINFO["sorted_in"] = "mycmp"; 
    for (i in line) print line[i]
  }
' file.csv

    请注意，在这两种情况下，标题行都保留在原位，因为列名的District Code数值计算为0，并且没有任何列值小于"1"。更正式的正确解决方案是用自己的NR==1 {print; next}规则处理标题行。

    请参阅GNU Awk 用户指南：12.2.1 控制数组遍历

    • 3

相关问题