如何在模式中合并多行字符串（但不包括模式）？答案

【问题标题】：How merge multi-line string within a pattern (but excluding the pattern)?如何在模式中合并多行字符串（但不包括模式）？
【发布时间】：2019-05-12 06:51:09
【问题描述】：

我有一个逗号分隔的 CSV 文件（来自数据库），但不幸的是，最后一个字段是双引号多行字符串，如下所示：

138749,CJIKMN,"d4IFtjCCBbIGCSqGSIb3DQEHAqCCBaMwggWfAgEDMQ0wCwYJYIZIAWUDBAIBMG4GBmeBCAEBAaBk
BGIwYAIBADALBglghkgBZQMEAgEwTjAlAgEBBCAeyMDmgdZS30d5JSraWWoUX50J1vKONjxUYxK9
iPZWWjAlAgECBCCzqs7CzH7+3j3trSz+/dcCmud3/Jo9ZYFmN4VTvTjB56CCBBowggQWMIIDnaAD
Lp69+Z3QgAIgHOYjzAQlDRHnDJ/zDtlkWN5pq7T7h3ef9Mnv4ocSuAA="
136065,CIJEPY,"d4IF4jCCBd4GCSqGSIb3DQEHAqCCBc8wggXLAgEDMQ0wCwYJYIZIAWUDBAIBMIGYBgZngQgBAQGg
gY0EgYowgYcCAQAwCwYJYIZIAWUDBAIBMHUwJQIBAQQgNQdsXvKebYUdH0JybzpY2evf+v9Xg86b
hkjOPQQDAjBBMQswCQYDVQQGEwJHQjEOMAwGA1UEChMFVUtLUEExIjAgBgNVBAMTGUNvdW50cnkg
LUxRjUXbTgfGwUKOFwemsc4KXbsLZ13MkbNfAQ=="

如何合并这些多行，不包括引号，保持其余部分不变？我能想到的都是： sed '/\"/{n;:l N;/\"/b; s/\n//; bl}' sampleOut.txt 但这不是我想要的。我正在寻找这个：

138749,CJIKMN,d4IFtjCCBbIGCSqGSIb3DQEHAqCCBaMwggWfAgEDMQ0wCwYJYIZIAWUDBAIBMG4GBmeBCAEBAaBkBGIwYAIBADALBglghkgBZQMEAgEwTjAlAgEBBCAeyMDmgdZS30d5JSraWWoUX50J1vKONjxUYxK9iPZWWjAlAgECBCCzqs7CzH7+3j3trSz+/dcCmud3/Jo9ZYFmN4VTvTjB56CCBBowggQWMIIDnaADLp69+Z3QgAIgHOYjzAQlDRHnDJ/zDtlkWN5pq7T7h3ef9Mnv4ocSuAA=

知道我该怎么做吗？ awk 也可以。

-桑

【问题讨论】：

标签： bash awk sed pattern-matching

【解决方案1】：

您能否尝试以下操作（稍后会添加说明）。

awk '
/,\"/{
  val=$0
  gsub(/\"/,"",val)
  next
}
/\"$/{
  gsub(/\"/,"")
  print val $0
  val=""
  next
}
{
  gsub(/\"/,"")
  val=val?val $0:$0
}
END{
  if(val){
    print val
  }
}'  Input_file

说明：现在为上述代码添加说明。

awk '
/,\"/{                 ##Checking condition if a line has comma with " in it then do following.
  val=$0               ##Assigning current line value to variable val here.
  gsub(/\"/,"",val)
  next                 ##Using next will skip all further statements.
}
/\"$/{                 ##Checking condition if a line is ending with " then do following.
  gsub(/\"/,"")
  print val $0         ##Printing variable val and current line value here.
  val=""               ##Nullifying variable val value here.
  next                 ##Using next will skip all further statements.
}                      ##Closing block for condition here.
{
  gsub(/\"/,"")
  val=val?val $0:$0    ##Creating a variable named val whose value is current line value and its concatenating its own value in it.
}
END{                   ##END section of awk is getting started here.
  if(val){             ##Checking condition if variable val is NOT NULL, if yes then do following.
    print val          ##Printing variable val value here.
  }                    ##Closing block of if condition here.
}' Input_file          ##mentioning Input_file name here.

【讨论】：

我认为您已经假设特定字段应该是最后一列，但事实可能并非如此，但我可以接受。我会对尝试在双引号中查找所有内容的解决方案感兴趣。我@我试试看，然后在这里发帖。
@MacUsers，请务必检查并让我知道您是否有任何其他条件，然后也可以编辑解决方案。

【解决方案2】：

sed 用于在单独的行上执行 s/old/new，仅此而已。使用 GNU awk 进行多字符 RS 和 RT：

$ awk -v RS='"[^"]+"' -v ORS= '{gsub(/[\n"]/,"",RT); print $0 RT}' file
138749,CJIKMN,d4IFtjCCBbIGCSqGSIb3DQEHAqCCBaMwggWfAgEDMQ0wCwYJYIZIAWUDBAIBMG4GBmeBCAEBAaBkBGIwYAIBADALBglghkgBZQMEAgEwTjAlAgEBBCAeyMDmgdZS30d5JSraWWoUX50J1vKONjxUYxK9iPZWWjAlAgECBCCzqs7CzH7+3j3trSz+/dcCmud3/Jo9ZYFmN4VTvTjB56CCBBowggQWMIIDnaADLp69+Z3QgAIgHOYjzAQlDRHnDJ/zDtlkWN5pq7T7h3ef9Mnv4ocSuAA=
136065,CIJEPY,d4IF4jCCBd4GCSqGSIb3DQEHAqCCBc8wggXLAgEDMQ0wCwYJYIZIAWUDBAIBMIGYBgZngQgBAQGggY0EgYowgYcCAQAwCwYJYIZIAWUDBAIBMHUwJQIBAQQgNQdsXvKebYUdH0JybzpY2evf+v9Xg86bhkjOPQQDAjBBMQswCQYDVQQGEwJHQjEOMAwGA1UEChMFVUtLUEExIjAgBgNVBAMTGUNvdW50cnkgLUxRjUXbTgfGwUKOFwemsc4KXbsLZ13MkbNfAQ==

【讨论】：

【解决方案3】：

试试这个 Perl 解决方案：

$ cat mac.txt
138749,CJIKMN,"d4IFtjCCBbIGCSqGSIb3DQEHAqCCBaMwggWfAgEDMQ0wCwYJYIZIAWUDBAIBMG4GBmeBCAEBAaBk
BGIwYAIBADALBglghkgBZQMEAgEwTjAlAgEBBCAeyMDmgdZS30d5JSraWWoUX50J1vKONjxUYxK9
iPZWWjAlAgECBCCzqs7CzH7+3j3trSz+/dcCmud3/Jo9ZYFmN4VTvTjB56CCBBowggQWMIIDnaAD
Lp69+Z3QgAIgHOYjzAQlDRHnDJ/zDtlkWN5pq7T7h3ef9Mnv4ocSuAA="
136065,CIJEPY,"d4IF4jCCBd4GCSqGSIb3DQEHAqCCBc8wggXLAgEDMQ0wCwYJYIZIAWUDBAIBMIGYBgZngQgBAQGg
gY0EgYowgYcCAQAwCwYJYIZIAWUDBAIBMHUwJQIBAQQgNQdsXvKebYUdH0JybzpY2evf+v9Xg86b
hkjOPQQDAjBBMQswCQYDVQQGEwJHQjEOMAwGA1UEChMFVUtLUEExIjAgBgNVBAMTGUNvdW50cnkg
LUxRjUXbTgfGwUKOFwemsc4KXbsLZ13MkbNfAQ=="
$ perl -ne ' chomp; if( /"$/) { s/\"//g;print $_,"\n" } else { s/\"//g; print } ' mac.txt  | nl
     1  138749,CJIKMN,d4IFtjCCBbIGCSqGSIb3DQEHAqCCBaMwggWfAgEDMQ0wCwYJYIZIAWUDBAIBMG4GBmeBCAEBAaBkBGIwYAIBADALBglghkgBZQMEAgEwTjAlAgEBBCAeyMDmgdZS30d5JSraWWoUX50J1vKONjxUYxK9iPZWWjAlAgECBCCzqs7CzH7+3j3trSz+/dcCmud3/Jo9ZYFmN4VTvTjB56CCBBowggQWMIIDnaADLp69+Z3QgAIgHOYjzAQlDRHnDJ/zDtlkWN5pq7T7h3ef9Mnv4ocSuAA=
     2  136065,CIJEPY,d4IF4jCCBd4GCSqGSIb3DQEHAqCCBc8wggXLAgEDMQ0wCwYJYIZIAWUDBAIBMIGYBgZngQgBAQGggY0EgYowgYcCAQAwCwYJYIZIAWUDBAIBMHUwJQIBAQQgNQdsXvKebYUdH0JybzpY2evf+v9Xg86bhkjOPQQDAjBBMQswCQYDVQQGEwJHQjEOMAwGA1UEChMFVUtLUEExIjAgBgNVBAMTGUNvdW50cnkgLUxRjUXbTgfGwUKOFwemsc4KXbsLZ13MkbNfAQ==
$

【讨论】：