awk 比较多行字段并合并

Question

我的文件中有数据，其模式如下所示 -

  TRSSOEC TRBSOEC      ADICOES                        DELETE            TROEC
  TRSSOEC TRBSOEC      ADICOES                        INSERT            TROEC
  TRSSOEC TRBSOEC      ADICOES                        SELECT            TROEC
  SOBCXYZ SCHEMA1      AGENTES_CARGA                  INSERT            TROEC
  SOBCXYZ SCHEMA1      AGENTES_CARGA                  SELECT            TROEC
  YASU999 ASDF123      BS_ADICOES_SISCOMEX            UPDATE            TRBSOEC
  USER123 WAREHOU      BS_ADICOES_TRIBUTOS_BKP        DELETE            TRBSOEC
  USER123 WAREHOU      BS_ADICOES_TRIBUTOS_BKP        INSERT            TRBSOEC

我想将行与字段 1、2、3 进行比较，如果它们相同，则捕获字段 4 进行显示，如下所示 -

grant DELETE,INSERT,SELECT on TRBSOEC.ADICOES to TRSSOEC;
grant INSERT,SELECT on SCHEMA1.AGENTES_CARGA to SOBCXYZ;
grant UPDATE on ASDF123.BS_ADICOES_SISCOMEX to YASU999;
grant DELETE,INSERT on WAREHOU.BS_ADICOES_TRIBUTOS_BKP to USER123;

基本逻辑是根据字段 1、2、3 进行分组，并打印不同的字段 4 以及其他公共字段。 授予 已提交 2.field3 到 field1 ；

到目前为止，我在下面尝试过，但不确定如何捕获字段 4，任何指导都会有帮助。

awk '
    # Each line
    {
        if ($1 == prev[1] && $2 == prev[2] && $3 == prev[3])
        {
            # Capture and use field 4 of last match
            if(!nrMatched) { nrMatched = prev[4] }
            print "grant " $nrMatched " on " $2"."$3 " to " $1";"
        }
        else
        {
            # No match, reset matched flag and just copy record through
            nrMatched = 0
        }
    
        # Save fields from this line for next comparison
        split($0, prev)
    }
' test.txt 

Answer 1

我会做什么：

awk '
{
    key = $1"|"$2"|"$3
    if (!(key in ops)) {
        groups[key]=$1","$2","$3
    }
    if (!seen[key"|"$4]) {
        ops[key]=ops[key](ops[key] ? "," : "") $4
        seen[key"|"$4]=1
    }
}
END {
    for (key in groups) {
        split(groups[key], fields, ",")
        print "grant " ops[key] " on " fields[2] "." fields[3] " to " fields[1]";"
    }
}' file