Question

我是awk的新手并尝试将ldif文件转换为csv，因为我在这里使用awk来解析和转换它。如果属性值中没有任何逗号，则下面的脚本工作正常。但是如果属性值中有逗号，则在csv中它创建一个单独的列。如果它包含使用awk的逗号，则需要帮助操作属性值字段。下面是我正在使用的awk文件和示例数据。 AWK脚本： -

#!/bin/awk -f
BEGIN {
        uid = ""
        omEntitytype = ""
        sn = ""
        givenName = ""
        initials = ""
        omUnit = ""
        departmentNumber = ""
        omCostCenter = ""
        title = ""
        omManager = ""
        omaffiliatedaccount = ""
        printf("USERID,USER_TYPE,USER_LASTNAME,USER_FIRSTNAME,USER_INITIAL,USER_UNIT,USER_DEPT,CHARGE_UNIT,USER_JOB_TITLE,SPONSOR_USERID,ACCOUNT_ID\n");
}
/^uid:/              {uid=$2}
/^omEntitytype:/       {omEntitytype=$2}
/^sn:/              {sn=$2}
/^givenName:/          {givenName=$2}
/^initials:/               {initials=$2}
/^omUnit:/              {omUnit=$2}
/^departmentNumber:/      {departmentNumber=$2}
/^omCostCenter:/      {omCostCenter=$2}
/^title:/ {title=$2}
/^omManager:/            {omManager=$2}
/^omaffiliatedaccount:/          {omaffiliatedaccount=$2}
/^dn/{
        if(uid != "") printf("%s,%s,%s,,%s,%s,%s,%s,%s,%s,%s,%s\n",uid,omEntitytype,sn,givenName,initials,omUnit,departmentNumber,omCostCenter,title,omManager,omaffiliatedaccount);
        uid = ""
        omEntitytype = ""
        sn = ""
        givenName = ""
        initials = ""
        omUnit = ""
        departmentNumber = ""
        omCostCenter = ""
        title = ""
        omManager = ""
        omaffiliatedaccount = ""
}
#capture last dn
END{
       if(uid != "") printf("%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s\n",uid,omEntitytype,sn,givenName,initials,omUnit,departmentNumber,omCostCenter,title,omManager,omaffiliatedaccount)
}

示例ldif要传递给它的数据

dn: uid=sample1,o=people,dc=om,DC=COM
uid: sample1
omEntitytype: Contingent, off-Site
sn: sample1 name1
givenName: sample1
initials: P
omUnit: 07
departmentNumber: 123
omCostCenter: 10
title: Analyst
omaffiliatedaccount: 12345

dn: uid=sample2,o=people,dc=om,DC=COM
uid: sample2
omEntitytype: Contingent, On-Site
sn: sample2 name2
givenName: sample2
initials: P
omUnit: 07
departmentNumber: 123
omCostCenter: 10
title: PLAT MGR, ENGINE,HYD,ELECT,DRIVES
omaffiliatedaccount: 12345

这里的示例数据中，我有问题（填写到csv中的新列）的字段omEntitytype和title，因为它们在值中有'，'

任何帮助都表示赞赏..谢谢。

Answer 1

请原谅我的无知，但你不能只使用不同的角色（例如'/'）来分隔多个标题吗？

Answer 2

每当你有名字时 - ＆gt;数据中的值对最好先构建一个类似的映射数组，然后按名称打印值：

$ cat tst.awk
BEGIN {
    OFS = ","
    numFlds = split("uid omEntitytype sn givenName initials omUnit departmentNumber omCostCenter title omManager omaffiliatedaccount",flds)
    print "USERID,USER_TYPE,USER_LASTNAME,USER_FIRSTNAME,USER_INITIAL,USER_UNIT,USER_DEPT,CHARGE_UNIT,USER_JOB_TITLE,SPONSOR_USERID,ACCOUNT_ID"
}
{
    name = value = $0
    sub(/:.*/,"",name)
    sub(/[^:]*:[[:space:]]*/,"",value)
    name2value[name] = value
}
!NF { prtRec() }
END { prtRec() }

function prtRec() {
    for (i=1; i<=numFlds; i++) {
        printf "\"%s\"%s", name2value[flds[i]], (i<numFlds?OFS:ORS)
    }
    delete name2value
}

$ awk -f tst.awk file
USERID,USER_TYPE,USER_LASTNAME,USER_FIRSTNAME,USER_INITIAL,USER_UNIT,USER_DEPT,CHARGE_UNIT,USER_JOB_TITLE,SPONSOR_USERID,ACCOUNT_ID
"sample1","Contingent, off-Site","sample1 name1","sample1","P","07","123","10","Analyst","","12345"
"sample2","Contingent, On-Site","sample2 name2","sample2","P","07","123","10","PLAT MGR, ENGINE,HYD,ELECT,DRIVES","","12345"

您实际上没有告诉我们您希望如何处理逗号，上面引用了它们，但如果这不是您想要的，那么只需更改prtRec()即可执行您想要的操作，例如：也许gsub(/,/,";",name2value[flds[i]])。

请注意，如果您的输入数据没有丢失某些您想要输出的字段和/或如果您只是希望字段按照它们出现在输入中的顺序输出，那么上面的内容会更加简单。

Answer 3

您可以检查＆＃34;，＆＃34;存在于该字段中，您可以将其更改为另一个字符或引用该字段（这是标准的）

这是后一种情况的原型

echo "a:b,c:d" | 
awk -F: -v q='"' -v OFS=, '{for(i=1;i<=NF;i++)
                                 if(match($i,",")) $i=q $i q}1'

a,"b,c",d

Answer 4

我会这样做（需要GNU awk）

generate LDIF | gawk '
    function quote(str) {
        if (str ~ /,/) str = "\"" str "\""
        return str
    }
    function print_record() { 
        if (data["uid"] != "") 
            printf "%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s\n",
                quote(data["uid"]),
                quote(data["omEntitytype"]),
                quote(data["sn"]),
                quote(data["givenName"]),
                quote(data["initials"]),
                quote(data["omUnit"]),
                quote(data["departmentNumber"]),
                quote(data["omCostCenter"]),
                quote(data["title"]),
                quote(data["omManager"]),
                quote(data["omaffiliatedaccount"])
    }
    match($0, /^([^:]+): (.*)/, m) {data[m[1]] = m[2]}
    NF == 0 {print_record(); delete data}
    END {print_record()}
'

根据您的样本输入

输出

sample1,"Contingent, off-Site",sample1 name1,sample1,P,07,123,10,Analyst,,12345
sample2,"Contingent, On-Site",sample2 name2,sample2,P,07,123,10,"PLAT MGR, ENGINE,HYD,ELECT,DRIVES",,12345

写入csv文件时忽略属性值中的逗号

4 个答案: