如果我理解你的意图，可能比你想的要简单。
您可以通过将制表符替换为逗号来使用几乎“原样”的数据，* 假设您要添加一个WHERE子句 。显然， 永远不要 * 更新没有WHERE的表;我添加了一个简单的占位符。

$: cat file
employee='Scott'        job='Manager'   Salary=10000    location='Chicago'      deptno=10       hike_in_perc=15 deptname='Accounts'
employee='bob'  job='peon'      Salary=100      location='hell' deptno=666      hike_in_perc=1  deptname='drudgery'

$: while IFS=$'\t' read -r line; do echo "UPDATE sometable SET $line WHERE somecondition='$somevalue';";done < <( tr $'\t' , < file )
UPDATE sometable SET employee='Scott',job='Manager',Salary=10000,location='Chicago',deptno=10,hike_in_perc=15,deptname='Accounts' WHERE somecondition='foo';
UPDATE sometable SET employee='bob',job='peon',Salary=100,location='hell',deptno=666,hike_in_perc=1,deptname='drudgery' WHERE somecondition='foo';

$: while IFS=$'\t' read -r line; do echo "UPDATE sometable SET ${line//$'\t'/,} WHERE somecondition='$somevalue';";done<file
UPDATE sometable SET employee='Scott',job='Manager',Salary=10000,location='Chicago',deptno=10,hike_in_perc=15,deptname='Accounts' WHERE somecondition='bar';
UPDATE sometable SET employee='bob',job='peon',Salary=100,location='hell',deptno=666,hike_in_perc=1,deptname='drudgery' WHERE somecondition='bar';

例如，如果您的前两个字段是您的条件...

$: while IFS=$'\t' read -r a b new; do echo "UPDATE sometable SET ${new//$'\t'/,} WHERE $a AND $b;";done<file
UPDATE sometable SET Salary=10000,location='Chicago',deptno=10,hike_in_perc=15,deptname='Accounts' WHERE employee='Scott' AND job='Manager';
UPDATE sometable SET Salary=100,location='hell',deptno=666,hike_in_perc=1,deptname='drudgery' WHERE employee='bob' AND job='peon';

对于大型数据集，使用awk会更快。

$: awk -F$'\t' '{printf "UPDATE sometable SET "; for(i=3;i<NF;i++)printf "%s,",$i; print $NF " WHERE " $1 " AND " $2 ";"}' file
UPDATE sometable SET Salary=10000,location='Chicago',deptno=10,hike_in_perc=15,deptname='Accounts' WHERE employee='Scott' AND job='Manager';
UPDATE sometable SET Salary=100,location='hell',deptno=666,hike_in_perc=1,deptname='drudgery' WHERE employee='bob' AND job='peon';

这些做了很多假设，比如文件的每一行都是更新一条记录所需的所有数据-但我认为这是值得一提的，以防这更像是一个XY problem。
如果你的文件可以是多行，但每个文件都是一个可能的UPDATE，那么也许另一种方法-

$: cat f2
employee='Scott'    job='Manager'
Salary=10000    location='New York' deptno=10
hike_in_perc=15 deptname='Accounts'

$: conditions="$(<f2)"
$: echo "select count(*) as cnt from emp where ${conditions//[$'\n\t']/ AND };"
select count(*) as cnt from emp where employee='Scott' AND job='Manager' AND Salary=10000 AND location='New York' AND deptno=10 AND hike_in_perc=15 AND deptname='Accounts';

祝你好运

awk：

$ awk 'BEGIN{c=0}{for (i=1; i<=NF; i++) print "var"++c"="$i}' cols.txt
var1=employee='Scott'
var2=job='Manager'
var3=Salary=10000
var4=location='Chicago'
var5=deptno=10
var6=hike_in_perc=15
var7=deptname='Accounts'

下面的脚本将解析文件声明的变量。

for var in $(cat cols.txt);do declare "$var"; done

[~] echo $employee
'Scott'
[~] echo $job
'Manager'
[~] echo $Salary
10000
[~] echo $location
'Chicago'
[~] echo $deptno
10
[~] echo $hike_in_perc
15
[~] echo $deptname
'Accounts'

不要填充一堆后缀为1-N的标量变量，填充一个数组，例如。arr[1]="employee='Scott'"，关联数组可能最适合您的需要，arr['employee']='Scott'。在这个论坛和stackexchange上有很多答案，展示了如何做到这两点，但你的答案有点复杂，因为每行有多个制表符分隔的赋值，所以你可以这样做：

declare -A arr
while IFS='=' read -r tag val; do
    arr["$tag"]="$val"
done < <(tr $'\t' $'\n' < file)

结果是arr[]被填充了标签（名称）到值的Map：

declare -p arr
declare -A arr=([employee]="'Scott'" [job]="'Manager'" [deptno]="10" [location]="'Chicago'" [deptname]="'Accounts'" [hike_in_perc]="15" [Salary]="10000" )

上面假设输入中没有重复的标记，因为OP提供的示例中没有。