Question

I have a xml which has several numbers, something like this:

<xml>
<case1>
    <amount>3.140000</amount>
    <amount2>3.0000</amount2>
    <detail>
        <name>something</name>
        <earning>3.1000</earning>
    </detail>
</case1>
<case2>
    <amount>3.10000</amount>
</case2>

The xml is in a single line, and I’m testing with a regex like this:

echo "433.000" | sed -e 's/\([0-9]*\.\)\([1-9]*\)\(0*\)/\1\2/g'

But it gives me something like 3. for 3.00. I’m not trying to parse anything, just remove the zeroes. And somehow, it doesn’t do anything when I try with:

sed -e 's/\([0-9]*\.\)\([1-9]*\)\(0*\)/\1\2/g' xmlFile.xml

I have to mention that I’m not good with regular expressions, and I’m on a hosted server, so I can’t install any library or program, so only shell is allowed.

Thanks.

Answer 1

The following should work:

sed -e 's/\(\.[1-9][1-9]*\)0*/\1/g' -e 's/\.00*//' xmlFile.xml

This will result in the following:

<xml>
<case1>
    <amount>3.14</amount>
    <amount2>3</amount2>
    <detail>
        <name>something</name>
        <earning>3.1</earning>
    </detail>
</case1>
<case2>
    <amount>3.1</amount>
</case2>