sed - Manipulate Phone Number in xml

Question

I'm trying to manipulate a xml file with a bash script and sed and can't get it. The structure in the xml looks like

<Name>title firstname lastname</Name><Home>+49 (30) 1234 94</Home><Mobile>+49 (171) 1234 94</Mobile>
<Name>title firstname lastname</Name><Home>+49 (30) 1234 94</Home><Mobile>+49 (171) 1234 94</Mobile>

I need to eliminate the space and ( and ) ONLY in the phone number. After a day whis regex and sed, I could not get it. I have the string that match as needed but I can't get the groups and the right substitution.

sed -e 's/([0-9]\s|[0-9]\s\([0-9]|[0-9]\)\s[0-9]|[0-9]\s[0-9])/gm'

Answer 1

Don't use sed to manipulate XML documents.
There are good tools for that activity.
xmlstarlet is one of them.

A valid XML structure requires a root element at the top of the node tree.
Let's say we have an XML fragment ( test.xml ):

<root>
    <Name>title firstname lastname</Name>
    <Home>+49 (30) 1234 94</Home>
    <Mobile>+49 (171) 1234 94</Mobile>
    <Name>title firstname lastname</Name>
    <Home>+49 (30) 1234 94</Home>
    <Mobile>+49 (171) 1234 94</Mobile>
</root>

The command :

xmlstarlet ed -u "//Home|//Mobile" -x "translate(normalize-space(.),'() ','')" test.xml

Details :

ed - enables edit mode

-u - to update xml structure

"//Home|//Mobile" - xpath expression to select the needed elements

-x - to update the needed values with xpath expression

. (period) - points to the current selected node(s)

normalize-space() - the function which returns the argument string with whitespace normalized by stripping leading and trailing whitespace and replacing sequences of whitespace characters by a single space

translate(string, string, string) - the function which returns the first argument string with occurrences of characters in the second argument string replaced by the character at the corresponding position in the third argument string.

---

The output:

<?xml version="1.0"?>
<root>
  <Name>title firstname lastname</Name>
  <Home>+4930123494</Home>
  <Mobile>+49171123494</Mobile>
  <Name>title firstname lastname</Name>
  <Home>+4930123494</Home>
  <Mobile>+49171123494</Mobile>
</root>

Answer 2

Assuming the format of the numbers remains same:

sed -r 's/(\+[0-9]{2}) \(([0-9]{2,3})\) ([0-9]{4}) ([0-9]{2})/\1\2\3\4/g' input
<Name>title firstname lastname</Name><Home>+4930123494</Home><Mobile>+49171123494</Mobile>
<Name>title firstname lastname</Name><Home>+4930123494</Home><Mobile>+49171123494</Mobile>

Answer 3

<Name>title firstname lastname</Name><Home>0049 (30) 1234567 94</Home><Mobile>+491711234577 0</Mobile>
<Name>title firstname lastname</Name><Home>+1 39976 1234 94</Home><Mobile>+49 (171) 1234 94</Mobile>    

sed -r 's/(\+|\d*?) ([0-9]{1,})|\s\(([0-9]{2,})\) ([0-9]{2,}) ([0-9]{1,4})/\1\2\3\4\5/g' input (see above)

<Name>title firstname lastname</Name><Home>004930123456794</Home><Mobile>+4917112345770</Mobile>
<Name>title firstname lastname</Name><Home>+139976123494</Home><Mobile>+49171123494</Mobile>