0

Ubuntu 16.04

I have a xml file that has about 50 lines of code. There is a certain line that contains the string "FFFedrs". ON this line is the word true.

If the structure was neat with only 1 space between like so .. <property name="FFFedrs" value="true"/> <!-- Enables/Disables EasyMoney -->

I could use a sed in place command like this:

$ cat file.xml

<property name="FFFedrs" value="true"/> <!-- Enables/Disables EasyMoney -->

$ sed -i 's/<property\ name=\"FFFedrs\"\ value=\"true\"\/>\ <!--\ Enables\/Disables\ EasyMoney\ -->/<property\ name=\"FFFedrs\"\ value=\"false\"\/>\ <!--\ Enables\/Disables\ EasyMoney\ -->/g' file.xml
$
$ cat file.xml

<property name="FFFedrs" value="false"/> <!-- Enables/Disables EasyMoney -->

But the file is not neatly formatted so the line that has the string "FFFedrs" looks something like ...

 <property name="FFFedrs"              value="true"/>       <!-- Enables/Disables EasyMoney -->

How do I sed the true to false on the line that has the string "FFFedrs"

Cyrus
  • 84,225
  • 14
  • 89
  • 153
Vituvo
  • 1,008
  • 1
  • 9
  • 29
  • Don't Parse [X]HTML With Regex: https://stackoverflow.com/a/1732454/3776858. I suggest to use an XML/HTML parser (xmlstarlet, xmllint ...). – Cyrus Jun 10 '18 at 07:51

2 Answers2

1

Just add \+ after your spaces:

sed -i 's/<property\ \+name=\"FFFedrs\"\ \+value=\"true\"\/>/<property\ name=\"FFFedrs\"\ value=\"false\"\/>/g' file.xml

This will replaces multiple spaces with a single space.

If you want to preserve the spaces:

sed -i 's/<property\(\ \+\)name=\"FFFedrs\"\(\ \+\)value=\"true\"\/>/<property\1name=\"FFFedrs\"\2value=\"false\"\/>/g' file.xml
Andriy Makukha
  • 7,580
  • 1
  • 38
  • 49
  • This is just what I needed. Hats off to you mate. – Vituvo Jun 10 '18 at 11:13
  • @needtoknow, it might be because you have tab characters in there. Try to replace space character `\ ` with `[[:space:]]` in your regex, like this: `sed -i 's///g' file.xml` – Andriy Makukha Jun 10 '18 at 13:17
0

This should do the trick

sed -i 's|\(<.*FFFedrs"\)\s*\(value.*>\)\s*\(<.*>\)|\1 \2 \3|' file

I use a pipe in sed as a separator because your line has slashes in. 3 groups of capture for the main values and output as wanted

You can use https://regex101.com/ to learn about regex and test it live.

David Peltier
  • 91
  • 6