单词匹配时，Notepad ++ RegEx会在标签之间删除

Question

I had a similiar question that was used for numbers this time I need to use it for keyword. Below is the sample data that I'm using from a KML file. I would like to remove all placemarks that contain the word footway.

 <Placemark>
        <styleUrl>#nothing</styleUrl>
        <ExtendedData>
            <SchemaData>
                <SimpleData>highway</SimpleData>
            </SchemaData>
        </ExtendedData>
        <LineString>
            <coordinates>0.0000,0.0000,0</coordinates>
        </LineString>
    </Placemark>     
    <Placemark>
        <styleUrl>#nothing</styleUrl>
        <ExtendedData>
            <SchemaData>
                <SimpleData>footway</SimpleData>
            </SchemaData>
        </ExtendedData>
        <LineString>
            <coordinates>0.0000,0.0000,0</coordinates>
        </LineString>
    </Placemark>

I tried to use the following however it is capturing everything

(?i)<Placemark>.*?footway.*?</Placemark>

Below is my notepad++ settings

Find what: (?i)<Placemark>.*?footway.*?</Placemark>
Replace with:
Warp around
Search Mode: Regular expression & mathces newline

Answer 1

Here is a way to go:

• Find what: <Placemark>(?:(?!<Placemark).)*footway(?:.(?!<Placemark))*</Placemark>
• Replace with: NOTHING

This will replace all <Placemark> blocks that contain footway and only them.

(?!<Placemark) is a negative lookahead that assumes there're no <Placemark> before footway , so, when you have many <Placemark> 's the regex matches a single <Placemark> at a time.

(?:(?!<Placemark).)* is a non capture group, that occurs 0 or more times and does not contain (?!<Placemark) followed by a character.

Answer 2

This is working for me with Notepad++ 6.9.2. It also works in this online python regex tester: https://regex101.com/r/BYGvzo/1

Are you sure you have the correct options ( regular expression + . matches newline ) set?

EDIT: Well, after your edit that's a different story! Not sure how to achieve it with a regex. I think it would be way easier to parse the XML and then get rid of the nodes containing the word footway.

See why: RegEx match open tags except XHTML self-contained tags

Answer 3

Simplifying your file, it looks like the first line below and your regular expression is matching as per the second line

<Placemark> ... </Placemark> <Placemark> ...footway ... </Placemark>
<Placemark>    .*?                          footway .*? </Placemark>

Need to prevent the first </Placemark> being included in the match.

If this is a one-off or seldom needed process then an approach I sometime use as it is very adaptable is as follows. Find a single character that does not occur anywhere in the file. For this example = is used. Do a replace-all of the regular expression (</?p)(lacemark>) with \\1=\\2 . Leading to the text:

<P=lacemark> ... </P=lacemark> <P=lacemark> ...footway ... </P=lacemark>

Then do a replace-all with the regular expression <P=lacemark>[^=]*footway[^=]*</P=lacemark> with nothing. Finally, remove all the = characters with another replace-all.

If there is no easy to use single character (ie something instead of the = ) then precede the above steps with some replacements to create an unused character. For example first replace all & with & then replace all = with &eq; . Now the = is free for use. After the above steps, undo the replacements, first replace all &eq; with = then replace all & with & .