Styjun

I have problem to find specific named xml tags within a text block which may or may not be empty. I am not sure if the problem is after all resolvable with regex.

For example I have the following text block:

...<item name="xxx">yyyy</item>....

To filter out the 'item' tag with its attributes I a use the following regex:

<item(.*?)>(.*?)</item>|<item(.*?)./>

This regex also works with empty tags like in the following example text:

...<item name="zzz" />....

My problem is that my text contains multiple 'item' tags. This works as long as I do not mix the empty with not-empty tags like in the following example textblock:

...<item name="zzz" />....
...<item name="xxx">yyyy</item>....

The regex I use did not work with this siutation as it finds the first result:

<item name="zzz" />.......<item name="xxx">yyyy</item>

My question is: Is this problem solveable with regex?
My first thougt was to use kind of "AND NOT" combination like

(<item(.*?))(?!/>)>(.*?)</item>

But I failed at the problem.

My Java Code to apply the regex to a given textblock looks like this:

Pattern pattern = Pattern.compile("<item(.*?)>(.*?)</item>|<item(.*?)./>", Pattern.DOTALL);
Matcher matcher = pattern.matcher(textblock);
while (matcher.find()) 
 String attributes = matcher.group(1);
 ....

Your regex will work if using tag attributes regex as defined by the w3c

you can get 100% accuracy in finding open and self-contained tags.

But, you need an engine that uses atomic groups you can also differentiate

between open and self contained as below:

Raw:

(?s)(?:(<item(?>b(?:".*?"|'.*?'|[^>]*?)*>)(?<=/>))|(<item(?>b(?:".*?"|'.*?'|[^>]*?)*>)(?<!/>))(.*?)(</items*>))

Double quoted:

"(?s)(?:(<item(?>\b(?:".*?"|'.*?'|[^>]*?)*>)(?<=/>))|(<item(?>\b(?:".*?"|'.*?'|[^>]*?)*>)(?<!/>))(.*?)(</item\s*>))"

https://regex101.com/r/R3iVVW/1

 (?s)
 (?:
 ( # (1 start), Self Contained item
 <item
 (?> # Atomic to trap self contained
 b 
 (?: " .*? " | ' .*? ' | [^>]*? )*
 > 
 )
 (?<= /> ) # Must be a self contained item

 ) # (1 end)

 | # OR, 

 ( # (2 start), Open item
 <item
 (?> # Atomic to trap self contained
 b 
 (?: " .*? " | ' .*? ' | [^>]*? )*
 >
 )
 (?<! /> ) # Not a self contained item
 ) # (2 end)
 ( .*? ) # (3), Item content
 ( </item s* > ) # (4), Close item
 )

Benchmarked:

Regex1: (?s)(?:(<item(?>b(?:".*?"|'.*?'|[^>]*?)*>)(?<=/>))|(<item(?>b(?:".*?"|'.*?'|[^>]*?)*>)(?<!/>))(.*?)(</items*>))
Options: < none >
Completed iterations: 50 / 50 ( x 1000 )
Matches found per iteration: 14
Elapsed Time: 4.73 s, 4728.21 ms, 4728214 µs
Matches per sec: 148,047