Look ahead with any character (.+)

[Bot-Factory 602]

Hello.

 

I want to scrape stuff that resides between two tags (all in one row!):

<TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!"§!"3.<TAG2><TAG1>asduha9h9h9h98h<TAG2><TAG1>asd24r32asd.<TAG2><TAG1>asf54z45t35"§$"§$<ypja898H<TAG2><TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!"§!"3.<TAG2><TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!"§!"3.<TAG2>

 

 

But the stuff in between those tags could be anything. 

 

So when I try something like:

(?<=\<TAG1\>).+(?=\<TAG2\>)

 

It's of course not going to work because .+ also includes all the <TAG2> in between. It just removes the one at the end.

 

Questions:

1. Can this be done in a single expression? The look ahead should stop when it finds the first match of <TAG1>

2. Is it possible to return just a specific position? Let's say, return only what is between the second <TAG1>****<TAG2> match, but ignore all the others?

Like: Start at the second <TAG1> you find and stop at the second <TAG2> you find.

 

 

Is stuff like this possible? 

 

 

Thanks in advance for your help

Dan

[Kreatus (Ubot Ninja) 422]

I suck at explaining things but this may help you out http://stackoverflow.com/questions/2503413/regular-expression-to-stop-at-first-match

 

Here's the code for getting the first instance of that code.

set(#var, "<TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!\"§!\"3.<TAG2><TAG1>asduha9h9h9h98h<TAG2><TAG1>asd24r32asd.<TAG2><TAG1>asf54z45t35\"§$\"§$<ypja898H<TAG2><TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!\"§!\"3.<TAG2><TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!\"§!\"3.<TAG2>", "Global")
set(#firstTag, $list item($find regular expression(#var, "(?<=\\<TAG1\\>).*?(?=\\<TAG2\\>)"), 0), "Global")
alert(#firstTag)

[UBotDev 276]

The easiest way for 2. is to just add all matches to UBot %list and get the offset you want with "$list item".

[Bot-Factory 602]

 

I suck at explaining things but this may help you out http://stackoverflow.com/questions/2503413/regular-expression-to-stop-at-first-match

 

Here's the code for getting the first instance of that code.

set(#var, "<TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!\"§!\"3.<TAG2><TAG1>asduha9h9h9h98h<TAG2><TAG1>asd24r32asd.<TAG2><TAG1>asf54z45t35\"§$\"§$<ypja898H<TAG2><TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!\"§!\"3.<TAG2><TAG1>asidug87G(&r67%$<>/ <&547gasdg8g!\"§!\"3.<TAG2>", "Global")
set(#firstTag, $list item($find regular expression(#var, "(?<=\\<TAG1\\>).*?(?=\\<TAG2\\>)"), 0), "Global")
alert(#firstTag)

Thanks a lot. 

Greedy non greedy... Ok that's something I didn't know. Still don't understand it completely.. But I'm getting closer :-)

 

Dan

[UBotDev 276]

Thanks a lot. 

Greedy non greedy... Ok that's something I didn't know. Still don't understand it completely.. But I'm getting closer :-)

 

Dan

The simple way to explain greedy is to say that it will try to match as much as possible, so instead of ending match at first <TAG2> occurance, it will end the match at the last one (since it's greedy and hungry for more ).

[Bot-Factory 602]

The simple way to explain greedy is to say that it will try to match as much as possible, so instead of ending match at first <TAG2> occurance, it will end the match at the last one (since it's greedy and hungry for more ).

But that always requires a look ahead or look behind right? Without that it wouldn't know where to stop?

 

Dan

[UBotDev 276]

But that always requires a look ahead or look behind right? Without that it wouldn't know where to stop?

 

Dan

No it doesn't always require it, since this regex also works:

\<TAG1\>(.*?)\<TAG2\>

The difference is that if you use look ahead/behind, the string that you use there won't be included as part of the match, it's only there as a rule (example above will contain <TAG1> and <TAG2> as part of the match, but with look ahead/behind it will only contain content between them - the example that Kreatues posted above).

[Bot-Factory 602]

Awesome guys!! Thanks a lot. That was really helpful!

 

So besides the regular match types, look ahead and behind and greedy.... What would you say are the next three most important regex options you use?

Something that everyone should have in their toolbox?

 

Dan

[UBotDev 276]

Awesome guys!! Thanks a lot. That was really helpful!

 

So besides the regular match types, look ahead and behind and greedy.... What would you say are the next three most important regex options you use?

Something that everyone should have in their toolbox?

 

Dan

I think those are the the most important ones, since they "simulate" scraping when you are extracting data from string that's already in memory.

 

I actually think that everyone who is in scraping seriously should know regular expression, since it's making job easier or even possible sometimes, when nothing else will work.