Stupid, Stupid Regex...

[jeepinjeff 4]

I humbly come to you all asking for your assistance with this stinking regex (which is what I THINK I need to use.)

 

I have this content that I scrape:

 

We discover apples most below and XfgtsX new information and trot yesterday to wolves.

Scouring the XgyrtX world for tasty and dog, we XiretX give fellow.

This scurry we floor - come take one XhhueoX with us.

 

 

 

I have a bit of script that will capture, save and replace the text UP TO the first 'X'(We discover apples most below and ) with ABCDEF.

 

        set(#aTemp, $find regular expression(#content, "^([\\s\\S]*?)(?=X)"), "Global")
        add item to list(%aContent, #aTemp, "Don\'t Delete", "Global")
        set(#content, $replace regular expression(#content, "^([\\s\\S]*?)(?=X)", "ABCDEF"), "Global")
        set(#aCount, 1, "Global")
        set(#count, 2, "Global")

 

The next thing I need to do is repeat it with everything BETWEEN the next 2 'X' and the next 2 and so on. I would end up with, in the sample above:

 

ABCDEFXfgtsXABCDEFXgyrtXABCDEFXiretXABCDEFXhhueoXABCDEF

 

(Later I'll be replacing the ABCDEF with a currently unknown value and the X's with curly brackets)

 

The regex that I have been trying to use is (?:[\s\S]*?X+){#count}([\s\S]*?)(?=X) where #count = the nth occurrence of X.

 

Regardless of how I try to escape characters, I simply can't get it to work properly - despite it doing exactly what I want it to at regex101.com

 

Here's the current code

 

loop while($comparison(#aTemp, "!=", $nothing)) {
    set(#aTemp, $find regular expression(#content, "(?:[\\s\\S]*?X+)\{{#count}\}([\\s\\S]*?)(?=X)"), "Global")
    add item to list(%aContent, #aTemp, "Don\'t Delete", "Global")
    set(#content, $replace regular expression(#content, "(?:[\\s\\S]*?X+)\{{#count}\}([\\s\\S]*?)(?=X)", "ABCDEF"), "Global")
    set(#aCount, $add(#aCount, 1), "Global")
    set(#count, $add(#count, 2), "Global")
}

 

PLEASE help!!

 

Thanks!

Edited February 11, 2016 by jeepinjeff

[Code Docta (Nick C.) 638]

Hi,

 

This is one approach...

 

This is not a solution for all cases but will head u in the right direction.

 

set(#text,"We discover apples most below and XfgtsX new information and trot yesterday to wolves.

Scouring the XgyrtX world for tasty and dog, we XiretX give fellow.

This scurry we floor - come take one XhhueoX with us.","Global")
clear list(%split)
add list to list(%split,$list from text($find regular expression(#text,"(X.*?X)"),$new line),"Delete","Global")
clear list(%replsce X)
add list to list(%replsce X,$list from text($replace(%split,"X",$nothing),$new line),"Delete","Global")
clear list(%add curlies)
set(#OPEN CURLY,"\{","Global")
set(#CLOSED CURLY,"\}","Global")
loop($list total(%replsce X)) {
    add item to list(%add curlies,"ABCDEF{#OPEN CURLY}{$next list item(%replsce X)}{#CLOSED CURLY}","Don\'t Delete","Global")
}
add item to list(%add curlies,"ABCDEF","Don\'t Delete","Global")
set(#string,$text from list(%add curlies,""),"Global")
alert($text from list(%add curlies,""))

 

should give you more ideas

 

Regex is bad ass but not the solution for everything

ubot Regex is one line at a time and .Net style

 

I use this when it gets tough for me.

 

http://regexhero.net/tester/

 

HTHelps,

 

CD

remove XdfsX.ubot

[jeepinjeff 4]

Nick,

 

Thanks, man. Being new to this stuff, it took me a bit to navigate my way through your suggestion.  One thing that I forgot to explain in the original post is that the values that AREN'T X...X (EX - "We discover apples most below and_" OR "_world for tasty and dog, we_"need to be saved so they can be re-associated with ABCDEF, in order, at a later time.  So, I'm working on using your script and after extracting the X...X values, replacing them with a pipe and then adding the content to a list, delimited by the pipe.

 

For my first programming/coding/Ubot project, I may have bitten off more than I can chew - I'm already dealing with php and SQL scripts as well as this "F-word-from-a-sweet-old-lady"-inducing regex.  It seems SO useful, but so insanely complicated and fickle.

 

Something I don't get is how I can run the same expression on different .NET regex testers and end up with different results.  As my father would have said "Enough to piss off the Pope!"

 

Thanks again!

[Code Docta (Nick C.) 638]

Maybe try splitting on X

 

set(#text,"We discover apples most below and XfgtsX new information and trot yesterday to wolves.

Scouring the XgyrtX world for tasty and dog, we XiretX give fellow.

This scurry we floor - come take one XhhueoX with us.","Global")
clear list(%cc)
add list to list(%cc,$list from text(#text,"X"),"Delete","Global")
alert($list item(%cc,0))
alert($list item(%cc,1))

 

and pick out what u want with list item

without knowing the whole story it is hard to understand what u really need

feel free to pm me

CD

[HelloInsomnia 1103]

I'm trying to read between the lines a little bit here so hopefully this isn't far off from what you want but I think what you want is to replace these occurrences of X (insert randomness here) X with spintax?
 
So the output you may want would look like this:

We discover apples most below and {forge|discover|embrace} new information and trot yesterday to wolves.

Scouring the {cruel|nasty|beautiful|amazing} world for tasty and dog, we {humbly|may} give fellow.

This scurry we floor - come take one {sip|last drink|bite} with us.

Let me know if I'm on the right track here because if so I can probably help. Ignore the colorization of the words above it's just because I wrapped code tags around it.