Parsing GMNotes from Token

I'm trying to parse the contents of a token's GM Notes field to find a particular line of text.      var gmNotes = npcToken.get("gmnotes");     var fullText = decodeURIComponent(gmNotes);    var allLines = fullText.split("<p>");     allLines.some(function(line){         if (line.startsWith("TargetText")){            //dostuff         }     }); This works, but it isn't sitting right with me. Splitting on the paragraph tag feels wrong, but I'm not sure of a more efficient/cleaner way to do it. If anyone has any ideas, I'd appreciate it. :)

Are you concerned that you might not catch all of the lines? If so, you could split on a regex of all of the style demarcations, capturing the type in groups: let allLines = []; fullText.replace(/(<(p|h1|h2|h3|h4|h5|h6)>(.*)<\/\2>)/gmi,(m,g1,g2,g3) => {     allLines.push({ type: g2, text: g3 });     return m; }); (air-coded, so normal testing caveats apply) Purists will dispute leveraging the replace operation this way (as a forEach that can quickly access the regex groups), so if that bothers you, you can rewrite it with as a split => forEach => regex exec => push, but this should work. In the end, you should have an array of tokens with a type property corresponding to the tag and a text property corresponding to the interior text. That the sort of thing you were looking for?

Alternatively, you might be able to use a regex that asserts the start of line but with the multi-line flag to quickly locate this one piece of text... let targetrx = /^<(p|h1|h2|h3|h4|h5|h6)>(Target Text.*)<\/\1>/mi; let theLineIWant; if(targetrx.test(fullText) theLineIWant = targetrx.exec(fullText)[2]; log(theLineIWant); You might have to escape the target text if it might have regex codes in it: const escapeRegExp = (string) => { return string.replace(/[.*+\-?^${}()|[\]\\]/g, '\\$&'); }; let escTgt = escapeRegExp(`Target(Text) with /*codes*/`); let targetrx = new RegExp(`^<(p|h1|h2|h3|h4|h5|h6)>(${escTgt}.*)<\\/\\1>`,'mi'); But you'll want to test that regex... I always have to double check escaping the slashes so that they show up in the string. =D

Thanks, Timmaugh. I like the second approach, but I'm not sure I understand exactly how the regex is working, so please bear with me. It looks like it's creating a pattern that should match any html tag, then the string I'm searching for. What does the <\/\1> bit do? Also, in the if statement, targetrx.exec(fullText)[2]; is executing the regex, and then element #0 will be the full string, element #1 will be the tag, and element 2 is the text that was being searched for?

Hey, Nick... yep, you're right about the groups and how they will be numbered. Group 0 is the full match, and group 1 is the tag type. The <\/\1> bit is using a way to refer to capture groups while still in the regex. The same as you can use $1 to refer to the first capture group in js in a replace operation, you can use \1 to refer to it in a regex. So if the regex is trigger on a <p> tag, the first group (not the zeroth group) will be 'p'. That means we're looking for </p> to close the capture group of the text we're looking for. </p> ...becomes... </\1> ...that needs to be escaped, so it becomes... <\/\1> Last... I'm sure you know about escaping the slashes in a string, so more for others who might be reading: if you have to use it as an actual string to supply a `new RegExp(...)` function, you have to escape those slashes again, since it will exist as a string before it is the source of the regex, and in that interregnum those regex-escaping slashes will become string-character-escaping slashes and disappear before it reaches the regex. That's where the extra slashes come in for this line: let targetrx = new RegExp(`^<(p|h1|h2|h3|h4|h5|h6)>(${escTgt}.*) <\\/\\1> `,'mi');

Thanks!