Question

I have a field with wiki style rendering on it that I’d like to bust up in Javascript.

The text I’m trying to parse looks like this:

{color:#47B}_name1_{color}
{color:#555}description1{color} 
---
{color:#47B}_name2_{color}
{color:#555}description2{color} 
---
{color:#47B}_name3_{color}
{color:#555}description3{color} 
---
etc

Where name1 and description1 belong together, name2 and description2 belong together, and so forth. The values for name and description are user supplied values, with description potentially spanning multiple lines.

My end goal is to be able to extract the values of each name and each description from the text (and be able to reliably associated name1 with description1, etc).

My question is: If I used a regex to match all the names into an array and all the descriptions into an array, can I be ensured that the items in the array are in the correct order? That is, will names[0] always be the first name in the parsed text (assuming I did a javascript regex match into the names array)? Also- is this bad practice/should I do this another way?

The regular expression I’m trying to use to match names is:

/^(\{color\:#47B\})(_)(\s*?)(.*?)(\s*?)(_)(\{color\})$/

And the regular expression I’m using to match descriptions is:

/(\{color\:#555\})(.*?)(\{color\})/

Answer 1

A regex search will always return matches in source order (i.e. in the order in which they occur in the source text.)

I assume you are asking this question because you’re hoping to do two regex matches (one for name, one for description) and then get two result arrays, and guarantee that namesmatch[i] always goes with descriptionmatch[i]. However, this will only be true if your source text is always exactly perfect.

In this case it may be better or safer either to use a single regex that matches both at once, or split your source up by those -- delimiters and then match within each block. The reason why it may be safer is that your source text may contain errors, and at least in this case you can detect that and have as much good data as possible.

A note about your regexes. The . does not match newlines, so if the text between your {color} braces might have a newline you need to include newlines explicitly. [\s\S] and [^] are common idioms for this. Alternatively, if all . in a regex should match newlines, set the dotAll flag (s).