Find text between key phrases - regex

Question

I have a var that have some text in:
<cfsavecontent variable="foo">
element.password_input=
<div class="holder">
<label for="$${input_id}" > $${label_text}</label>
<input name="$${input_name}" id="$${input_id}" value="$${input_value}" type="password" />
</div>
# END element.password_input
element.text_input=
<div class="ctrlHolder">
<label for="$${element_id}" > $${element_label_text}</label>
<input name="$${element_name}" id="$${element_id}"
value="$${element_value}" type="text"
class="textInput" />
</div>
# END element.text_input
</cfsavecontent>
and I am trying to parse through the var to get all of the different element type(s) here is what I have so far:
ar = REMatch( "element\.+(.*=)(.*?)*", foo )
but it is only giving me this part:
element.text_input=
element.password_input=
any help will be appreciated.

Answer 1

Your immediate problem is that by default . doesn't include newlines - you would need to use the flag (?s) in your regex for it to do this.
However, simply enabling that flag still wont result in your present regex doing what you're expecting it to do.
A better regex would be:
(element\.\w+)=(?:[^##]+|##(?! END \1))+(?=## END \1)
You would then do ListFirst(match[i],'=') and ListRest(match[i],'=') to get the name and value. (rematch doesn't return captured groups).
(Obviously the #s above are doubled to escape them for CF.)
The above regex dissected is:
(element\.\w+)=
Match element. and any alphanumeric, placed it into capture group 1, then match = character.
(?:
[^##]+
|
##(?! END \1)
)+
Match any number of non-hash characters, or a hash not followed by the ending token (using negative lookahead (?!...)) and referencing capture group 1 (\1), repeat as many times as possible (+), using a non-capturing group ((?:...)).
(?=## END \1)
Lookahead (?=...) to confirm the variable's ending token is present.