matches

<template>
  Everything inside of here would be tag_contents
  
  <template> <-- this is where the parser fails with a syntax error 
    everything inside of here would still be tag_contents
  </template>
</template>

<script>
  everything in here would be tag_contents
</script>

spt_lexer.xrl

Definitions.

% Helper Regexes (non-token)

ValidInTagName = [A-Za-z0-9]
ValidTagAttributeName = [A-Za-z\-]
ValidTagAttributeValue = [A-Za-z0-9\s="\-/:;,\.?\(\)@]
ValidTagAttribute = {ValidTagAttributeName}+=("|')?{ValidTagAttributeValue}+("|')?
ValidTags = (\s{ValidTagAttribute})*
ValidInsideTag = (.)

% Core HTML tokens
ROOT_TAG  = \<root{ValidTags}\>
TEMPLATE_TAG  = \<template{ValidTags}\>
SCRIPT_TAG  = \<script{ValidTags}\>
START_TAG = \<{ValidInTagName}+{ValidTags}\>
END_TAG   = \<\/{ValidInTagName}+\>
CHARACTER = {ValidInsideTag}

% Useful tokens
NEW_LINE  = [\r\n]\s*
SELF_CLOSING_TAG = \<{ValidInTagName}+{ValidTags}\s\/\>

Rules.

{START_TAG} : {token, {start_tag, TokenLine, TokenChars}}.
{END_TAG}   : {token, {end_tag, TokenLine, TokenChars}}.
{CHARACTER} : {token, {char, TokenLine, TokenChars}}.

{NEW_LINE}  : skip_token.

Erlang code.

spt_parser.yrl

Terminals start_tag end_tag char.
Nonterminals tag tag_contents chars.
Rootsymbol tag.

tag -> start_tag end_tag : {unwrap('$1'), []}.
tag -> start_tag tag_contents end_tag : {unwrap('$1'), '$2'}.

tag_contents -> chars : ['$1'].

chars -> char chars : unicode:characters_to_binary([unwrap('$1')|'$2']).
chars -> char : unwrap('$1').

Erlang code.

unwrap({_,_,V}) -> V.