// Two new lines before text content followed by three new lines and one explicit new line with <br>

var input = '\r\n<div>\r\ntext\r\nwith newlines\r\n<br>\r\n</div>\r\n';

console.log("String to parse: '" + input + "'");

​

// Current xlsx impl

var output = input

  // Remove new lines and spaces from start of content

  .replace(/^[\t\n\r ]+/, "")

  // Remove new lines and spaces from end of content

  .replace(/[\t\n\r ]+$/,"")

  // Replace remaining new lines and spaces with space

  .replace(/[\t\n\r ]+/g, " ")

  // Replace <br> tags with new lines

  .replace(/<\s*[bB][rR]\s*\/?>/g,"\n")

  // Strip HTML elements

  .replace(/<[^>]*>/g,"");

​

// This will result in ' text with newlines \n '

console.log("xlsx implementation output: '" + output + "'");

​

// modified impl

var outputMod = input

  .replace(/^[\t\n\r ]+/, "")

  .replace(/[\t\n\r ]+$/,"")

  // Added line which removes any white space characters after and before html tags

  .replace(/>\s+/g,">").replace(/\s+</g,"<")

  .replace(/[\t\n\r ]+/g, " ")

  .replace(/<\s*[bB][rR]\s*\/?>/g,"\n")

  .replace(/<[^>]*>/g,"");

// This will result in 'text with newlines\n', which seems more appropriate

console.log("Modified implementation output: '" + outputMod + "'");