consider adding templating

[gkatsev]

Seems like there is a lot of string manipulation going on in the creation of the controls. Would be nicer if we were to use a templating system.
Even one as simple as:

function template(str, arrSubs){
  var i, l;
  i = 0;
  l = arrSubs.length;
  for (; i < l; i++) {
    str = str.replace("{"+i+"}", arrSubs[i]);
  }
  return str;
}
template('foo {0} {1} {2}', [1,2,3]);
// "foo 1 2 3"

For example, we currently have the same string in the content and updateContent of the durationDisplay.
Instead, each component, could have a template associated with it that can be changed.

[heff]

Yeah, the current way definitely isn't the most optimal. In this specific case it might be cleaner to just build a span for the time specifically and reference the innerHTML of that instead of overwriting the whole content el.

On Mar 4, 2013, at 3:34 PM, Gary Katsevman notifications@github.com wrote:

Seems like there is a lot of string manipulation going on in the creation of the controls. Would be nicer if we were to use a templating system.
Even one as simple as:

function template(str, arrSubs){
var i, l;
i = 0;
l = arrSubs.length;
for (; i < l; i++) {
str = str.replace("{"+i+"}", arrSubs[i]);
}
return str;
}
template('foo {0} {1} {2}', [1,2,3]);
// "foo 1 2 3"
For example, we currently have the same string in the content and updateContent of the durationDisplay.
Instead, each component, could have a template associated with it that can be changed.

—
Reply to this email directly or view it on GitHub.

[heff]

To clarify this more, every component has a contentEl property that returns the el() by default, but you can have them be different, for instance if you want children to be embedded in a sub-element of the component's main element.

In this specific case we could set up the span as a contentEl, so we don't have to create a new element each time.

But there still use for templates other places. If we can decide on how to pull in lodash we could make use of those templates.

This could probably be broken up into two feature issue, but confirming this either way.

[mmcc]

Just to revisit this one, if we're going to use an ES6 transpiler (#1812), this is somewhat moot. Even something as simple as string substition would improve things.

[mmcc]

I think we can consider this closed now that we have things like #2015 pulled in.

[gkatsev]

There is probably some extra stuff we can do to increase DRY, but yeah, we did get a lot of stuff for this via template strings.