Wrong question. The question is whether you should block it absolutely-- in config or htaccess-- because Preview, Translate and so on aren't bound by robots.txt.
Are you asking whether google can find the file, or whether they can read it? They can definitely find it; quick detour to raw logs confirms that they occasionally ask for .js files. Presumably for indexing purposes, which means it's time to whip out the
:: shuffling papers ::
<FilesMatch "\.(js|txt|xml)$"> Header set X-Robots-Tag "noindex" </FilesMatch>
You mean, where in htaccess does it go? Doesn't matter. Ordering in htaccess only makes a difference within individual modules. This one's a free-standing directive. Unless you've got something else in htaccess that says "yes! index all you like! override all other instructions!" and it overlaps the js/txt/xml group. This does not seem likely.
They probably use js when constructing Previews. But I can't personally check, because my js tend to involve fonts-- or piwik tracking-- so the default version is the same as the <noscript> version.