Bugfix for parsing xml that starts with whitespace

[rasmuserik]

Rhino has problems when parsing xml that starts with whitespace (and XML.ignoreWhitespace === true), ie:

js> (new XML(" <a/>")).toSource()
&lt;a/&gt;

This patch fixes this issues such that:

js> (new XML(" <a/>")).toSource()
<a/>

Unit test also added.

This is implemented by removing the special case for strings not starting with <. According to ECMA-357 10.3, numbers and booleans are passed to the same xml-conversion as strings, so as far as I can see the special case is not needed.

[rasmuserik]

trim leading/trailing space within xml-textnodes as in spidermonkey and ecma-357

Added another patch+unittest on whitespace and e4x: Rhino treats leading/trailing whitespaces in e4x differently from SpiderMonkey, with XML.ignoreWhitespace and XML.prettyPrinting. ie. Rhino does:

js> XML.ignoreWhitespace = true; XML.prettyPrinting = true; uneval(<a> b </a>)
<a>b</a>
js> XML.ignoreWhitespace = true; XML.prettyPrinting = false; uneval(<a> b </a>)
<a>b</a>
js> XML.ignoreWhitespace = false; XML.prettyPrinting = true; uneval(<a> b </a>)
<a> b </a>
js> XML.ignoreWhitespace = false; XML.prettyPrinting = false; uneval(<a> b </a>)
<a> b </a>

where the latest SpiderMonkey does:

js> XML.ignoreWhitespace = true; XML.prettyPrinting = true; uneval(<a> b </a>)
"<a>b</a>"
js> XML.ignoreWhitespace = true; XML.prettyPrinting = false; uneval(<a> b </a>)
"<a> b </a>"
js> XML.ignoreWhitespace = false; XML.prettyPrinting = true; uneval(<a> b </a>)
"<a>b</a>"
js> XML.ignoreWhitespace = false; XML.prettyPrinting = false; uneval(<a> b </a>)
"<a> b </a>"

The second commit changes Rhinos behaviour, such that leading/trailing spaces are not trimmed when building/trimming the xml-tree, but is trimmed when prettyprinting, - which seems to be same behaviour as SpiderMonkey and in the e4x-standard.

[mbd-dbc-dk]

Would be great to have this patch merged - we do a lot of E4X manipulations using Rhino, and could use this patch (disclaimer: from same org. as Rasmus).

[rasmuserik]

What is the status of this,
ie. what needs to be done
to get this one pulled?

[p-bakker]

Converted to draft as this has been lying around since forever and needs someone to go through to determine whether we want/needs this PR not