Last modified: 2013-07-10 01:03:33 UTC
See the parserTests added in gerrit change I194167f3c0621394ab4fb40e242711ccb86615ae Round-trip of <b>x<b>y</b>z</b> is tricky --- we need to mark the fact that the <b> tag comes from html, not from single-quote wikitext.
That is what we are doing already with stx:html, which is why this works: echo '<b>x<b>y</b>z</b>' | node parse --wt2wt <b>x<b>y</b>z</b>