This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
It seems to me that a start tag for CAPTION, TBODY, THEAD, TFOOT, TR, TD, TH, TABLE in 'in select in table' causes an infinite loop in the fragment case when there's no <select> in table scope, because the "as if" </select> doesn't change any state before the token is reprocessed. Maybe it's impossible to reach this state in the fragment mode. If that's the case, the assertion about fragment mode in the spec should be replaced with a "this can't happen" assertion.
Sorry about the bad bug report. I inlined the bit from "An end tag whose tag name is "select"" from "in select" into the "Act as if an end tag with the tag name "select" had been seen" part of "in select in table". This resulted in a non-sensical "(fragment case)" assertion, but as a result of inlining, it isn't really a spec bug. It would be nice to mention that it's impossible to reach "in select in table" without having a "select" element in table scope, though. (If that's true!)
Am I correct in assuming that > It would be nice to mention that it's impossible to reach "in select in table" > without having a "select" element in table scope, though. (If that's true!) ...is the only request here? If so it would be helpful to understand where such a mention should be placed.
Marking this NEEDSINFO since Ian asked for info. Henri, I hope you're ok with this lacking the formal response template.