[20:04:57] <jdlrobson>	 tgr: nice write up. I think the thing that's gonna be a pain is looking for parentheticals across nodes e.g. <span>this is some text (hello</span><span><span> and goodbye)</span>
[20:05:04] <jdlrobson>	 especially given they can be nested :/
[20:05:39] <jdlrobson>	 i was trying to find some libraries that do this yesterday but so far no luck
[20:46:24] <tgr>	 jdlrobson: as long as they are on the same level in the DOM tree it does not seem too bad
[20:47:02] <tgr>	 things like (hello<span>)</span> would be more involved, but hopefully not really done in practice
[20:48:52] <tgr>	 from an internal POV ther is not that much difference beteen <span>(hel</span><span>lo)</span> and <span>(<a>hello</a>)</span> since all the text fragments there become separate nodes, and the latter is definitely something that will happen a lot
[20:50:17] <tgr>	 there is the HTML5 selection / DOM range API which would make it trivial but I wouldn't hold my breath for finding an implementation it that does not involve running something Electron-like
[20:50:49] <jdlrobson>	 tgr: i think as long as the existing tests still pass we have a good starting point
[23:29:54] <tgr>	 bearND, mdholloway: is someone working on T184557? I kind of cookie-licked that and then walked away which in hindsight is not a nice thing to do
[23:30:03] <tgr>	 so let me know if you want me to follow through
[23:35:41] <bearND>	 tgr: I'm trying but it's getting quite complicated, which is not good.
[23:36:54] <bearND>	 tgr: We also have this rule to not remove parentheses with there is not space inside them. Not sure how to handle this across different elements/text nodes.
[23:37:16] <bearND>	 s/with/when/
[23:41:08] <tgr>	 walk the tree twice, maybe? first walk, make a list of ranges to remove, and you can easily discard from there if there is no space or no closing brace or whatever
[23:41:24] <tgr>	 second walk, actually discard them
[23:56:43] <bearND>	 tgr: How would you store the ranges? AFAICS domino doesn't implement Document.createRange().
[23:57:20] <bearND>	 I guess I would have to implement my own structure.
[23:57:34] <tgr>	 start node, end node, character position in those nodes
[23:58:13] <bearND>	 yeah. Man this all adds up.
[23:58:46] <tgr>	 it's a lot of work just to get rid of some parens, for sure