Jan Moringen

> My impression from the documentation was that the parse-tree package is intended to be lower-level than CSTs and to leave the question of what is constructed to the client....

I will try to improve the documentation, maybe after settling on a way of composing skipped and "proper" results.

I experimented with something like this a bit. I used the `:arc-drawer` option of `format-graph-from-roots` to allow for different arc drawing styles: https://techfak.de/~jmoringe/mcclim-graph-layout-1.png (the choices at the top are for...

@john-a-carroll I think your application would not be affected at all if we provided alternate edge drawing styles as additional functions to supply via `:arc-drawer`.

Here is a sorted version with progress indicated: * [ ] `clim:accept-values` * [ ] `clim:accept-values-resynchronize` * [x] `clim:add-presentation-translator-to-command-table` * [ ] `clim:augment-draw-set` * [ ] `clim:cache-output-record` * [x] `clim:check-box-selections`...

In my opinion, graphical changes such as changing shape dimensions or maybe even line breaks in response to layout changes are best done by adjusting existing output records, if practical...

> is this really clouseau-centric enhancement? Because of the label? No, it is not Clouseau-centric. It just has something to do with Clouseau.

Sorry for taking a long time to reply. On my system, the test uses the font `DejaVuSans.ttf` which doesn't seem to have a glyph for the character. Maybe McCLIM should...

Thanks. Applied in https://github.com/s-expressionists/dpans/commit/8c75767627d41db43ec36109905bffa333546c91 (except the `#'` part since we only address errors for now).

> it would be nice to have an option in the `defrule` macro to add custom error messages. > > in the case of complex rules there might be a...