Vitalie Spinu
Vitalie Spinu
CI build on mac and all CRAN checks on mac are ok. I guess it has to do with the GCC on your machine. Any chance you can compile it...
`iter_rows` also happily runs on rows with object columns.
I think these should be better done on every mode basis. It's not sustainable to keep adding altering modes inside polymode proper. Each mode's author knows best which variables are...
Yes, an option to return a matrix is a much better idea. Could get rid of the print method. Those are for debugging purposes anyhow.
> One idea was to drop buffer argument in handlers and provide a macro that would create a handler with current buffer and current connection stored in its closure. The...
> but why do we need the current/another buffer? Not everything acts on it, so maybe the handlers should just obtain the buffers they need. I see at least two...
> A caller making an nREPL request will then simply construct their response handler using function composition: > > (->> (cider-repl--out-response-handler buffer) > (cider-repl--value-response-handler buffer) > (cider--ns-response-handler) > ...) This...
> (BTW, -handler postfix is inappropriate here. CIDER's usage of the term handler also seems wrong.). Never mind. It's a handler constructor, not a handler itself, so it's fine to...
Same here. Would be more than happy to through a few coins in support!
> It doesn't. You can use any memtable with a direct con.execute. I see. I was not aware of this. So this would execute regardless of the default backend: ```...