@lethosor - any idea why Travis is silent?

It was disabled temporarily because of the current state of the develop branch (e.g. the ruby plugin won't compile anywhere, besides the issues on Windows addressed in another PR). I'll see if there's a better way to check PRs, but I doubt this PR would pass the compilation stage. I can definitely test this locally, though (although it might be a couple days).

Rebased on develop to fix merge conflicts, added a commit to improve docs/Core.rst wrt use of DFHack commands and document the new syntax from #767.

autodump-destroy-here is actually slightly different from autodump destroy-here (although I thought they were the same myself until now) - the former uses Gui::cursor_hotkey, so DFHack will never even attempt to call it if a cursor isn't available. With the latter bound to a hotkey, you'll get an error logged to the console if you press Ctrl-Shift-K, which is an issue if you have Ctrl-Shift-K bound to another command that works in another UI state.

Incidentally, you do get an error if you run either version from the console. As far as I know, plugin commands can't detect whether they're invoked by a keybinding (only the DFHack core can), so merging the logic into autodump_main and making it silent when invoked as a keybinding isn't possible. Even if it were, it's easier to delegate the task of determining whether the UI state is valid to the hotkey guards (Gui::cursor_hotkey, Gui::any_item_hotkey, etc.).

Anyway, thanks for documenting the +-style arguments - that showed up in 5-ish changelogs around 0.43.03, so I must have assumed it was documented already, but apparently not. :)

OK, looks like the core docs need some extra work - I'll split that out into a new pull request. I'll also remove the autodump changes, and squash the rest for clarity.

The good news is that Dwarf Fortress.exe" +devel/print-args test1 +devel/print-args test2 prints test1 and test2, working as expected and with multiple commands (which I'll now document).

Rebased on develop to remove merge conflict in NEWS.rst, and removed use of lru_cache (again).

@lethosor - I've fixed the merge conflict, any other blockers?

I've been dealing with OS upgrades and changes to Lua GUI scripts/hooks for a few days here, and I'll probably be busy in real life later this week. Sorry for the timing being somewhat inconvenient, but I'll get to this when I have time. No need to delete/recreate the comment to notify me again - I do check the list of PRs periodically, and definitely before releases.

Sorry - just checking that it wasn't a problem that I needed to fix.

(I've also added a commit to expand the workNow docs, per a request via Reddit PM)

Rebased to fix merge conflict in NEWS.rst