Rename location_palette::can_swap() to the correct name so that it
overrides as intended, thus making the UI disable the "swap fg and bg
items" button.
The class structure doesn't match the responsibilities here, as part
of UX isn't part of the palette, rather it's part of the tool that
decides whether there are fg and bg items - alternatively, the tool
decides whether right-click is "place bg item", "delete items", or
"show unit tool options". However, this change seems to be enough to
make the UX correct.
The "virtual" is redundant, but add it for consistency with the other
code in these classes.
When I use an ability id=A and include [filter][filter_adjacent]ability_id_active=A, the more units with the adjacent ability I add to it, the slower the game becomes, and at the third unit the game freezes, whereas with the direct comparison of the configs the game only slows down significantly after the 5th unit added.
The only place that used these was the lua show_dialog implementation. We can just construct the window object directly instead. Since this skips out on the finalize_build step (which in the case of modal/modeless_dialog was being called by those respective ctors), I've removed that function and merged it into the window ctor. No need to do it separately.
Builds on work in af81bba53b and 247e5ff055
Note that location_palette::get_help_string is not virtual as it does not inherit from editor_palette (where the pure virtual get_help_string is defined) but rather common_palette.
Resolves https://github.com/wesnoth/wesnoth/issues/7926
[filter_ability] and [filter_ability_active] have been merged into a single filter equipped with the 'active' attribute which when the value is 'true' checks the units affected by the sought ability and when the value is 'false' or unset will check the units carrying the ability even if they are not affected by it.
Because of the attribute strategy, I'm afraid that the developer will be mistaken about its function and that's why I'm not at all keen on it but I want to get out of this impasse.
all_directions better reflects the purpose of the former. Also made it return a value,
since the only places that used it immediately assigned it to a local variable.
The first SDL_KEY_DOWN handler is being called twice, while the second one is being called once.
The tab order handling is done by the first handler, which caused the focus to move *twice*, resulting in the erratic behavior when tab was pressed.
Moving the tab handling to the second one stops this and now focus moves only once when Tab is pressed.
With custom gui2 theme support in #9057, it is now possible to create addons that supply UI Themes or Skins.
This adds support so such addons are properly recognized and validated.
Resolves#9009, closes#9384. Besides the issues that came from having both int and bool conversions (even with bool marked explicit), it doesn't make sense to use int for all numeric assignments. I didn't test whether it would be an issue in practice, but it seems better to let callers be explicit about what type they want for numeric values than risking overflow or wrapping for very large values.
* Removed duplicate map_location hasher, removed unused teleport_id variable, dst_node reference is used at other possible locations, teleport destination heuristic distance now calculated before path finding loop only once, teleport source heuristic distance now stored in nodes, only calculated once and reused.
---------
Co-authored-by: SomeName42 <>
Change Jumpcat from level 1 to level 2
Increase HP from 32 to 40.
Reduce MP from 9 to 7.
Reduce arcane resist from 20 to 10.
Increase cost from 15 to 26
Replace claws' parry with backstab, but reduce damage from 6-3 to 5-3
Remove accuracy from tail, but increase damage from 11-2 to 13-2
* address bluffs-mountains transitions
* revise visual indicator of flood-filled elevation regions in editor
* small border versions of dry and basic mountain
* fix some frozen/elevated transitions
* address mini-map concerns in #8924
