- WIP: AI and multi targets, human and X=0 use cases, human and impossible targets use cases;
- improved stability and shared logic (related to #13606, #11134, #11666, continue from a53eb66b58, close#13617, close#13613);
- improved test logs and debug info to show more target info on errors;
- improved test framework to support multiple addTarget calls;
- improved test framework to find bad commands order for targets (related to #11666);
- fixed game freezes on auto-choice usages with disconnected or under control players (related to #11285);
- gui, game: fixed that player doesn't mark avatar as selected/green in "up to" targeting;
- gui, game: fixed small font in some popup messages on big screens (related to #969);
- gui, game: added min targets info for target selection dialog;
- for devs: added new cheat option to call and test any game dialog (define own dialogs, targets, etc in HumanDialogsTester);
- for devs: now tests require complete an any or up to target selection by addTarget + TestPlayer.TARGET_SKIP or setChoice + TestPlayer.CHOICE_SKIP (if not all max/possible targets used);
- for devs: added detail targets info for activate/trigger/cast, can be useful to debug unit tests, auto-choose or AI (see DebugUtil.GAME_SHOW_CHOOSE_TARGET_LOGS)
* added DebugUtil.AI_ENABLE_DEBUG_MODE for better IDE's debugging AI code;
* it's a target amount optimizations;
* it's use a grouping of possible targets due same static and dynamic stats (name, abilities, rules, damage, etc);
* instead of going through all possible combinations, AI uses only meaningful targets from particular groups;
- reworked skip and phase buttons to use layouts instead IDE designer;
- added GUI scale support skip and phase buttons (depend on dialogs font size from preferences settings);
- fixed broken database in some use cases (example: AI and choose name dialog, related to #11285);
- added docs and debug tools for sql queries, caches and memory analyse (see DebugUtil);
- refactor code to use shared settings;
- deleted outdated and un-used code (db logs, stats, etc);
* server: fixed that a critical errors ignored in user commands threads (now it will be added to the logs);
* network: fixed frozen user responses in some use cases;
* network: fixed accidental and incorrect user responses (only latest response will be used now);
* network: improved freeze logs, added problem method name and code's line number;
* cheats: removed outdated deck and card load logic (only init.txt commands supports now);
* cheats: fixed wrong priority after add card dialog (closes#11437);
* cheats: improved stability and random errors on cheat executes (related to #11437);
* docs: added details on network and thread logic, human feedback life cycle, etc (see HumanPlayer, ThreadExecutorImpl);