In D461#22270, @elexis wrote:

Just like all the other patches that disable things if the game is officially paused

Such as?

I'm not a fan of this patch since I've always used pauses to already queue some commands in advance.

This doesn't seem like it should be allowed to happen, as if the game is paused, it should be (by definition) actually paused, and prevent players from spamming commands (possibly unintentionally, from new players) that can build up without any response or indication from the gui.

Besides preventing players from sending useful diplomacy changed messages, it also doesn't prevent sending of useless (duplicate) diplomacy cahnged messages, so the title is misleading.

Actually, if one follows the test plan for this diff, they would notice it does prevent sending of useless/duplicate diplomacy changed messages, so I'm not sure where exactly this is coming from.

Actually fix the hero healer bug.

In D544#22420, @mimo wrote:

Thanks for the patch. It looks good, although i'm a bit afraid that the logic is becoming quite complex (and difficult to extend in the future).

Fix comments.

In D545#22430, @mimo wrote:

Could you be more explicit? this replay has 24000 turns, and watching such a long game to catch a unit jerking is too much for me. Do you know approximately the turn number, otherwise if it happens again, the simpler is to save a game when this happen and upload it here. I've tried several times on my side, and could not reproduce it.

I believe it may happen in this one, for player 4 in particular.

In D544#22311, @mimo wrote:

I'll review it tomorrow, but in the meantime (and as it is one of the comments i had from a quick look at the patch), could you add some comments in the diplomacy manager explaining what are the differents objects, in particular the difference between this.diplomacyRequests and this.sentDiplomacyRequests and what is the exact role of each one.

Add some more comments to be more clear as to the purpose of the different DiplomacyRequests Maps.

I have noticed something else. When territory boundaries are close, it can happen that units can get stuck in a loop between retreating to the nearest structure and moving out towards enemy units, and this results in jerky movements for the units back and forth (It's kind of hard to explain). Maybe there needs to be a threshold set up with metadata or something (I'm not entirely sure).

Don't send the diplomacy network command if it a pointless one (i.e. clicking on the enemy button with a player you are already allies with).

Don't send requests to defeated players, and don't include gaia when determining mutual enemies.

It will be good to have these improvements to Petra's behavior towards "kiting" :)

It seems gaia entities can spawn completely submerged, although I'm not familiar with the relevant code:

This is was discussed again, and I believe that this would be good to have for the next release. Being able to copy the replay path is not as important as
actually making it clear for new-comers where replays are stored.

Nice! I will test.

Address remarks.

(I didn't do any proper testing/measuring, but it seems this rain impacts performance quite a bit.)

In D532#21817, @mimo wrote:

I missed something previously: it is not clear to me if the targetedRelics should only contains gaia relics (which i would support as a raid may not be adapted to fight for an enemy relic) or not: if yes, i've indicated inlined some changes which are needed. If not (i.e. if targetedRelics also contains player relics), we should add a way to remove them if the relic has been moved into enemy territory.

Set m_WaterUgly to true.

Missing semicolon.

Remove duplicated code for raids, and rename targetEntity function to raidTargetEntity.

Update.

Thanks mimo for the remarks!

More style fixes after discussion with @elexis. Move the regicide garrison option below the name of the victory condition in gamedescription, check for capture_the_relic in Setup.js, rename potentialSpawnPoints to potentialNeutralSpawnPoints in TriggerHelper.js

In rP19615#21243, @elexis wrote:

Thanks for the fixes rP19622 and rP19623!

And rP19627 for rP19622.

Move a function to TriggerHelper and use it for regicide and capture the relic to remove duplication.

Check for UnitMotionFlying.

Regicide.js cleanup.

Update after rP19614, remove unneeded check in unit_actions.js, and handle maps with no neutral territory.

Check for getStance() in setStance().

Check for getStance() on init, remove uneeded setMetadata, and properly get garrisonEmergency.

Allow getStance() to be used by the AI.

In D333#20534, @mimo wrote:

Sorry, i forgot to add you on the commit message :(

Thanks mimo, I've updated the patch.

Thanks for the comments.

Use "Hero Garrison: Enabled" and "Hero Garrison: Disabled" for the label strings in gamedescription.js.

Ok, I have done some updates.

Update.

Petra is fairly easy to defeat when regicide garrison is disabled, as it will only pick one location to "retreat" the hero to when its health decreases (the closest base/CC), but I am working on some improvements for another diff similar to the regicide AI additions in rP18731 and rP19085.

In D495#19828, @elexis wrote:

Is it a must have? because read-only input text fields are not implemented yet and I don't see me having time for this in the next 2 weeks

While at it, the replay filename should become copyable.

Rebase

Toggle diplomacy and barter buttons of menus that are open when the game becomes paused.

There is also a check for target.healthLevel() < 0.55 in defenseManager.js and some other places, but these are special cases.

Move the range visualization options to "General" in the options menu, and use "display" instead of "rendering" in the comment in the config file.

I've noticed on this map that sometimes trees spawn very close to each other and clip sharply, kind of like what happens sometimes on the "Kerala" random map.

Also, I think at some point it would be best to explicitly target melee siege units, as they do a lot of damage to structures.

When starting the map, this warning always appears: WARNING: Path with name 'danubius' doesn't exist

Disable the buttons in the diplomacy menu when the game is paused, and do the same in the barter menu for consistency (after discussion with @elexis).

This has already been done for CaptureTheRelic.js in rP19345.

Nice! These sorts of improvements to garrisoning attacked structures are great to have.

Well, this diff doesn't seem to break cinematic camera functionality anymore :)

Update after the changes made in the commits in the last day. The failing test is expected, as this patch adds a new component.

Reset only if the original team of mutual allies is altered, removing the need to modify the player component.

@elexis believes that the timer shouldn't be reset when a player on the team-to-win changes their diplomacy with some other player from neutral to enemy - that makes sense. But what if player A is allied with player B, but player B has their diplomacy stance with player A set to enemy, and then later, when player A's team starts the countdown, B switches to become allies with A? Should the countdown reset then? Maybe not, but it certainly comes down to a design decision.

Now with the recent changes in gamesetup.js, no warnings are thrown when a map specifies a Victory Duration value that is not in victory_times.json, but when looking under "More Options", the Victory Duration value is displayed as "Unknown" (even though it is shown correctly in the game-description, and the value is also passed correctly to the simulation).

Using a heart-shaped texture/mask:

Use a Garrisonable component based on the way that cmpWonder is used, and update after the gamesetup changes in rP19504.

Tested, fixes the bug as advertised.

Update after rP19517, and reduce the thickness for the inspiration aura so that it is not very visible unless zoomed in.

Update after rP19518.

Thanks for the comments!

Add a missing check before the submit call in cmpselectable line 672; if the game was paused while hovering an entity with an aura overlay, then the cursor was moved away and then back onto the entity, it would crash. There are already equivalent checks for these null cases before submit calls in cmpselectable (f.e. line 663 if (m_UnitOverlay)).

This will be a great option to have for this victory condition! This diff also showcases the ease-of-use of the gamesetup code after the changes in rP19504.

Remove an uneeded continue in GuiInterface.SetRangeOverlays, remove a redundant comment in RangeVisualization.js and change the formatting of cmpSelectable.AddRangeOverlay.