In rP25869#54232, @Jammyjamjamman wrote:

In rP25869#54228, @Langbart wrote:

I found some strange behavior that could be caused by this changeset, but I'm not sure, could you take a look at it? #6365

This bug is same fog of war bug. The other units aren't missing, just hidden. (You will see them if you send your units to attack.)

PS: I tested ​Phab:D4305 as there was another error related to [25869] (#6356), but it did not fix the problem.

When I applied D4305, It did fix that bug. However, 0AD does some strange caching + retrieval when loading previous scenario maps played. So if you initially played the map with 0ad svn on or after this commit, then apply D4305 and reload this game again, it will load the map with the same (buggy) settings as used on the game before this patch.

Long story short, apply D4305, start the game, select a different map, then select the Combat Demo (Huge) again and the map should load correctly.

In D4317#183850, @Angen wrote:

Wait, didn't you say they are promoted to next rank?

In D4317#183843, @Angen wrote:

Yes, but you can see in tooltip that required is 150, so why calculation above produces trash

Removing unnecessary square brackets. Now it is the same as in the introductory_tutorial.js, which also does not use square brackets.

In D4317#183836, @Angen wrote:

Anyway why is req 0

Adding || 1

In D4317#183833, @Angen wrote:

More like req || 1 but if entity is promoted, division by 0 should not happen

I found some strange behavior that could be caused by this changeset, but I'm not sure, could you take a look at it? #6365

In D4317#183814, @Angen wrote:

what happens when you place basic mercenary on map and runs it normally in game?

A small problem was found caused by this changeset number. #6362

In D4305#183758, @andy5995 wrote:

[...] I wanted to suggest enabling "allied view" as the default option. I just anticipate that the majority of hosts or players would want to see their allies when the game starts. Any opinions on this?

Remove the striking "orange" color and replace it with a more gold-like color.

Adding some Square brackets to match the style in the rest of the document.

Apply the patch and compile the game

Ok

Run the game and open the developer overlay

Ok

Check that checkboxes and dropdowns work

Ok, but I would change the style of the dropdown. Increasing the devCommandsText even more would also look good, sans-16 fits nicely.

The auto-produce icons have all a hand on it.

How do you like the extra hand and orange color?

Added Valihr to the reviewers since I need someone.

Revision that changes gameplay should have [Gameplay] in the title, this will automatically subscribe the Balancing Advisors, see e.g. D4309.

In D3998#183438, @JCWasmx86 wrote:

After an hour I destroyed all her four CC and she started to retreat to a central point in the middle of the map, after a short time the Ai committed suicide and I won.

@Langbart I wasn't able to reproduce it. (The red ones won at the end, the blue one did nearly nothing).

Optional:
(1) if the variation would be switchable after selecting the structure that would be good.

In D4305#183394, @andy5995 wrote:

Also I think more courteous to the developer, who agreed to fix this bug, and did so in a timely manner.

In D4307#183399, @Freagarach wrote:

Please consider D2927 (and its commit).

Glad you made it, good to have some new gameplay patches being made.

Keeping all existing sizes, as 384/448 are good for nautical maps and we don't want to limit the possibilities.
Only a renaming of the maps above "Normal" so that we can get rid of "Medium".

In D4305#183348, @Jammyjamjamman wrote:

Could be done. But there would end up being 5 options:

• Normal
• Explored
• Allied View
• Explored + Allied View
• Revealed

This also helps new players, because sometimes people ask on IRC how to change the number of AI players or the size of the map. Random maps allow this.

The naming could be done better instead of Tight Arc/ Circle just call it Teams Together/ Equally distanced.

Works, would you agree to put those 3 options under one dropdown menu?
AOE HD calls it Reveal Map

Should the option appear if there are only two players? Better just disable and gray them out.

I would remove the Random option, players expect a certain type of team placement by default. There is no Random for the map size either.

Reduction of duplicate code via const hasSecondary = !secondaryName || primaryName == secondaryName;

In D4304#183219, @chrstgtr wrote:

I would warn, however, this can create a real mess with pathfinding as units may inadvertently walk by these and get damaged. That would be neither realistic nor desirable from a gameplay perspective

Yes, you are right. The AI keeps sending cavalry units to the same place and the horses keep dying. There is no need to test this patch further.

Redo the patch with the help of bb.
Looks better.
The Ishtar Gate of Babylon has only one name and the name is always nicely centered.

Removing capture points and the rangeoverlay.

In rP25869#54158, @Jammyjamjamman wrote:

I can push a fix soon. Should this be done in a new patch?

An issue has been discovered that can be observed for the first time with this changeset. See #6356

Noticed:
(1) The sentence This item is paused.
Why is it in paranthesis? Why not add some color, for example orange. I would even put the little hand symbol in front of it, then everyone who hovers over it will see faster what it is about.

I would only give this to special units like ships or battering rams, but not normal human units.

This change makes it very cumbersome to use formations when moving units around the map. Since the formation is set to the box formation by default (rP24480), new players will be shot to pieces. At least change it to "none".

There is only a very short moment when the idle number on the minimap still "twitches". This is definitely an improvement.

With D4217, the number of units is now displayed on the minimap, idle units are alternating the isIdle state from true to false, which causes the number of idle units to change constantly. When I apply this patch to my working copy rP25953, the problem described in ticket #6279 no longer occurs. It fixes it.

In D4297#183023, @Freagarach wrote:

Instead of removing the templates, one could create a template_bridge with no health/resistance and let the two used templates inherit from that. But since the templates are wonky and don't add anything, deletion seems like the correct choice.

In D4297#183000, @bb wrote:

If we just want the AI to stop attacking them, I suppose we could also remove the health component and be fine.

If I remove the Health tag it throws errors at me, is there another way?

Changing the buttons from ModernButtonRed back to their original StoneButton style.

In D4296#182920, @Freagarach wrote:

It seems that in general, the red buttons are used for pupups, but either way, we should make them not less consistent, but more consistent.

Small discussion on IRC 0ad-dev (4/Oct/21), increasing width to 1280 meets little resistance, increasing width to 800 meets some opposition who would rather see 720 or 768. Below 768 we would have even less vertical space e.g. for displaying control groups, I would be against it and would take either 800 or 768.

• 1280x768 or 1280x800 seem to be suitable candidates.

The most important part is the increased width, which would allow a larger minimap, and increasing the width of the summary screen columns, which are too small for some translations (see #6025 - Summary page problem with long words (15/Feb/21)).

Trying to solve with getTextSize().height.

Change a variable from let to const after the linter has recommended it and improve the onPlayersFinished function.

Removing the size from the .xml file and adding two elements.

A new problem has surfaced that can be noticed for the first time with this changeset, see ticket #6340 for details.

I tried this patch on a random map called Sahel Watering Holes for a 1v1 against a Medium Balanced Petra Bot.

In D2926#127767, @Nescio wrote:

Scenarios and skirmish maps are listed in alphabetical order and can't be randomized (fixed number of players and map sizes), therefore the first map people encounter is always the “Acropolis Bay (2)” skirmish map. While it may not be that bad per se, it's probably not the best 0 A.D. has to offer. Moreover, many (most?) new players don't explore other possibilities, they just start a match.

Fixed a duplicate description entry in the GameSetup.cpp file.

Two issues related to the battalion function:
Issue:1 - movement of several battalions at the same time
In the image below I have got 2 battalions.
One with 20 swordsmen in the Testudo formation and one with 28 skirmishers in the Close Order formation. When I select both and move them forward, they change to the Box formation and become one battalion. I would prefer them to be separate after advancing. If I change the small default icon in the formation to None, both battalions disappear after I move them forward.

Verify that (de)selecting behaves as usual. Test with the mentioned diff to verify the reported bug is fixed.

In D4282#182561, @Freagarach wrote:

In D4282#182543, @Langbart wrote:

When I try to deselect some units by holding down the Ctrl key and left-clicking on the icon, it only works for battalions with one unit type. I am unable to deselect battalion 2.

Yes, you will select the entire formation by default, unless using the modifier to not do so.

But a ctrl + left click on any unit icon that is part of battalion 2 should deselect the entire battalion 2, at the moment it does nothing.

In rP25884#53743, @Freagarach wrote:

Can source/lib/tex/tests/test_tex.cpp be ignored somehow?

In D4282#182543, @Langbart wrote:

This image was only to illustrate my point with deselecting units, but the layout looks nice. Having a bunch of units grouped together to see who is part of a team looks appealing.
Hyrule conquest does it but with control groups.

When I try to deselect some units by holding down the Ctrl key and right-clicking on the icon, it only works for battalions with one unit type.

In D4282#182494, @marder wrote:

But why is there only one gather at this tree?

Yes this solves the problem described earlier in D2175#182343. The woodcutters look for a new tree after they are done with the first one.

In D4108#182477, @s0600204 wrote:

In D4108#182476, @Stan wrote:

@s0600204 any idea what we could do for macos and pkg config ?

Rebase to take into account recent changes, then add the below suggested line so that the generated .pc file is in the correct place for our scripts to find it. (There's no need to relocate the header files.)

In D4280#182376, @wraitii wrote:

@Langbart: Check that it works with non Kushite units but if so yes.

Seems to work. 0AD has special civ related structures (e.g. camp_noba, temple_amun, ...) still in the units/{civ} templates. Would it make sense to create a patch and get all the structures for all civs into the mixins/builder.xml file? See below, one needs to get into seven files and remove the redundant line now.

@wraitii Is it allowed to move the structures/{civ}/pyramid_small into the builder.xml file? Layout looks better this way.

The units in a battalion, who were tasked with cutting down trees will only cut down the first tree and then remain idle. They will not look for a new tree.

In D2175#182341, @marder wrote:

Since the battalions are now tied to formations (as far as I understand), is it still possible to have a battalion and send it to gather something or will this disband it, since the formation can not be used for that?
Cause I liked it to be able to select one wood cutter battalion with one click and task them with something different.

Tested it on the map Hindu Kush (2) with steep hills, Farmland and the scenario map Units Demo. The problem in the referenced ticket can no longer be reproduced. For a player/user this solution works.
EDIT: Also tested in Atlas, works.

IRC 0ad-dev (22/Sep/21)

In D2175#182306, @Freagarach wrote:

http://irclogs.wildfiregames.com/%230ad-dev/2021-09-20-QuakeNet-%230ad-dev.log
I'll try to merge this this week then, unless objected.

In D4278#182263, @Stan wrote:

@Langbart can you check cursorbackend="system"

IRC 0ad-dev (20/Sep/21)

19:12 <@Stan> @tell Langbart have you got a retina screen ? Can you test D4278 anyway ?

Yes, I do use a retina screen. Ok.

In D3611#182259, @Stan wrote:

Any errors/warnings? Else it's good to go I suppose @s0600204

In D3611#182257, @Stan wrote:

pass --force-rebuild to build-osx-libs.sh

In D3611#182254, @Stan wrote:

Can you run the script without --preserve libs ;) You need to rebuild the thing for the pc files to be copied.

Same error as above.
Entire terminal output: https://termbin.com/47vw

In D3611#182252, @Stan wrote:

Can you see if you can find the fmt.pc file? It might be name libfmt.pc or something similar.

In D3611#182197, @Stan wrote:

@Langbart can you test this maybe we'll be able to debug what goes wrong on macos :)

Re