Does the 15% stack?

I'm no historian, but I wonder if we could tweak the animation to make it look more realistic rather than tweaking the attack rate (I assume the realistic repeat rate of a spear is far slower than that of a sword?).

Addressed by r19735.

String change looks fine, although we should probably change capitalization to be consistent with other buttons: "Mark as Buddy"

Are there any cases where we would get no player selected (like selected player leaving or finishing a game or something?) and the enabled=true still that would cause undefined behavior?

In terms of use for a typical user, wouldn't we also want this setting in the gamesetup windows? I'd imagine that the options page on the main menu would be the global default while gamesetup could hold match/this instance specific settings. e.g. if we want to set assigned by default and find that too many irrelevant players are joining, we could switch the setting from within gamesetup so we don't have to re-host to change it.

Definitely agree with this change.

After discussion on IRC, we agreed to also add an "unassign by default" option.

Couple reasons:

Definitely agree with this change. :)

Not sure this is the right way to implement this. Other games often have a "spectate" button and a "join" button, with the join button (and/or the spectate button) settable to require a password by the host. IMO this seems like like over-use of the buddy system (and makes it harder to reason through).

Have you been able to successfully run newer versions of Ejabberd?

Why are there rating parsing changes in the context of this patch?

Sounds good to me.

Go for it! :)

Just to confirm, that error was when you tried to connect your local ejabberd instance, correct?

I think target time for a single fireship to destroy a dock should be around 15-25 seconds.

Perhaps reduce HP of the fire ship? It doesn't sound unreasonable that a dock is vulnerable to fire and may require extra defenses around the dock if an opponent has Iberians.

Maybe increase repeatTie/prepareTime and decrease Hack and Pierce attack? Crush should be the primary damage as both crush and fire should be the most effective against structures.

The dots seem like the best solution to me.

I'm going to have to agree with vladislav, "Create" rather hides the fact that they are the host (even if we have NAT traversal and stuff, it probably won't work on some networks and they might wonder why).

Why not just "Registration Forbidden"?

I'd lean toward name sorting, but either is fine.

I'd agree with elexis. The extra header doesn't add much in terms of helping the user understand the meaning of the column while making it less attractive.

Can you add a screenshot as well?

What is the function of the Buddies radiobutton?

D280

Unnecessary imo.

Wouldn't you check all settings before you click "I'm ready"? If we can't take the time to look over all the settings before we press "I'm ready", I doubt that you're going to catch much more in 10 seconds.
In any case, if the process of readying-up involves too many steps, I think the overall better way would be to change the "Game settings have been changed." message to a more descriptive message mentioning exactly what has been changed.

Maybe yellow for regular users, orange for friends, green for mods, light-blue for mod-friends?

Seems like a good concept. :)

I'd actually be in favor of making all players one color, moderators in another, buddy in another, mod+buddy in another.

New numbers look acceptable.

I'm quite reluctant to add another radio button at the top. Can we make buddy games appear in bold instead?
Also, still don't really like buddy filtering. It seems like a lot of trouble for something that not many would use (and if many did use, new players might feel left out etc.) Could also increase the number of bug reports of "why is the lobby empty?" if someone accidentally presses the button.

I'm not too sure about having chat filters...it may lead to an assumption on the part of the player that his or her messages are only visible to the filtered party, which I think (?) is not the case.

Removes broken cleanup code.

New buddy column rather makes an already cramped list even more cramped. :/

Other sections still need some English proofing (which I will do later if I have time)

Would make it easier for the translators (and anyone who isn't fluent in English)

"Filter out open games" or "filter by leaving only the open games in."

"Filter open games" seems a bit ambiguous (filter out?). Perhaps "Only show open games"?

gonna change some things to how iqhandler works

Some English suggestions. :)

Agree with fatherbushido. There are also a lot of changes in this diff that aren't mentioned in the summary. and thus, has no visible justification within this page. This revision should probably be split into multiple tickets.

I'd rather see the game list only hold concrete values (attributes of the game itself) instead of derived fields, but after talking with elexis, this feature seems like it would only be used by a subset of the lobby; perhaps a config option customizing columns?

Personally, I'm not sure really sure having an extra column really helps that much (1200 could mean a 1500 + 900 vs two 1200s), especially when we're already pressed for space in that box.

Dunno about you guys, but in about every game I've seen, catapults have more HP than ballistas :P

Ah okay, I was using the sandbox 2 for ptol.

I'm still getting 9 units for the mercenary camp. I'll check on my end to see if it's a caching issue or something.

I'd say nah. What's the purpose behind the tokens deletion line, btw?

I'm not really liking having 9 units in the mercenary camp stretching onto two lines...perhaps we can remove the non-mercenary units from that building?

Or we could move most heroes to the CC (like Macedonians currently).

Ah ok. Sounds good. I'll review the other parts soonish.

Perhaps we could include a warning the first time the player clicks this about the host being able to pull a fast one on them? :P

I'm probably quite out of touch for balance, but I envisioned the mercenary archers to be a bit like a handicap early game for Ptolemies (low health, costs metal) in terms of early game military might (with the exception of camel archers). With free basic structures, are there any handicaps for Ptolemies to stop developing quicker than other civs?

Looks good to me.