The command line options themselves were removed in commit
168313bed9da570d3bf7d4b6bc7fc93db1e7935f, wherein someone clearly didn't
do the full research.
Specifically, the SDL_net-based implementation of the network API never
supported or used these options. Only libana did.
Without this, log messages with --log-precise enabled tend to look just
a bit *off*:
20201019 23:09:21.786752info general: attempting to generate locale by name 'en_US.UTF-8'
^
This is semantically correct since we can't claim we own a console
before we find out if we have one or need to set one up.
Nobody should be calling using_own_console() before
early_log_file_setup() finishes, but you never know. People in this
project have surprised me before.
This had the confusing side-effect of making some code in wesnoth.cpp
believe it was safe to prompt the user to press Enter when NOT using
--wconsole on my system, because the flag ended up with a true value by
default.
This is why you don't leave integral-type variables uninitialized, ever.
previously the code put the brackets in the wrong
line and handfixed that but a wmlindent comment,
but that then broke possible other uses of that
macro.
wesnothd has four timers, three were using steady_timer, one was using deadline timer. This changes the final timer to be a steady_timer, which won't be affected by clock changes whereas deadline_timer would be affected. It also allows using std::chrono instead of boost::posix_time.
* SoF S9 - some cleanup to the final scenario
* Sof S9 - change time to tint red, replace remove key, add some small dialog to explain the reduced recall cost
* SotA - units - new sprites for sea captain, including non-skeleton undead version
* SotA - Sc4 Intro revision for ghoul background
* SotA - remove y,z characters from terrain codes, all campaign codes should now end with S.
* core terrain - new iron fence terrain *^Eqf
* SotA S2 - change large custom fence to iron fence
The remove attribute for [event] was never added to the validation schema,
and using [remove_event] seems easier to search WML for. Let's just recommend
the new key and deprecate the old one.
Instead of using the allied orb during their turn, the unmoved, moved and
partial orbs are shown. The player's own units will be shown with the moved orb
during the ally's turn, which matches the previous behavior. This is intended for
co-operative MP campaigns, where it will make it easier for players to discuss
possible moves. The UX is also visible in SP scenarios with allied sides, HttT's
first scenario is an easy place to see what it does.
If another status is added to the orb_status enum, this code is likely to
support it with no changes needed for the parts in this commit. For example,
issue #5155's proposed orb_status::disengaged has been tested with this.
Updated the documentation about orb colors. In these docs I've moved all of
the images to a single line above the bullet-pointed list, as otherwise the
layout looked bad once a line wrapped - the first line appeared
horizontally-aligned with the center of the image, the second started under the
image.
Added documentation about the delay in multiplayer games, as the purpose of this
orb-coloring feature is to assist discussion in multiplayer games.
Instead of creating lots of image::locators and calling a family of
similarly-named preferences functions, this refactor encapsulates that logic
and uses an enum.
The new orb_status files still need adding to the MacOS build.
An assert has been added to `unit_drawer`'s constructor. There are two callers
- the `display` class guards it with an explicit check before constructing
unit_drawer. The `game_display` class doesn't have an explicit check, but it's
clear that other code in that class assumes the teams are already valid (and
would crash if they weren't).
