| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
|
|
|
| |
nothing otherwise.
|
|
|
|
|
|
| |
immediately cast
to std::string.
|
|
|
|
| |
This should remove the current abundance of more prompts.
|
|
|
|
| |
This reverts commit 6efa910975ee9276ad90a84016a4bbece541448c.
|
| |
|
|
|
|
|
| |
Otherwise, the line count is off when prompts are involved,
causing the "more" prompt to overwrite the last line.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
The newline parameter to message_out is gone. Instead, a
pointer to the message line is passed which may lie outside
the message window. The window is then scrolled to make
that line the last line.
This also updates libnonunix.cc, but I haven't been able
to test those changes.
|
| |
|
|
|
|
|
|
|
|
|
| |
This fixes the level up message disappearing with
delay_message_clear and is probably the right thing to do
anyway.
Before, the ---more--- would be printed before older messages
were shown.
|
|
|
|
|
| |
This gives a clear indication of which messages happened in the
current turn.
|
| |
|
| |
|
|
|
|
|
| |
You can now also turn on autoclear-more status from outside message.cc,
rather than just turning it off.
|
| |
|
|
|
|
| |
This mostly puts && and || on the proper lines, per the style guide.
|
| |
|
| |
|
| |
|
|
|
|
|
| |
I'm pretty sure I've managed to replace every instance of a reference to
these files, so I hopefully haven't missed anything.
|
| |
|
|
|
|
|
| |
In most places Feawn is now referred to as "Fedhas," its long name
(used in the religion screen) is "Fedhas Madash"
|
| |
|
| |
|
|
|
|
| |
New: colour.cc, coord.cc, coordit.cc, random.cc, rng.cc.
|
| |
|
| |
|
|
|
|
|
| |
Signed-off-by: Brendan Hickey <brendan@bhickey.net>
Acked-by: Steven Noonan <steven@uplinklabs.net>
|
|
|
|
| |
Signed-off-by: Steven Noonan <steven@uplinklabs.net>
|
|
|
|
| |
Signed-off-by: Steven Noonan <steven@uplinklabs.net>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
at the more after "You have lost your religion!" to avoid the negative
consequences of excommunication): when a HUP signal is received, instead of
immediately saving the game, close the stdin stream, which will cause all of
the Curses calls which normal block on key-presses to unblock, so that the code
after a more() call will execute before the disconnected game is saved (which
will happen in _input() in acr.cc). This only affects Curses
(non-tiles/console) UNIX builds which have both USE_UNIX_SIGNALS and
SIGHUP_SAVE defined.
Exceptions to the above:
* If HUP is received during a call to yesno() with safeanswer == 0, then
the game will be saved immediately, since there's no way for the
function to tell which answer is safe to give automatically.
* If HUP is received while selecting the target for a controlled teleport
then the teleport will be canceled, since otherwise a target other than
the intended one might be the one under the cursor when the disconnect
happened.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@10443 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
| |
Also fix potions of porridge's menu colour being yellow for Mummies.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@10355 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
| |
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@10271 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
| |
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9984 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
|
|
|
|
| |
previous command (`) (at least for actions involving selecting an item from
inventory, selecting an ability to use, selecting a spell to cast, and
selecting a spell to memorize).
Not tested with the tiles build.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9839 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
|
| |
from the save file. This fixes a two-fold bug with the tutorial "welcome
back" message, a lack of a --more-- prompt and some line-wrapping glitches.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9826 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
| |
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9709 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
|
|
| |
actually allow eating of rotten stuff.
* Tweak Xom's mood descriptions -> shortened to 7.
* Increase chance of Xom acting under tension: 5% -> 10%
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9535 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
|
|
|
| |
by adding a boolean to keep track on whether we already called
flush_prev_message() this round, reset once a new message is pushed into
prev_message. There's probably a better solution, but this one works
well enough.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9285 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
| |
mpr_formatted_output(). Fixes 2647846.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9275 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
|
|
|
|
| |
* Fix grid descriptions only showing belatedly.
* Fix tutorial starting prompt not being shown.
There are likely to be more instances where flush_prev_message() needs
to be called although the turn isn't over yet and it's not a prompt.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9265 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
recent message in prev_message and only output it when another non-matching
message rolls in or the player gets a turn. Matching messages (must be
identical and issued in direct succession, like is the case with the
message history condensation) increase the counter, so you end up with
messages like:
You feel sick.
The killer bee misses you. (x3)
The killer bee stings you.
The killer bee stings you but doesn't do any damage.
The killer bee misses you. (x5)
...instead of the 11 lines it would have been in total.
This behaviour is controlled by the same option as the condensation in the
message history, msg_condense_repeats.
There's definitely room for improvement. At the moment the bracketed
information is written into the message itself, which will prevent
further merging in the message history, or, probably worse, cause ugly
double-merging like "The killer bee misses you. (x3) (x2)".
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9262 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
| |
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@9049 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
|
|
|
|
|
| |
to immediately move out of view, by introducing the notion of the currently
acting monster to crawl_state, and only flushing out the "comes into view"
message in mpr() for the currently acting monster. Not sure if it's worth it
just for the sake of avoiding doing a "has this monster just now come into
view" check in every place that a monster might issue a message, but at least
this way we won't miss any places such a check should be placed.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@8623 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
messsages. Such messages are not not flushed immediately upon the monster
coming into view, but:
1) After world_reacts() is done.
2) When mpr() is called and the player is either delayed or repeating a
command.
This should make sure that the "comes into view" message comes before any other
messages the monster might give, and also fix the problem of the player moving
into a monster's LOS, getting a "comes into view" message, and the monster then
moving back out of LOS during the immediately following world_reacts(), which
would leave the player puzzled as to why there's no monster in view.
As part of the change you.turn_is_over is set to true as soon as world_reacts()
starts, which might cause bugs in other places, yet some play testing indicates
that it's okay.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@8621 c06c8d41-db1a-0410-9941-cceddc491573
|
|
|
|
|
|
|
| |
* Better feedback when quiver cycling doesn't do anything.
git-svn-id: https://crawl-ref.svn.sourceforge.net/svnroot/crawl-ref/trunk@8552 c06c8d41-db1a-0410-9941-cceddc491573
|