Age | Commit message (Collapse) | Author | Files | Lines |
|
then you won't be able to type spaces in the NPC integer or string
dialogs. And no, assigning focus to the ok button doesn't resolve
anything, but makes things worse, as then the user has to click on the
input field to input text, then must click the ok button, as the chat
window already overrode it's confirmation action), this commit allows for
both the chat input and the NPC dialog confirming to share the same key,
but allows for people to change one or the other if they want to. This
will still allow for the player to use the keyboard fully for NPC
dialogs, but allow for people who don't like not being able to use chat
with an NPC dialog open to assign confirmation to another key and get
their way as well.
This is a rather ugly way of resolving both sides of the issue, but it's
the only way to appease both sides. The Aethyra players want things the
way they were, while Bjorn wanted things a different way.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
strings.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
complaint commit d2876ad85935c5d0b181d050bd4c19a4682123b1 (btw, if
you're concerned about that, return is also a bad choice for the chat
window in general, which overrides the GuiChan default for use widget.
Tab would also be a bad choice, as it cycles through the widgets), next,
I made the text displayed by the keyboard configuration setup dialog
translatable, and lastly, I exposed the rest of the previously
unconfigurable keys to be configurable through the keyboard setup tab.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
scrolling through the table upward, while the other didn't check upper
bounds for rows and columns to see if the event is in the table, but
rather just the lower bounds.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
to the input fields. Now the only NPC dialog which isn't completely
navigatable by the keyboard alone is the NPC list dialog (seems to
have regressed slightly recently).
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
has an inventory or an equipment.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
that uses it anyways.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Mazzeroth. At the moment, I think this is the last of the missing
pointer deletes, but if more are found, then I'll add them to the
proper classes and delete them.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
the same lookup code. This fixes left clicking on the last commit.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
mininum width for the text wrapping code to be the length of the
player's name instead.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
silly. Condensed the sizing code for speech bubbles to do the same for
speech bubbles of all sizes.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
set content size functions to use the width variable.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
that speech bubbles are always at least as long as the name is.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
extracted to the wrong directory.
|
|
commit. Also, while the overall performance improved a little, it
doesn't appear to be as drastic as the last commit was suggesting.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
include anything not needed by that specific widget or window. This
appears to have cleaned up system performance a bit on my current setup,
where it went from idling on 45% in game with opengl down to 30% now.
Also moved iptostring to the tostring header, as importing all of
network.h is a little overkill to use that function, and it goes along
with the basic functions that are in that header file anyways. TODO:
find out a way to get rid of warnings when a class doesn't use this
function.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
needed so much, I think that now that opacity actually works in tables
now, it looks a bit bettwe with the scroll area visible.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
to trust that the resolutions presented are legal resolutions to try.
This shouldn't be an issue, as the resolutions listed are pulled
straight from SDL, and it avoids showing the user a visual artifact.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
alpha values under SDL.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
as defining whether the given table is opaque or not.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
listbox.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
|
|
recorder over to the layout code (not that it helps much at the moment).
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
default font.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
subimage class so that when OpenGL isn't enabled, GUI opacities can
still be applied.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
updating.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
spots that could also be fixed up as well, but require other methods to
fix them (the filled color rectangles on the progress bars, as well as
the text on them as well, as an example).
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
when the client was last exited.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
even when there's already a saved window configuration.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
back to the chat class, where they really belong. Since the party class
parses its own suboptions, I left the suboptions within that class.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
enough to contain the map's name. Also fixed an unnoticed logic flip
that was done unintentionally. This class now manages to work
perfectly when the config file has no previous dimensional influences,
but still won't resize properly without removing the ability to resize
or getting rid of those stored configs. TODO: Find out why loading
window configuration for the default minimum width or height overrides
the Minimap's supplied dimensions it wants.
Signed-off-by: Ira Rice <irarice@gmail.com>
|
|
|
|
"no-c-format".
|
|
Signed-off-by: Ira Rice <irarice@gmail.com>
|