Tap-to-time provides the user the ability to tap to the lyrics/syllables
of the song in order to time lines or karaoke. It consists of these
extra UI interactions:
- **Indicator**: tap marker: a designated marker that can be moved to
the current audio position; indicated in:
- the audio display by a green arrow underneath a marker
- the karaoke display by a green-colored syllable
- **Control**: tap marker: the tap marker can be changed by selecting
syllables on audio display in karaoke mode, or clicking the markers on
audio display in dialogue mode
- **Control**: ctrl-right-click audio display: starts playing the audio
from that exact position until the end of the file
- **Option**: Timing/Tap To Time: enables the tap marker indicator and
commands
- **Button**: time_opt_tap_to_time: toggles the Timing/Tap To Time option
- **Button**: time_tap_connect (hotkey I): a command that:
- moves the tap marker's position to the current playing audio
position
- sets the next marker to be the tap marker
- if the tap marker is already the last marker AND BOTH autocommit AND
next-line-on-commit is ON, will move onto the next line
- if moved on to the next line, also sets the start marker to the current
audio position, so the two lines are connected, and moves to the
next tap marker (essentially reinvoking time_tap_connect once)
- **Button**: time_tap_no_connect (hotkey O): similar to
time_tap_connect, except it will not set the next line's start
position even if moved to the next line
Expected workflow:
1) User loads song lyrics
2) User splits each line into syllables
3) User turns on tap-to-time, autocommit, and next-line-on-commit
4) User plays audio from beginning, tapping time_tap_connect to each
syllable, occasionally tapping time_tap_no_connect when a break between
lines is desired
5) If user messes up a line, they can set the tap marker to where they
want to restart from, and ctrl-right-click to start the audio a few
seconds before it
6) Syllables can be split/merged at will, and adjustments to timing can
be done using normal karaoke timing controls
Not the perfect place for it by any means, but that's where the rest of
the application-global stuff is these days and it removed the need for
automation stuff to include main.h (and thus all the wxApp garbage).
It's modestly faster, significantly more type-safe, and doesn't assert
when there's too few arguments, which causes problems for plural forms.
Closes#1733.
Fixes errors when trying to automatically open audio from dummy video
and improves UX a little by disabling the Open Audio From Video menu
item when there are no audio tracks.
Most of the dialogs in Aegisub have a public interface consisting of a
single function, so there's really no need to expose the actual dialog
classes to the rest of the program. Add dialogs.h with the declarations
of all of those functions (most of which are just ShowFooDialog()), and
kill a million other headers.
Add a new Project class which is responsible for everything related to
opening and closing audio, video, subtitles, timecodes and keyframes.
This pulls almost everything not directly related to playing audio/video
out of the audio and video controllers, pulls more crap out of
FrameMain, and happens to make things a little simpler in the process.
Some of the toolbar buttons are seriously wrong when mirrored and there
doesn't seem to be a way to disable mirroring for the images while using
RTL layout for the toolbar itself, so double-mirror them so that they
end up with their original appearance.
Ideally the button images would be localizable as some of them actually
should be mirrored, but that's probably not worth the work involved.
See #1354.