41c4c41eaa
fix apparent logic error for MidiRegionView when redisplaying note model
2014-03-06 23:16:42 -05:00
2a6fc5af74
tweak Polygon::render() to work whether or not fill or outline are unset
2014-03-06 23:15:59 -05:00
feebabb330
trivial API change for MidiRegionView::update_hit()
2014-03-06 13:27:51 -05:00
2784ce99e1
comment edit
2014-03-06 13:27:26 -05:00
72517680cc
use CANVAS_DEBUG_NAME instead of explicit #ifdef to set canvas item name
2014-03-06 13:26:55 -05:00
37de3e2f67
make Hit (percussive note display item) actually draw something and fix up its coordinates
2014-03-06 13:26:36 -05:00
7d17db09ac
when in smart mode, context-click on a region item should behave as if in object mode.
...
Specifically, select the region before displaying the context menu
2014-03-06 10:40:31 -05:00
7bb305b8ae
separate ArdourButton::set_colors() and ArdourButton::build_patterns()
2014-03-06 09:53:38 -05:00
f9dbc28086
add ::on_name_changed() to ArdourButton, to use recently added ::on_name_changed() in CairoWidget
2014-03-06 09:49:00 -05:00
e4e6010cd4
add on_name_changed() virtual method to CairoWidget
...
If a CairoWidget does not a GtkRC-defined style, then changing its name does not trigger on_style_changed(). Since we want to use CairoWidget::set_name()
to trigger changes in the rendering of a widget, this is ... bad. Adding on_name_changed() provides a workaround for that.
2014-03-06 09:44:33 -05:00
4a915ee541
Add ArdourButton::set_fixed_colors() (equivalent to ::set_bg_colors() in Valeriy's original changes)
2014-03-06 08:05:33 -05:00
b360274e34
revert 29b1e706dd and then redo separation of ArdourButton::set_colors() from ArdourButton::build_patterns.
...
More to come
2014-03-05 17:51:51 -05:00
0c9120de8f
do not show region trim handles unless in (effective) object mouse mode
2014-03-05 14:40:49 -05:00
6605b2a542
when an automation/region gain line is hidden, so are its points
2014-03-05 14:40:49 -05:00
142439b9a8
no dot prefix for VST cache files on windows.
2014-03-05 19:44:01 +01:00
a08c0ea1da
manually revert change to canvas expose handling in fee026c5ef
...
Breaking out the rectangles implies a z-axis ordering of drawing,because the rect drawn last will implicitly be "on top".
But redraw areas are not submitted with any z-axis information, and so drawing like this breaks canvas layering. It
would be more efficient to draw the rects separately, but we don't have any ordering information and so we cannot do it
correctly.
2014-03-05 13:12:18 -05:00
983bf548cd
alter name of AutomationLine::show() to much more sensibly named ::update_visibility()
2014-03-05 12:11:22 -05:00
f5233f11a2
remove debugging output
2014-03-05 11:47:58 -05:00
58a30da03d
use a centripetal catmull-rom curve to smooth ArdourCanvas::Curve
...
See http://en.wikipedia.org/wiki/Centripetal_Catmull-Rom to understand the benefits of this.
2014-03-05 11:38:30 -05:00
f3300ec03c
move vst detection code into ARDOUR namespace
2014-03-05 08:47:54 +01:00
dd1c397344
clean up and document VST scanner src, add error logging to file
2014-03-05 08:41:29 +01:00
e6849e9c7e
"correct" curve drawing (no artifacts during redraw)
...
Interpolation errors still need addressing.
2014-03-04 21:58:17 -05:00
c29d392b0c
tentative intermediate state for ArdourCanvas::Curve
2014-03-04 21:58:17 -05:00
dde4d3bbc3
only queue an item redraw from Item::show() or Item::Hide() if the call changes the visibility status of the item (i.e. make no-ops really be no-ops)
2014-03-04 21:58:17 -05:00
836e95d964
fix/adjust logic used to display automation lines
2014-03-04 21:58:17 -05:00
6863bb9d28
reduce number of points used to define x-fade curves
2014-03-04 21:58:16 -05:00
fee026c5ef
break down GdkEventExpose into distinct rectangles for canvas expose rather than drawing the entire region as a single rect
2014-03-04 21:58:16 -05:00
1c839ed541
some more mingw x-compile fixes
2014-03-04 18:53:43 +01:00
912f07b919
mingw build fixes (tested with i686-w64-mingw32 on linux-x86_64)
2014-03-04 15:25:58 +01:00
3e250d5f06
even more small mingw/win fixes
2014-03-03 18:44:30 +01:00
5cfdb7d1ce
yet another mingw/win fix
2014-03-03 18:39:30 +01:00
43421f0b46
minor fix for libpbd wscript on windows
2014-03-03 10:31:53 -05:00
5db9599e7e
some windows build fixes
2014-03-03 10:31:53 -05:00
9952f774e9
some mingw related tweaks to the build-environment.
2014-03-03 16:09:14 +01:00
6442f316d4
detect VST midi-in/midi-out ports separately.
2014-03-03 13:35:09 +01:00
76fb7278dd
paths-dialog gtkmm-2.20 compatibility (part two)
2014-03-02 22:51:04 +01:00
f303c9ebb2
paths-dialog gtkmm-2.20 compatibility
2014-03-02 22:38:29 +01:00
52b127a35b
implement VST midi-output
2014-03-02 21:18:52 +01:00
e6c3cece64
remove recently added conflation of mingw64 and msvc with respect to symbol export
2014-03-02 14:59:08 -05:00
86ad316180
fix missing const for windows vst_search_path()
2014-03-02 12:46:33 -05:00
40c426fc5d
install small-splash image
2014-03-02 12:24:16 -05:00
deb48ff22d
add missing extra file
2014-03-02 11:33:35 -05:00
3b5cc5a9fd
move fixup_bundle_environment() code into per-platform files
2014-03-02 11:30:51 -05:00
8040df1897
move export_search_path(), previously in gtk2_ardour/main.cc, into libs/pbd
2014-03-02 11:25:54 -05:00
b8419ae7cd
use "new" VST search path code and adjust build system to reflect new source files
2014-03-02 10:53:25 -05:00
8ba449bc4e
remove fallback_folders code, add simpler more focused get_vst_search_path() and breakout get_win_special_folder() into its own file
2014-03-02 10:52:52 -05:00
c10e11eca5
for windows build, add fallback_folders.cc to libpbd source list
2014-03-01 22:47:19 -05:00
fc88eaf392
add likely stub code for adding linkage to libole when building libpbd on windows.
...
I may have gotten the precise linkflags string incorrect, but this my best guess.
2014-03-01 22:44:56 -05:00
fd64bba1de
add -DPROGRAM_NAME to libpbd compile flags
2014-03-01 22:43:46 -05:00
08c89809ee
REALLY REALLY CORRECT template export macros for libpbd
2014-03-01 22:31:47 -05:00
