Update API usage, incorporate new features, and refactor a bit:
- Use new progress callback to show progress bar
- Use new frame info functions to get timecodes/keyframes instead of
manually decoding
- Use new tracklist API to show track selection dialogs
- Expose option of whether to apply RFF
- Show a warning when bestsource had to revert to linear decoding
- Reuse the swscale context across frames and ensure the format
stays constant
- Add xxhash wrap
- No longer mark the bestsource video provider as slow
[src\meson.build] Add DirectWrite has dependency
[src\font_file_lister_gdi] Rework GDI FontCollector to use DirectWrite
This replaces all the logic of using the Windows registry to obtain the font path by using DirectWrite. The goal is simply to improve the quality of the code. This doesn't change any functionality
[src\meson.build] Remove Uniscribe has dependency
Uniscribe was only used for the FontCollector. Since we now use DirectWrite, we don't need it anymore.
[src\dialog_fonts_collector] Catch exceptions that FontCollector may raise
On Windows, the initialization of the FontCollector can raise an exception
[src\font_file_lister] Document the exception that GdiFontFileLister can throw
[src\font_file_lister_gdi] Correct possible memory leak when an error occur
Fix error caused by AddFontResource on Windows 10 or higher
[meson.build] Replace add_project_arguments with conf.set for HAVE_DWRITE_3
[src\dialog_fonts_collector] Update message error and optimisation
[src\font_file_lister_gdi] Correct documentation typo
[src\font_file_lister_gdi] Cosmetic nit - Initialize hfont in one line
[src\font_file_lister_gdi] Cosmetic nit - Remove if statements brace
[src\font_file_lister_gdi] Replace WCHAR param of normalizeFilePathCase to std::wstring
[src\font_file_lister_gdi] Replace WCHAR by std::wstring
[src\font_file_lister_gdi] Use IDWriteFontFace::GetSimulations to detect fake_italic/fake_bold
See this comment: https://github.com/arch1t3cht/Aegisub/pull/107#issuecomment-1975229652
[src\font_file_lister_gdi] If Win7/8 has Win 10 SDK on compile time, correctly verify if font has character(s)
With the Visual Studio 2019 toolchain on Windows 7, it installs the Windows 10 SDK by default. Because of this, ``HAVE_DWRITE_3`` is true, so the ``QueryInterface`` always fails. Now, if the ``QueryInterface`` fails, we try to verify if the font has characters with a Windows Vista SP2 compatible code.
[src\font_file_lister_gdi] Support facename that contains only whitespace AND truncated facename
Problem 1:
Previously, if a user wrote "\fn ", it would return the font Arial, which is not what we want. This is because when we request EnumFontFamiliesEx with whitespace or an empty lfFaceName, it will enumerate all the installed fonts.
Solution 1:
To resolve this issue, let's implement a solution similar to libass to determine if the selected facename exists: 649a7c2e1f/libass/ass_directwrite.c (L737-L747)
Problem 2:
GDI truncates font names to 31 characters. See: https://github.com/libass/libass/issues/459
However, since I changed the method to determine if a facename exists, I ensured that we still support this "feature".
To test this, I used the font in: https://github.com/libass/libass/issues/710
[src\font_file_lister_gdi] Add a FIXME comment regarding the utilization of std::wstring over WCHAR
[src\font_file_lister_gdi] Add FIXME comment about charset
This can be useful when scripts want to read the entire frame
and pass it on somewhere else (say, write it to a file or use some
library function on it).
This still breaks when the 0 digit is narrower than some other
digit, but that's unlikely and the current behavior now matches
the original behavior when all times are < 10h.
Fixesarch1t3cht/Aegisub#111 .
This is what users will want in the majority of cases, and switching
to this makes using this function much easier.
However, this does break backwards compatibility. Luckily to my
knowledge this function is not actually used in any existing published
script (all scripts using get_frame just use getPixelFormatted instead)
so the damage shouldn't be too large. But this is also why I'd rather
rip off the band-aid of breaking backwards compatibility now than later.
When destroying a frame handle, the previous logic would copy the
shared_ptr in the userdata and then free it twice (once explicitly
and once at the end of the function), which is actually UB, even
if if worked fine so far. This commit now ensures that it's the
actual userdata's shared_ptr that's freed in the gc function.
An unfortunate consequence of this is that the grid widths have to
be computed on every change to line timing, which will probably
have an impact on timing performance when autocommit is enabled.
It's probably not a huge impact but it's worth keeping an eye on.
All current renderers and muxers have no issues with longer timestamps.
However, classic VSFilter keeping track of time with a 32-bit int,
leads to a more meaningful limit at 596h, so enforce that instead.
Since the stringification code for SRT and ASS (but not SMPTE) relied on
hours being single digit, replace them with a simple format string.
GUI inputs are still limited to <10h, but at least
pre-existing timestamps no longer get mangled.
The vfr class (understandably) does not like timecode lists with only
one element, so the current default script breaks when opening images.
So to not bloat the default script too much we just ignore
__aegi_timecodes in the provider in these cases.
