The Windows build was erroring immediately when trying to
initialize cpal because of changes in winit 0.20:
https://github.com/RustAudio/cpal/pull/348
Moved the cpal initialization to a separate thread to work around
this issue.
The premultiplied alpha was not properly considered when there was
a color transform on a bitmap. Now the shader unmultiplies the
alpha before applying the color transform, and the remultiplies it.
When a stream sound uses ADPCM compression, the ADPCM header is
included in each SoundStreamBlock (as opposed to stream sounds
in the other formats). This header wasn't being parsed, resulting
in corrupted audio (see https://homestarrunner.com/main12.swf).
Converts the Bitmap character to a proper display object. This can
be instantiated directly in a PlaceObject tag in SWFv9 movies,
compared to the previous versions which indirectly references
bitmaps from Shape tags.
Renamed `AudioBackend::play_sound` to `start_sound`, and this
also takes a `SoundInfo` parameter with the event sound settings
from the SWF file.
Desktop now obeys the loop and start/end point settings. Envelopes
are still TODO.
Gradients could render with uninitialzied memory (usually colors
from previous gradients). Now the interpolation value is clamped
in the shader to prevent out-of-bounds access of the gradient
colors array.
This will be cleaned up further when the rendering backend
switches to gfx-rs.
Run cargo update. Remove git dependency on jpeg-decoder (the latest
version was published to crates.io). Update swf-rs to latest to
disable smoothing in SWFv7 and lower (addresses #28).
lyon vomits a lot on various tessellations. Previously this would
bail out immediately for the entire shape, resulting in missing art.
Now we just skip that particular path, and the rest of the shape
should get tessellated as normal.
Player is now in charge of scaling/cropping/translating the content
to fit the viewport size supplied by the frontend.
Added backend::render::Letterbox, which stores the margin sizes
for letter/pillarboxing.
The current version of jpeg-decoder on crates.io (0.1.15) panics
when encountering a valid JPEG marker of length 2. Some SWF files
contained JPEGs with a COM segment of length 2. This is fixed
on the master branch of jpeg-decoder.
DefineBitsJPEG tags can have some extraneous bytes inside the
stream. Splice these out when decoding the JPEG.
TODO: Docs say this should only happen for SWF versions before 8?
Lyon seems sensitive to overflow on large coordinates, so passing
twips resulting in overflow panics on very large shapes.
Avoid panics on tessellation failure: instead, return an empty
mesh and log an error.
Switch from pixels to twips for internal calculations. swf crate
was updated to add a Twips wrapper-type. Ruffle will use this type
internally for transforms and shapes. This fixes some precision
errors in rendering (see heroes_of_cybertron.swf) and will match
what the Flash Player does internally more closely. (e.g.
DisplayObject x and y will be quantized to 1/20 pixel units). It
is up to the renderer to convert units from twip space to pixels
when rendering.
dependabot-preview[bot]