* The properties parameter is evaluated last. This is observable by
the order the `toString`/`valueOf` methods of the parameters are
called.
* Only `null` in the property list parameter configures all properties,
as opposed to `undefined`, `null`, numbers and booleans previously.
* Objects in the property list parameter are not handled specially.
Instead, they're also coerced to string and split by comma.
* tests: add tests for scroll
* avm1: implement scroll, maxscroll, bottomScroll
* chore: fmt
* docs: note that scroll is 1-based
* fix: non-word wrapped text with manual breaks is scrollable
* chore: move magic number to const
* chore: avoid mut with extra if
* chore: moving clamping behaviour into core
* refactor: eagerly compute line data
* fix: make scroll work when text is aligned right
* chore: clippy
* docs: add more information about line_data
* tests: add more test cases for scroll
The current version just doesn't make any sense.
The fixed version is akin to the `target.starts_with("_level") && target.len() > 6` line a bit earlier in this file.
Wire up the op so that it affects the quality setting, although the
setting is still unused by Ruffle.
This op will remember whether the stage was in `High`/`Best` quality.
Split out the "bitmap downsampling" flag in `Stage` so that we can
persist this state.
This interface handles returning a bitmap given an ID and is used
by the render backend to get the bitmap used for a bitmap fill.
This will allow for bitmap fills in the drawing API, as these will
manage their own list of bitmaps.
This also removes the MovieLibrary dependency from render backends
and will allow for better decoupling in the future.
* Rename movie_clip::ClipAction to movie_clip::ClipEventHandler.
* Store the swf::ClipEventFlag event flags that trigger the event
directly in the event handler. Previously we split up any event
that had multiple event flags into separate events. Now these
can be kept as a single event.
* Remove `MovieClip::has_button_event`, and instead store the
union of all event flags in `MovieClip::clip_event_flags`. This
will be useful for other cases in the future.
Matrices in an SWF file store their scale/skew components in
in 16.16 format (fbits).
Split `ruffle_core::Matrix` and `swf::Matrix`. `swf::Matrix` now
stores its data as `Fixed16` instead of immediately converting to
`f32`.
This add two knobs to the impl_custom_object! macro:
- `set(...)`, for using a specific prototype in the `set`
method;
- `bare_object(...)`, for objects that are convertible to a raw object
type.
When reading an SWF, search for FileAttributes and
SetBackgroundColor and return this along with the header data
because it's useful (in particular, the AS3 flag).
This allows for reduced boilerplate when defining native methods,
properties, and constants (but not values depending on other runtime
values, like built-ins objects and prototypes).
Also remove ScriptObject::force_set_function (replaced by DeclKind::Method)
Move `MovieClip::is_swf` flag to `DisplayObject::is_root`, and use
this flag to handle the behavior of `DisplayObject.root` crawling
upwards until it hits a top-most loaded SWF/Bitmap.
Simplify `root` and `stage` so that they don't have to consider
buttons. Instead, do some trickery to ensure the button's states
see the proper values of `parent`, `root`, and `stage` during
construction.
`parent` is now a bare accessor, `avm1_parent` returns what AVM1 thinks the parent should be, and `avm2_parent` returns what AVM2 thinks the parent should be.
Both InitArray and InitObject should bail-out without popping anything
off the stack when the elements/properties count is negative or greater
than or equals 2147483648.
Previously, if the arguments count was greater than the actual
stack size, then a stack underflow occurred which resulted in a
sequence of undefined values. That didn't match Flash's behavior.
Also, this prevents potential huge allocations that hang Ruffle.
In addition, num_args seems like it should use coerce_to_u32
(wraps at 4294967297). This also means that -1 ends up acting like
u32::MAX and would pop all values off of the stack.
Toss out any shared objects that contain ".." in the name
to avoid accessing files outside of the Ruffle data directory.
The DiskStorageBackend also will fail any requests with a ".."
component as an extra precaution.
Fixes#3961.
MovieClip.getBytesLoaded and getBytesTotal return the size of that
specific clip, even if it's not a loaded SWF. The previous logic
only returned the size of the parent SWF.
If the clip is an SWF, the uncompressed size of the SWF is returned.
Otherwise, the length of the tag list inside the clip's DefineSprite
tag is returned.
Clamp the index parameters to the length of the array. This fixes
`[].splice(1, 0, 'obj');` to add the object at the proper
position (0, because the start index 1 is larger than the length).
BitmapData.loadBitmap is not subject to the size limit. That is,
it can load larger bitmaps than the BitmapData constructor.
Fix panic for large bitmaps by not invoking the BitmapData constructor
directly. This should also improve performance.
Do the same for BitmapData.clone.
* Extract is_size_valid, which respects 3 different limits:
SWF<=9, SWF10 and SWF>=11.
* Delay is_size_valid check after all arguments coercing.
* Swap the transparency and fill_color parameters of init_pixels.
Use the proper types for ColorTransform:
* Fixed8 (8.8) format for multiplicative component
* i16 format for additive component
This matches the behavior of Flash (for example, alpha only changes
in units of 1/256).
If a child clip is named `_level0`, accessing `_level0` should
return the level and not the child clip.
Move `DisplayObject::get_level_by_path` to `StageObject`, and
change it to return an `Option<Value>`, and return
`Some(Value::Undefined)` if the path is a valid level path but
the level is not occupied. This causes get/set of `_levelN` to
be swallowed, even if the level isn't populated.
Children instances should shadow display object magic properties.
For example, a child named "_x" will be returned instead of the
parent's _x position in GetMember.
Rust nightly 4/13 allows f64::parse to handle "infinity", case
insensitive. This broke cases such as `Number("Infinity")`, which
should return `NaN` in AVM1.
Additionally, Rust will now print "-0" for negative zero, when
previously it would print "0".
* Return NaN for inf cases ("inf", "-Infinity", "+INF", etc.)
* Add a test for `Number("inf")` (this was also incorrect before
the latest nightly)
* Add a special case for zero in `f64_to_string` to ensure
that -0.0 gets coerced to "0".
For more info, see:
https://github.com/rust-lang/rfcs/issues/1074
`XMLNode.{appendChild, insertNode}` can now be used to move
child nodes from a parent to another.
There are two special cases:
- if the node is already a child of the destination, nothing happens
- if moving the node would create a cycle, nothing happens
Prototypes can be primitive, not just objects. However, TObject's
proto() and set_proto() work with Object and not Value. To allow
easier and more incremental migration from Object to Value, this
commit introduces a new pair of proto_value/set_proto_value functions
in TObject, preserving the existing proto/set_proto functions.
Both pairs have default implementations, so it's enough to implement
only one pair in each TObject implementation.
For now only ScriptObject uses the new pair, which fixes the behavior
of incorrectly coercing __proto__ to object.
Follow-up migration plan:
1. Replace all old proto/set_proto implementations by the new
proto_value/set_proto_value in all TObject implementations on a
one-by-one basis.
2. After proto/set_proto become unused, delete them and rename
proto_value/set_proto_value to proto/set_proto for more concise code.
The "this" argument should fall-back to the global object only for
undefined or null. Other primitives should be coerced to an object
rather than falling-back to the global object.
As a drive-by replace `Vec::new` by `Vec::with_capacity` in
`Function.prototype.apply` to minimize unnecessary reallocations.
relrelb