| Age | Commit message (Collapse) | Author |
|---|
|
This commit implements line wrapping in the terminal, and tries its best
to move the cursor to the "correct" position.
|
|
|
|
|
|
When resizing a terminal window the number of columns may change.
Previously we assumed that this also affects lines which were in the
terminal's buffer while that is not necessarily true.
|
|
This is our way of implementing the [[ErrorData]] internal slot, which
is being used in Object.prototype.toString().
|
|
|
|
|
|
We had a cached shape for environment records to make instantiating
them fast. Now that environment records don't inherit from JS::Object,
we can just get rid of this. :^)
|
|
Previously, EnvironmentRecord was a JS::Object. This was done because
GlobalObject inherited from EnvironmentRecord. Now that this is no
longer the case, we can simplify things by making EnvironmentRecord
inherit from Cell directly.
This also removes the need for environment records to have a shape,
which was awkward. This will be removed in the following patch.
|
|
|
|
|
|
Our environment records are currently weird in that they inherit from
Object, but don't have a connection to the global object.
I'd like to remove this inheritance, and the first step is giving them
their own pointer to the global object.
|
|
This patch adds the concept of variable bindings to the various
environment record classes. The bindings are not yet hooked up to
anything, this is just fleshing out all the operations.
Most of this is following the spec exactly, but in a few cases we are
missing the requisite abstract operations to do the exact right thing.
I've added FIXME's in those cases where I noticed it.
|
|
eval only has direct access to the local scope when accessed through
the name eval. This includes locals named eval, because of course it
does.
|
|
For now this only allows us to single-step through execution and inspect
part of the execution environment for debugging
This also allows to run to function return and sending signals to the VM
This changes the behavior of SIGINT for UE to pause execution and then
terminate if already paused
A way of setting a watchpoint for a function would be a good addition in
the future, the scaffold for this is already present, we only need to
figure out a way to find the address of a function
On a side note I have changed all occurences of west-const to east const
|
|
|
|
|
|
Also make parameters static so they aren't in every node of the tree
this saves a substantial amount of memory.
|
|
before:
sizeof(Board)=344, sizeof(Move)=36, sizeof(Piece)=4, sizeof(Square)=8
after:
sizeof(Board)=108, sizeof(Move)=9, sizeof(Piece)=1, sizeof(Square)=2
|
|
This more than doubles the number of nodes that MCTS can search for a
given time limit, and greatly reduces memory usage.
|
|
"High score" should be two words, and this matches other games in the
system.
|
|
Previously, the high score was only in-memory, so only persisted for as
long as the FlappyBug window was open.
|
|
|
|
Steps to reproduce:
1. Start TextEditor and make some changes to the document.
2. Try to open an existing file.
3. When prompted choose to save the changes to the existing document.
4. Close the file picker by clicking 'Cancel'.
5. Note how the file was opened anyway and your changes were lost.
Same applies to the 'New File' action.
|
|
There's no point in saving the file - and potentially having to ask the
user for a file name - if the user abandons the 'Open' action by
clicking 'Cancel' in the file picker. This now also matches TextEditor's
behavior.
|
|
There was already 'Save As' and now we also have 'Save'.
|
|
This makes sure to ask the user whether they want to save changes to
their current document when opening a file even if the document has
never been saved before.
|
|
This adds access X_OK check in discover_apps_and_categories()
to see executable specified in .af files exist before
registering them for start menu.
|
|
The parser doesn't always track lexical scopes correctly, so let's not
rely on that for direct argument loading.
This reverts the LoadArguments bytecode instruction as well. We can
bring these things back when the parser can reliably tell us that
a given Identifier is indeed a function argument.
|
|
This was accidentally deleting a property from the object record object
itself, rather than from the object record.
It's quite confusing that Environment Records are objects, but moving
away from that will require some large changes.
|
|
This does break two TypedArray test262 test but those really were not
correct because hasProperty was not implemented
|
|
|
|
This fixes getting values from double proxies:
var p = new Proxy(new Proxy([], {}), {});
p.length
|
|
|
|
In get_own_properties:
Entries which are deleted while iterating need to be skipped
In PropertyDescriptor::from_dictionary
If the getter/setter is undefined it should still mark it as present
|
|
Most of the code is taken from put_own_property however the attributes
need to be handled slightly differently it seems
|
|
Mapping:
Set -> put (does not throw by default)
CreateDataPropertyOrThrow -> define_property
|
|
Now put uses is_strict_mode to determine define_property should throw
|
|
|
|
|
|
This was missing step 3 from the spec:
3. If Type(O) is not Object, return O.
Also use RequireObjectCoercible() for a better error message and make
the rest of the code a bit easier to read and more similar to the spec
text.
|
|
These represent the outermost scope in the environment record
hierarchy. The spec says they should be a "composite" of two things:
- An ObjectEnvironmentRecord wrapping the global object
- A DeclarativeEnvironmentRecord for other declarations
It's not yet clear to me how this should work, so this patch only
implements the first part, an object record wrapping the global object.
|
|
To better follow the spec, we need to distinguish between the current
execution context's lexical environment and variable environment.
This patch moves us to having two record pointers, although both of
them point at the same environment records for now.
|
|
This patch adds FunctionEnvironmentRecord as a subclass of the existing
DeclarativeEnvironmentRecord. Things that are specific to function
environment records move into there, simplifying the base.
Most of the abstract operations related to function environment records
are rewritten to match the spec exactly. I also had to implement
GetThisEnvironment() and GetSuperConstructor() to keep tests working
after the changes, so that's nice as well. :^)
|
|
According to the spec, [[HomeObject]] is an internal slot on function
objects, and should always be accessed through there.
|
|
This is so they can find their associated resources and it's
the same behavior as in Lagom.
This also required changing some tests so that they could
write their resources in a writable location.
|
|
If we consumed whitespace and/or comments after a RegexLiteral token,
the following token must not be RegexFlags - no whitespace or comments
are allowed between the closing / and the flag characters.
Fixes #8201.
|
|
Previously we'd send mouse events twice if the target window had
global cursor tracking enabled.
|
|
This replaces the naive copy algorithm that only supported rectangular
and 100% opaque selections with a more general approach that supports
any shape and alpha value.
Note that we now make a brand new bitmap with a hardcoded format instead
of just cropping the layer's existing bitmap. This is done to ensure
that the final clipboard image will have an alpha channel.
|
|
A "feather" value sets by how much the borders of the selection will be
smoothed, and a "mode" value sets how the newly selected region will
interact with an existing image selection (if any).
|
