Summary:
Values returned from C# Then callbacks must release ownership of the
underlying native value, so Eina code can clean it up nicely and avoid
the Wrapper flushing it early.
The same issue applied to the Async wrappers. In this case the value
passed as the Task parameter could be released by an `using` block
awaiting the value.
Also Future creation was then-ing the wrong handle.
Also add better exception messages.
Reviewers: vitor.sousa, felipealmeida
Reviewed By: vitor.sousa
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D9197
Summary:
Iterator and Accessors are views only, not owning the data they point
to.
Also updated the tests by handling some test data that were leaking.
Fixes T8036
Reviewers: vitor.sousa, felipealmeida
Reviewed By: vitor.sousa
Subscribers: cedric, #reviewers, #committers, segfaultxavi, q66
Tags: #efl
Maniphest Tasks: T8036
Differential Revision: https://phab.enlightenment.org/D9189
Summary:
C# `Eina.Value` now has builtin support for `EINA_VALUE_TYPE_OBJECT`.
To avoid ambiguity with the `Set` method overloads, explicit casting
operators were used for wrapping/unwrapping `Efl.Object` instead of
implicit ones like for other value types.
Thus, to initialize an `Eina.Value` from an object, you can use the
following:
`var v = (Eina.Value)myObj;`
Reviewers: felipealmeida, vitor.sousa, segfaultxavi, Jaehyun_Cho
Reviewed By: Jaehyun_Cho
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D9164
Summary:
This commit makes use of the `ownership,shared` and `ownership,unique`
events from Efl.Object in order to avoid the C# wrapper from being
collected while C code holds a reference to the object.
For example, creating a list of items in a for loop and attaching events to
them would fails without this commit, as the C# GC may collect the wrapper.
The basic idea is that we use a `WrapperSupervisor`, which is stored in
the Eo data storage, with a GCHandle allocated for the lifetime of the
underlying Eo object. This supervisor takes care of holding either a
weak C# reference (when in unique mode, allowing the wrapper to be GC'd)
or a hard C# reference (when in shared mode, making the wrapper
non-collectable while the Eo has extra references).
One limitation is that object graphs can leak if a shared object in the
graph - an Eo child for example - stores a hard reference to another
object in the graph as a C# field. In this example, this causes the
parent to always have a hard C# reference (from the child) as the child
is non-collectable due to the parent holding an Eo reference to it.
Depends on D8678
Test Plan: `ninja test` and `make test`
Reviewers: lauromoura, felipealmeida, woohyun, segfaultxavi
Reviewed By: lauromoura
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D9014
Summary:
dotnet's GC seems to be more agressive, showing some issues that usually
do not appear when running under Mono's.
This commit uses for free_cb's the same scheme we already use for
regular Eo refs, using a GC handle to keep the cleaning callback alive.
Reviewers: vitor.sousa, felipealmeida, woohyun
Reviewed By: vitor.sousa
Subscribers: segfaultxavi, cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D8593
Summary:
This commit mainly fixes errors caused by deallocating resources in the garbage
collector thread. Using `ecore_main_loop_thread_safe_call_async` to queue
resource deallocation in the main thread seems to solve it.
Also, some `efl_ref` calls are added in places they were missing, mainly
objects that unref in the destructor thus taking ownership if efl_ref is not
called.
Also fix improper resource deallocation in tests that were causing it to crash,
enabling it to call Efl.All.Shutdown again. This allocation and the deallocation
process was moved from the Eo class constructor to static class methods that are
called in the test 'set up' and 'tear down' methods.
Queuing resource deallocation in the main thread make it mandatory that tests
call `Efl.App.AppMain.Iterate()` if they want to check proper resource
deallocation (like TestFunctionPointers.set_callback_inherited_called_from_c).
Extras:
Remove duplicated declaration of 'eflcustomexportsmono' in meson in order to fix
some linking problems.
Remove some unused code around deallocation functions that had to be reworked.
Object allocation is now supplied with the call site information it expects
(file name and line for _efl_add_start).
Depends on D8550
Test Plan: meson test
Reviewers: felipealmeida, lauromoura, cedric, segfaultxavi
Reviewed By: lauromoura
Subscribers: segfaultxavi
Tags: #efl_language_bindings, #do_not_merge
Differential Revision: https://phab.enlightenment.org/D8431
Summary:
Abstract Eo classes are now proper C# abstract classes.
As a side effect, returning Eo instances from native code was reworked
to return instances of their actual Eo classes instead of previous
behavior of returning a generic Efl.Object and using static_cast.
Instead of `var window = Efl.Ui.Win.static_cast(widget.GetParent());`
Use `var window = widget.GetParent() as Efl.Ui.Win;`
Another side effect was that `efl_constructor` was removed from the list
of supported `Efl.Object` overrides. It is invoked inside
`efl_add_internal_start`, before the bindings makes the association of
the newly created EoId with the C# instance that created it, making the
managed delegate meaningless. C# users then can use regular C#
constructors to initialize fields.
Also changed to set the private data of C#-inherited classes before the
call to constructing methods (aka constructor parameters) so C# classes
can override them correctly.
Fixes T7778
Fixes T7757
Reviewers: vitor.sousa, felipealmeida, segfaultxavi
Reviewed By: vitor.sousa, segfaultxavi
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7778, T7757, T7702
Differential Revision: https://phab.enlightenment.org/D8550
Summary:
Added basic documentation for things that were missing.
Some other files are silent due to a pragma disabling CS1591. They
should be handled later.
Also, removed `Efl.Io.Positioner` from the blacklist as it is referenced
from the `Efl.Io.Reader.eos` event documentation.
Reviewers: segfaultxavi
Reviewed By: segfaultxavi
Subscribers: felipealmeida, cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D8370
Summary:
This commits adds dotnet as a supported C# platform for EFL# bindings.
Due to differences between Mono and Dotnet regarding DllImport, the
bindings now are using an imperative approach to load the function
pointers through the NativeModule and FunctionWrapper classes. These
classes handle the dlopen/LoadLibrary and dlsym/GetProcAddress calls.
Also, the previous caching of non-owned strings returned to native code
was removed until further memory checks.
We also had to create workaround for bool and chars in Structs for C#
marshaling. Going through System.Byte instead and Marshaling manually
to their respective types.
In order to actually build efl_mono.dll with dotnet right now,
issue #4782 from Meson should be fixed to make it properly detect and
used the Dotnet compiler. Also use "-Ddotnet=true" when running meson.
Fixes T7394
Reviewers: felipealmeida, vitor.sousa, bu5hm4n
Reviewed By: vitor.sousa
Subscribers: cedric
Tags: #efl
Maniphest Tasks: T7394
Differential Revision: https://phab.enlightenment.org/D8069
Efl.Class (in practice, the return from the *_class_get() functions) can
be used as argument to functions, like in Efl.Object.provider_find and
Efl.Ui.Widget_Factory.item_class(get/set).
This commits adds support by representing Efl.Class instances
as System.Type in the C# API, allowing someone to do things like:
`factory.ItemClass == typeof(MyFramework.MyButton)`
It also supports user-defined classes that inherit from efl classes.
Summary:
It was marshalling erroneously data into and out of arrays and lists.
Instead of passing data by value (or by address of correct size), it was
stuffing data into IntPtr and trying to parse out afterwards.
This commit changes the binding to use the same approach of plain
Get/Set, with proper overloads.
Reviewers: vitor.sousa, segfaultxavi, felipealmeida
Reviewed By: vitor.sousa
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D8057
this removes the need for the calling a Init function.
Reviewed-by: Felipe Magno de Almeida <felipe@expertisesolutions.com.br>
Differential Revision: https://phab.enlightenment.org/D7556
Summary:
Previously, any unhandled Eina_Error would cause an exception
to be thrown when the control returned to C#.
This commit changes this behavior to only raise it when an exception
went unhandled from a C# callback back to C, like in an event handler,
for example.
Test Plan: run tests
Reviewers: segfaultxavi, Jaehyun_Cho, felipealmeida
Reviewed By: Jaehyun_Cho
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D7537
Summary:
new Eina.Value(0) is a special case. The 0 is silently converted
to an enum (Eina.ValueType) and therefore the call is ambiguous
with the 0 being first converted to an Eina.Value via the implicit
conversion operator (calling the Eina.Value deep copy constructor).
Adding constructors for all supported types solves the problem because
they have higher priority. Also, they avoid one deep copy of the
Eina.Value.
Includes test case to catch this problem in the future. This was discovered
in the tutorials, where new Eina.Value(0) is being used.
Test Plan:
The src/efl_reference_core_event.exe example from the examples repo was
not compiling before, and now it is.
make check and make examples still work as expected.
Reviewers: lauromoura
Reviewed By: lauromoura
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D7598
Summary:
For basic types, this will make it easier to pass Eina.Values into
functions, without requiring to setup and later Set() or Get() calls.
As discussed on irc, this seems to be a better way to improve the Value
C# API than using method chaining.
Fixes T7388
Test Plan: run tests
Reviewers: segfaultxavi, felipealmeida
Reviewed By: segfaultxavi
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7388
Differential Revision: https://phab.enlightenment.org/D7526
This brings in the possibility to receive the app object from bindings.
With the app object you can listen to pause / args / terminate / resume
events.
fix T7509
Differential Revision: https://phab.enlightenment.org/D7480
Summary:
As discussed in T7204:
- Eo Interfaces/mixins -> C# Interfaces with concrete class
implementations
- Eo Regular/Abstracts -> Proper C# classes
- Added some new generators and helper methods.
- Refactored the class generator, splitting into helper methods
Eo handles now are stored only in the "root" class in any given
inheritance tree (generally, Efl.Object), and accessible to each child.
Methods also are defined in a single place instead of repeatedly
generated in everyfile, reducing the size of the generated .dll from
30MB to around 4.5MB.
Mixins are generated as C# interfaces but any regular class it inherits
from is lost, as we can't have interfaces inheriting from regular
classes. This will be dealt with in a later commit.
Summary of API Changes:
- Merged Inherit/Concrete classes. (These suffixes disappear from
regular classes).
- Interface still have implementations with 'Concrete' suffix for when
they are returned from methods.
- Removed 'I' from interface names.
- Removed interfaces for regular/abstract Eo classes.
- Concrete classes for interfaces/mixins hold the event argument struct.
- Removed '_' from classes, enums, structs, etc, as indicated in C#
naming conventions.
- Namespaces are now Camel.Cased.
- Renamed IWrapper's raw_handle/raw_klass to NativeHandle/NativeClass
Also renamed the test classes as after the namespace change, the
test namespace Test can conflict with the helper Test namespace.
(And use more meaningful names than Test.Testing...)
Also Fixes T7336 by removing a deprecated example and adding
efl_loop_timer_example to build system.
Fixes T7451 by hiding the class_get DllImports and renaming the IWrapper
fields. The native handlers are used in the manual binding.
Still need to work:
- As there are still some events names clashing (e.g. Efl.Ui.Bg with "resize"
from Efl.Gfx.Entity and Efl.Gfx.Image), Events are currently declared on
the interface and implemented "namespaced" in the classes,
requiring the cast to the interface to access the event.
- The Mixin Conundrum. Mixin inheritance will be dealt in a future
commit.
Depends on D7260
Reviewers: segfaultxavi, vitor.sousa, felipealmeida, Jaehyun_Cho
Reviewed By: vitor.sousa
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7451, T7336
Differential Revision: https://phab.enlightenment.org/D7262
The tests are added and build. For running C# code please see the wiki.
you can enable -Dmono=true
Differential Revision: https://phab.enlightenment.org/D7203
Summary: This will allow deep copying the given value from C#.
Reviewers: felipealmeida, vitor.sousa, segfaultxavi
Reviewed By: vitor.sousa, segfaultxavi
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D7017
Summary:
eina_list_reverse() returns the list, the C# version should do the same.
This fixes the reference/csharp/eina/src/eina_list.cs example.
Fixes T7112
Test Plan: The eina_list.cs example works after this change.
Reviewers: lauromoura, vitor.sousa, felipealmeida
Reviewed By: vitor.sousa
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7112
Differential Revision: https://phab.enlightenment.org/D6995
Summary:
Using malloc/free as it was used before would cause double frees and
other issues when mixing with eina_values created from the value
mempool inside Eina.
Fixes T7359
Reviewers: felipealmeida, vitor.sousa, segfaultxavi
Reviewed By: vitor.sousa
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7359
Differential Revision: https://phab.enlightenment.org/D6958
Summary:
Plain conversion to IEnumerable, which is the base of LINQ
Depends on D6189
Reviewers: felipealmeida, vitor.sousa
Reviewed By: vitor.sousa
Subscribers: cedric, #committers, zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6190
Summary:
As futures are tied to a given promises and are usually handled in
a 'attach and forget' scheme, we can't cancel them as it would cancel
the whole chain.
Reviewers: felipealmeida, vitor.sousa
Reviewed By: vitor.sousa
Subscribers: cedric, #committers, zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6189
Summary:
Besides the normal methods returning Futures, we now generate
a wrapper with the "Async" suffix. This wrapper returns a
Systems.Threading.Tasks.Task which can be awaited on and reflect the
status of the Future.
When an eina.Future fails with ECANCELED, TaskCanceledException is
raised in the Task. Otherwise, an efl.FutureException(eina.Error) is
raised.
Depends on D6174
Reviewers: felipealmeida
Reviewed By: felipealmeida
Subscribers: cedric, zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6175
Summary:
Promise/Future cleanup:
In the promises, we use a wrapper Eina_Promise_Cancel_Cb to
invalidate the wrapper if it ever gets cancelled from outside. When
invalidating from C#, we can do it directly.
For the futures, likewise, in order to be able to invalidate the
wrapper when the chain it belongs to gets resolved we then() an
internal future with a callback to invalidate the wrapper we return to
C#. The return of this intermediate then() is the future we actually
return to the user.
Also added ECANCELED to the list of default eina.Errors
Depends on D6173
Reviewers: felipealmeida
Reviewed By: felipealmeida
Subscribers: cedric, zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6174
Summary:
eina.Value.Empty now means that we have an zeroed (empty) eina value.
For optional values that are empty use eina.Value.OptionalEmpty.
This was required to support the empty values passed with
EINA_VALUE_EMPTY in some Ecore futures.
Also, returning an eina_value by value is not supported in eolian
for safety reasons, so we removed some tests that tried to use this
behavior.
Depends on D6171
Reviewers: felipealmeida
Reviewed By: felipealmeida
Subscribers: cedric, zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6172
Summary:
Due to the absence of typedef from C#, we generate thin structs with
implicit operators to allow reference the data from their typedef'd name
from C#.
The other alternatives would be always converting to the lowest base on
the alias stack (losing the meaningfulness of the typedef name) or using
the 'using' directive. The latter has the restriction that it makes an
alias visible only in the file they are declared.
Reviewers: felipealmeida, cedric, segfaultxavi
Reviewed By: segfaultxavi
Subscribers: zmike
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6157
Summary:
Now the generated classes use an approach more familiar to C#
developers:
Interfaces: efl.Object -> efl.IObject
Concrete (implementation) classes: efl.ObjectConcrete -> efl.Object.
During this change, some methods that could clash with the
implementation class name (CS0542) had the prefix "Do" added (like in
efl.Duplicate.Duplicate() and efl.Pack.Pack()).
Depends on D6049
Reviewers: felipealmeida, vitor.sousa
Reviewed By: vitor.sousa
Subscribers: cedric
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D6050
When we have an eina.Value_Native (representing an Eina_Value passed by
value) and assign it to an eina.Value (a class with an IntPtr to an
underlying Eina_Value) we copy it so the eina.Value can take ownership
and free the data normally.
A possibly better alternative would be adding an extra flag to
eina.Value (something like OwnsPointer) to check whether we should free
the struct we point to or not.
When transferring the ownership of a ptr(struct) from Unamanaged to
managed, we should marshal the reference-typed fields or they can point
to Managed memory that would make no sense to access from C (for
example, strings would return garbage).
The downside is that it can cause potential leaks if the receiver of the
memory do not free it. In the current state of the EFL api this can
happen in two places:
- Efl.Ui.List.SegArray::remove
return -> ptr(Efl.Ui.List.LayoutItem) @owned
- Efl.Ui.Focus.Manager::fetch
return -> ptr(Efl.Ui.Focus.Relations) @owned
The resources copied by both structs may leak when those functions are
overriden in C# and the values returned to the C code.
Also hide some internal stuff instead of exporting it and generate
implicit conversion operators for struct external/internal.
This commit adds the "documentation" generator, which gets the
documentation_def attribute of the given item and generates xml comments
to be exported by MCS.
For items requiring some customization of the generated comments (e.g.
functions and its parameters), the helpers to generate the preamble
(summary), body (paragraphs) and epilogue (currently just the @since
tag) were added.
Currently we do not support converting Eolian references into xmldoc
references.
As we explicitly generate Get/Set methods for properties, for now the
generator tries to get the get/set specific documentation first. If it
is not present, fallback to the common docs.
Later this could be changed to generate the common one as paragraphs of
the Get/Set.
Also some generated code like the wrappers for calling C# methods
from C can be private. This will cleanup the introspection results
and warnings when generating documentation.
Due to this visibility change, the binbuf tests had to be changed
to add redirect calls to the native methods instead of directly
calling the DllImport'd methods.
Fix several integer binding type deduction based in its size on C.
Generation for function pointers no longer use modified argument name
which is different from the parameter name.
New generation context for structs.
bool from UnmanagedType.I1 to UnmanagedType.U1 (correct use
inside structs according to mono documentation).
byte (signed char) and int8 now is correctly represented by
sbyte in C#.
Check parameter direction in some out generators in parameter.hh.
Add efl_libs.csv to gitignore.
Make eina.Value pointer constructor public.
Add missing fields to efl.kw_event.Description struct.
Remove eina.File workaround (let struct gen handle it).
Remove is_function_ptr bool from regular_type_def and
add a typedecl_type enum to it. Also add some helper
methods for easier comparison.
Left some test cases commented for when pointer parameters
are properly working.
Instead of messing around with varargs, create individual wrappers for
each type supported.
The va_list approach was getting problems with float/double on Windows.
Lauro Moura
Felipe Magno de Almeida
Vitor Sousa
Jaehyun Cho
Xavi Artigas
Marcel Hollerbach
Vitor Sousa