what is happening is that a file gets announced through eio_model
listing code, at this point of time, the monitor does not yet know about
the file. If the file now gets deleted between the annoncing and the
learning of the file from the monitor, then the file got an ADD event,
but no DEL event. Which is a bug.
With this commit there is a new API which asks the monitor if the file
already has the knowledge about the files existance, or not. A few
monitors like win32 inotify or cocoa do not have context about the file
directly, if the OS is now having the same bug, then we are again in
trouble, however, we canot do anything about that. In the case of kevent
or poll, this asks the context of the monitor if the file is already
there.
Reviewed-by: Cedric BAIL <cedric.bail@free.fr>
Differential Revision: https://phab.enlightenment.org/D10006
Summary:
this logic brings a few problems:
- When files are annonced here, without the monitor having the context
of the files, immidiat actions like deletion of the file are resulting
in no deletion event. Which is definitly a bug, so we need to wait here
for the monitor.
- When a new count is annonced before the files are there from the
monitor, we are heading towards the same issue - additionally, we might
already have files available, due to the custom counting that are not
annoncned through the monitor, which means, at a later pointer we are
annoncing a new file which is already available via the API.
Right now i do not see a way to continue with this code, it caused major
issues with our CI and the POLL backend, for now things do work again.
Depends on D9624
Reviewers: zmike, stefan_schmidt, cedric, felipealmeida
Reviewed By: zmike
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D9625
i saw a segv in e/efm .. an eio call was called after it had been
canceled. this should never happen. ever. so ensure it does not once
an eio file has been canceled.
@fix
Summary:
when a monitor is destroyed, it unconditionally and immediately deletes
the monitor struct. this means that as soon as the monitor is dead, the
backend must never access the parent pointer again if its lifetime exceeds
the lifetime of the monitor (such as in threads)
the only member of the monitor data used by the fallback monitor is the
monitor path, so we can just copy it to the fallback data to avoid ever
needing to dereference this pointer
fixes reliability issues with efl sentry unit tests
@fix
Depends on D9708
Reviewers: cedric
Reviewed By: cedric
Subscribers: #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D9709
Summary:
if an event is emitted for a child that is added to the model during a
call to _efl_io_model_children_list(), it's possible that this child
will never be detected by the model's monitor/sentry if it is deleted
before the monitor can detect it, which means there will never be a
corresponding eio event emitted
in this case, ensure that we manually remove this child from the model
since we know we've just deleted it
this fixes reliability issues with efl io model monitor unit test
@fix
Reviewers: cedric
Reviewed By: cedric
Subscribers: #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D9708
Summary:
integrate mman.h to make Evil private to the EFL, as mman.h does not exist on Windows. After a discussion with raster, i include sys/mman.h only on non Windows platform.
One issue, though, is that src/modules/emotion/generic/Emotion_Generic_Plugin.h has inlined functions using mmap()
Test Plan: compilation on Windows
Reviewers: cedric, raster, zmike
Subscribers: #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D9542
do it once and remember the result from the first one. drops overhead
for sure by a chunk i actually could see in perf reports like about 1-2%
of cpu...
It kind of worked by luck for 6 months, but really shouldn't. This
is now a custom iterator that will walk the string part of the properties
only. It could become a generic iterator if necessary one day, but
for now, it is fine just here.
T7771
Reviewed-by: Marcel Hollerbach <mail@marcel-hollerbach.de>
Differential Revision: https://phab.enlightenment.org/D9075
Summary:
if the wref automatically removes this pointer then there's no need to
remove the wref on the pointer
@fix
Depends on D8973
Reviewers: cedric
Reviewed By: cedric
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D8974
Summary:
we need to to this here, otherwise we effectfily start the monitor
later, which means, we can miss a change in the filesystem. However,
this makes things a lot slower.
Reviewers: cedric
Reviewed By: cedric
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D8526
Summary:
Little introduction into what eio did before this commit:
Efl.Io.Model creation:
- direct ls of a directory in a thread (A)
- take all the contents of a directory and feed it slowly back into
the mainloop
- when all events have been feeded back to the mainloop and have been
processed: start monitoring (B)
However, any file created between (A) and (B) will not be in the model,
since not the listing nor the monitoring did caputure it. Hence we need
to start monitoring before we actaully start listing. In the callbacks
we then check if we already published something.
ref T7311
Reviewers: zmike, cedric
Reviewed By: cedric
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7311
Differential Revision: https://phab.enlightenment.org/D8525
Ecore_Thread excpect resolution of the error and done case to be instantaneous,
while Eina_Future default scheduler linked with Ecore main loop is build around
asynchronous answer. This create a lot of potential. A better patch would be
to provide an Ecore_Thread helper that does the integration properly. Sadly
we are in release now, so this is basically what an helper would do, but
contained inside Efl_Io_Manager.
This also solve the same problem as D7970 and D8053, but it should avoid its
side effect.
Reviewed-by: YeongJong Lee <yj34.lee@samsung.com>
Differential Revision: https://phab.enlightenment.org/D8371
There is no point in keeping a pointer to the main loop now that we
are using efl_future_then. This resolve potential bug with leftover
dangling weak reference as efl_future_then do require a free case
otherwise.
Reviewed-by: Marcel Hollerbach <mail@marcel-hollerbach.de>
Differential Revision: https://phab.enlightenment.org/D8374
Summary:
All legacy @since tags have already moved to the *.legacy.h files.
EO files are now devoid of @since tags (except some eldbus still
needed for legacy).
Upcoming patches will add @since 1.22 to those APIs which come out
of beta in this release.
APIs marked @beta do not need @since tags.
Test Plan: Everything builds, EO docs (like DocFX) have no Since tags.
Reviewers: zmike, bu5hm4n, lauromoura, cedric
Reviewed By: cedric
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D8430
Summary:
Since we're now going to be shipping some eo classes as stable,
there is no point in keeping the eo api behind a macro, and it
should be enabled by default. Another case is beta classes, but
those are behind the EFL_BETA_API_SUPPORT guard.
This also changes includes around the place where things are
clearly broken (such as an included header needing something
from another header but that other header being guarded, notably
efl_ui_widget.h needing focus manager but focus manager being
behind beta in Elementary.h)
Reviewers: zmike, cedric, bu5hm4n, stefan_schmidt, segfaultxavi
Reviewed By: cedric, segfaultxavi
Subscribers: segfaultxavi, #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D8322
Summary:
this is done because .eo files are not stable, and in order to stop
people depending on it, its better for now to disable the installation
of them for now.
ref T7676
Reviewers: stefan_schmidt, cedric, zmike, devilhorns
Reviewed By: zmike
Subscribers: #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7676
Differential Revision: https://phab.enlightenment.org/D7897
Summary:
This enables all the checks unconditionally, without ignoring
classes that don't have an Efl namespace. This required a lot
of beta marking to make it build. It most likely doesn't
mark types correctly, as that is not fully enabled yet.
Reviewers: zmike, cedric, segfaultxavi, bu5hm4n
Reviewed By: segfaultxavi
Subscribers: #reviewers, #committers
Tags: #efl
Differential Revision: https://phab.enlightenment.org/D8266
Summary:
these are all types that we do not currently want to release
Depends on D8102
Reviewers: segfaultxavi
Reviewed By: segfaultxavi
Subscribers: segfaultxavi, cedric
Tags: #efl_api
Differential Revision: https://phab.enlightenment.org/D8241
Summary:
This make sure that the object returned by children_slice_get are properly
destroyed when the refcount drop to only the parent holding a reference on
it. This make it clear that the user of the api can rely on efl_ref/efl_unref
to actually manage its use of the returned object.
Additionnaly we are cleaning up the created object that we are using to build our own
request inside the Efl.Io.Model and avoid internal leak.
Depends on D7864
Reviewers: felipealmeida, segfaultxavi, SanghyeonLee, zmike, bu5hm4n
Reviewed By: zmike
Subscribers: #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7528
Differential Revision: https://phab.enlightenment.org/D7865
if length of path arguments are longer than PATH_MAX,
there could be a BOF problem potentionally.
this patch fixes it.
Reviewed-by: Cedric BAIL <cedric.bail@free.fr>
Differential Revision: https://phab.enlightenment.org/D7919
Summary:
Instead of surrounding all the #include "*.eo.h" lines in Efl.h
with #ifdef EFL_BETA_API_SUPPORT, include these files unconditionally, but mark
all classes as @beta in the eo files.
This will allow taking them out of beta one by one as we deem them stable enough.
Otherwise, the current procedure involves moving the #include line out of the
ifdef block, which is cumbersome and messes include order.
Depends on D7950
Fixes T7692
Test Plan: Nothing changes
Reviewers: zmike, bu5hm4n, cedric
Subscribers: #reviewers, #committers
Tags: #efl
Maniphest Tasks: T7692
Differential Revision: https://phab.enlightenment.org/D7951
Summary:
Eolian adds a per-class BETA guard (like EFL_UI_WIN_BETA) to any method tagged
as @beta. This means that any app (and the EFL code) wanting to use BETA features
has to enable them class by class, which is cumbersome.
This commit replaces the individual guards with the global EFL_BETA_API_SUPPORT
guard, so apps only need to define one symbol to access BETA features.
Any usage of the per-class guards has been removed from the EFL code and examples.
When building EFL the global guard is defined by configure, so all EFL methods
already have access to BETA API.
Efl_Core.h and Efl_Ui.h no longer define EFL_BETA_API_SUPPORT. Apps wanting to
use BETA API have to define this symbol before including any EFL header
(It has been added to the examples requiring it).
Test Plan:
make && make check && make examples still work, but there's a lot less #defines
in the code
Reviewers: zmike, bu5hm4n, q66
Reviewed By: q66
Subscribers: cedric, #reviewers, #committers
Tags: #efl
Maniphest Tasks: T6788
Differential Revision: https://phab.enlightenment.org/D7924
Summary:
As the result of discussion in T7458,
we need to rename all efl_model based classes with efl_XXX_Model sequence.
I've run few vote for this, see V42, V43
few classes are totally renamed as our consideration of misnaming.
| Efl.Model_Loop | Efl.Loop_Model |
| Efl.Model_Item | Efl.Generic_Model |
| Efl.Model_Container | Efl.Container_Model |
| Efl.Model_Container_Item | Efl.Container_Model_Item |
| Efl.Model_Composite | Efl.Composite_Model |
| Efl.Model_Composite_Boolean | Efl.Boolean_Model |
| Efl.Model_Composite_Boolean_Chlidren | Efl.Boolean_Model_Item |
| Efl.Model_Composite_Selection | Efl.Select_Model |
| Efl.Model_Composite_Selection_Chlidren | Efl.Select_Model_Item |
| Efl.Model_View | Efl.View_Model |
| Eio.Model | Efl.Io.Model |
| Efl.Ui.Model_State | Efl.Ui.State_Model |
| Efl.Ui.Model_Size | Efl.Ui.Size_Model |
| Efl.Ui.Model_Exact | Efl.Ui.Exact_Model |
| Efl.Ui.Model_Average | Efl.Ui.Average_Model |
| Efl.Ui.Model_Homogeneous | Efl.Ui.Homogeneous_Model |
I worried about Efl.Io.Model changes, cause it is widely used,
but as I tested, there is no issue found yet.
Eldbus.Model also may can changed Efl.Dbus.Model,
but I cannot found any class who using Efl.Dbus namespace,
so I left it future work.
Test Plan:
Run the Make Test,
it works well except Efl test about selection.
there are class inheritance problem exist in select_model,
I reported it and will fix it in another patch.
Reviewers: cedric, felipealmeida, woohyun, Hermet
Reviewed By: cedric
Subscribers: lauromoura
Tags: #efl, #do_not_merge
Differential Revision: https://phab.enlightenment.org/D7533
If you need data, use a efl_future_then as done in every case here to get the same feature.
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Differential Revision: https://phab.enlightenment.org/D7577
eina_promise_data_set is a misleading API and result in conflicted use by
multiple independent piece of code leading to bug and crash. It is also not
necessary and we can avoid using it completely.
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Differential Revision: https://phab.enlightenment.org/D7572
This reverts commit 9b5155c9f1.
For now lets revert this, this breaks copy and paste, further more it
has the potential to break a lot more things, as eio_model tends to use
efl_loop_promise new, and then eina_promise_data_set, which is
explicitly forbidden.
This fixes crashing terminology instances.
I am not sure this is the right way to do it as binding would have to likely
to bind it manually.
Reviewed-by: Lauro Neto <Lauro Moura <lauromoura@expertisesolutions.com.br>>
Differential Revision: https://phab.enlightenment.org/D7492