Some glibc declare write(2) with the attribute warn unused result.
So we now ensure that the calls to write(2) are successful. Otherwise,
we print an error and update errno accordingly.
Some glibc versions declare pipe(2) with a warn unused result attribute,
leading to compile-time warnings when pipe(2)'s return value is not
checked.
If pipe(2) fails, we now print an error and make the calling function
fail.
when i did the vpath transition i made a copy and paste error, that
might made a few people miss theire config and appear weird directories
~/elementary.elementary/ in theire homedirectory, sorry for that!
This finally enables looking up things from the current unit
rather than from a backing storage in the Eolian state. This
also means that the benefits of having a unit system will
finally be visible.
This makes sure variables are stored as well as types within their
respective units. Also, declarations are now refcounted just like
any other Eolian object.
Summary:
Again, ptr_null/nonnull were added in check 0.11, while we depend on
0.9.10.
Test Plan: Run make check
Reviewers: marcelhollerbach, cedric, felipealmeida
Reviewed By: felipealmeida
Differential Revision: https://phab.enlightenment.org/D5820
Summary: Among them, IO interfaces are needed by the new efl.Task.
Test Plan: Run make check
Reviewers: felipealmeida
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D5821
If we have a blacklisted gl implementation then we'll create a canvas,
it'll fail to be useful, and we'll free it - which calls
ecore_wl2_shutdown(), but then we'll also call ecore_wl2_shutdown()
a second time further down the failure path.
Take a bonus reference before freeing the failed canvas to keep refcounts
sane.
Eolian now properly merges all unit hashes (e.g. unit A has
children B and C, hashes of B and C are merged into A's hashes
in order to be able to look up B and C's contents via A) and
maintains a (potentially cyclic) children graph.
Not everything is yet added into all hashes, but at least the
building blocks are there.
if its a non-legacy loop handler that also binds an fd handler
internally (the 2 things are linked as a pair) then on loop del, dont
del the objet as the obj del should handle things elsehwere.
this is astart of the work for having a common task class/interface
between loops, threads ane exe's so the i/o is all symmetric and works
the same way between all of them as well as similarly for launching
and knowing when the exit etc. etc.
this is not final and not perfect, but it's a start. comments of
course welcome
odd - i found ecore fd handlers basically ignored hangups from the
other end so we never knew if the other end went or not... crazy. now
we at least have all the read/write/error flags on and the next read
should fail indicating the broken pipe etc. ...
@fix
when there is only one object in a manager, and this object has a
redirect manager, then right now nothing will happen and nothing will be
called on the redirect manager.
With this patch the redirect manager will be unset when the manager
cannot give a correct candidate.