message header into account. It worked before at some point as BLooper was
letting the looper handle messages to invalid targets - they are now just
dropped.
This fixes bug #93.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16045 a95241bf-73f2-0310-859d-f6bbb57e9c96
task_looper() again.
* removed BMessenger::fPreferred - whenever you had to specify "usePreferred" separately,
you don't have to do that anymore - use B_PREFERRED_TOKEN instead.
* fixed BTokenSpace::GetToken() semantics: it will no longer touch the "object" argument
in case of failure.
* Introduced a BWindow::_DistributeMessage() that will be part of the event dispatcher
counterpart to the app_server (the other will be _DetermineTarget()).
* Made it easier to use Michael's Message4 implementation: just add the following line
to your UserBuildConfig:
AppendToConfigVar DEFINES : HAIKU_TOP src : USING_MESSAGE4 : global ;
* Introduced ServerWindow::HandlerMessenger() and FocusMessenger() - the first will
target the client handler, while the other will target the preferred handler of the
client looper (usually the view having focus).
* Fixed dano message unflattening in the Message4 code.
* Changed BMessage::PrintToStream() to no longer use macros in the Message4 implementation.
* I hope that's all - it's a huge change, but it's all connected.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15046 a95241bf-73f2-0310-859d-f6bbb57e9c96
cleanup.
* made BMessageQueue::IsLocked() const - the non-const version is still provided
for binary compatibility.
* Both BMessageQueue::FindMessage() versions are now thread safe, the queue's BLocker
is now mutable to allow for this.
* renamed BMessage::link to fQueueLink as the "Message4" implementation uses it.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14955 a95241bf-73f2-0310-859d-f6bbb57e9c96
optimazation for ReplaceFlat() (also done for AddFlat() before).
Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13662 a95241bf-73f2-0310-859d-f6bbb57e9c96
I think BMessageBody::FindData<>, BMessageBody::AddData<>, BMessageBody::ReplaceData<> could be untemplatized
Works here, not unit tested though
Axel, please review this.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13653 a95241bf-73f2-0310-859d-f6bbb57e9c96
the magic is not part of the header anymore when read from a stream (because
BDataIO does not support seeking, and we need to know the magic before actually
reading the header).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13429 a95241bf-73f2-0310-859d-f6bbb57e9c96
Not really suited for message sending/retrieval (because target stuff is
missing), but should work more or less with disk based messages.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13428 a95241bf-73f2-0310-859d-f6bbb57e9c96
* In BMessage::Header::ReadFrom(DataIO &) the checksum was checked before
the fBodySize field was initialized, which could make
BMessage::Unflatten() fail.
* Calculate the header checksum now by writing the header to a buffer and
checksumming that. This is simpler and avoids problems when something
else is written than is actually stored in the header (e.g. when a
preferred target token is replaced by B_PREFERRED_TOKEN).
While the BMessage unit tests always passed, now the other App Kit tests
pass as well.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11130 a95241bf-73f2-0310-859d-f6bbb57e9c96
a separate class. This allowed to improve _SendFlattenedMessage() to
deal properly with flattened BMessages as well.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11116 a95241bf-73f2-0310-859d-f6bbb57e9c96
proper private static BMessage members and made them accessible through
BMessage::Private. Got rid of unused _reconstruct_message_().
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11106 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Removed 1500 lines of unused non-template-madness code. We have version
control to resurrect it, if we ever wish to.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11105 a95241bf-73f2-0310-859d-f6bbb57e9c96
a buffer is given.
* Added method for sending flattened messages. The KMessage support is
complete, but for BMessages this is a bit tricky. We currently unflatten
the BMessages and send those.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11060 a95241bf-73f2-0310-859d-f6bbb57e9c96
* In handle_reply(), in case the port message code is not as expected,
don't return B_OK and let the caller proceed with an uninitialized
BMessage, but return an error.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@10762 a95241bf-73f2-0310-859d-f6bbb57e9c96
preferred handler is addressed. So make sure, it has a usable value in
this case, otherwise ReturnAddress() usually won't work.
* If the message to be read from the port is bigger than the on-stack
buffer, the chunk of memory to be allocated on the heap has to be the
size of the message, not the size of the on-stack buffer. This caused
all kind of nasty, hard-trackable problems.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@10677 a95241bf-73f2-0310-859d-f6bbb57e9c96
messages which did not explicitely request a reply would fail. The member
is now initialized if an unflattened message has the MSG_FLAG_INCL_REPLY
flag set.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9193 a95241bf-73f2-0310-859d-f6bbb57e9c96
Added a ToDo item because gcc is right and itemSize may be used uninitialized;
someone with a better understanding of this code should fix it.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8233 a95241bf-73f2-0310-859d-f6bbb57e9c96
You don't have to check against NULL when deleting an object.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@6674 a95241bf-73f2-0310-859d-f6bbb57e9c96
Added calls to _init_message_(), _delete_message_(), and
_msg_cache_cleanup() to InitTerminateLibBe.cpp
Finished first implementation of BMessage::SendReply(), BMessage::_send_(),
and BMessage::_send_message()
Add BMessage to app.src, removed BBlockCache from support.src.
New BMessage::Private class has functions for twiddling BMessage internals
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@4371 a95241bf-73f2-0310-859d-f6bbb57e9c96
as the template param (where everything else was using void*) and
BMessage::AddData() was adding the pointer to the pointer as the data.
Unit tests save the day again!
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3385 a95241bf-73f2-0310-859d-f6bbb57e9c96
adding entry_refs and BMessages. Lots of stuff for specifiers and also for
sending replies (with thanks to Mike Nordell).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3295 a95241bf-73f2-0310-859d-f6bbb57e9c96
it. Also, minor fix in entry_ref handling in BMessage::AddData().
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2967 a95241bf-73f2-0310-859d-f6bbb57e9c96
out all over the place). There is now a giant
#ifdef USING_TEMPLATE_MADNESS
separating the two versions. At some point, the common code should be
reconciled.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2953 a95241bf-73f2-0310-859d-f6bbb57e9c96
Axel Dörfler