fixed size flag at once. Not sure if this can be merged into one of the
existing GetInfo() variants without breaking binary compatibility.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@40374 a95241bf-73f2-0310-859d-f6bbb57e9c96
in BMessages. Also added type codes for them in TypeConstants.h. Closes ticket
#6302. Thanks!
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37511 a95241bf-73f2-0310-859d-f6bbb57e9c96
unarchiving protocol to support archival of arbitrary object graphs.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37431 a95241bf-73f2-0310-859d-f6bbb57e9c96
or we were using the previous form incorrectly. Anyway, this doesn't change
the functionality and fixes ticket #5676.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36052 a95241bf-73f2-0310-859d-f6bbb57e9c96
but only 2 were used, leading to various problems (like bug #5596).
I added another variation of print_type which handles that case.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35911 a95241bf-73f2-0310-859d-f6bbb57e9c96
known length to print the stream so printf() doesn't potentially crash when
trying a strlen() on the string. Fixes for example printing B_KEY_DOWN messages,
where such strings are used, in certain conditions.
Review welcome, printf() formatting is not one of my strengths. :-)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35629 a95241bf-73f2-0310-859d-f6bbb57e9c96
The size instead of the data would be printed. The fixed/non-fixed handling is
generic though, so no need to special case the always non fixed elements.
Since adding of basic types through their Add*() function always generates fixed
size fields, this would only be visible if the data was added through AddData()
with isFixedSize == false. This is the case in compiled resources for example,
hence you would see it when printing the returned messages of BMimeType::Get*()
functions for pre-installed types.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35336 a95241bf-73f2-0310-859d-f6bbb57e9c96
returns valid data. Since the data is still valid in the copy, we don't need
to clear the corresponding flag when copying a message. This is most probably
the same behaviour as in R5 as Tracker seems to expect that in the case where
a drop is done after a menu was displayed as in #4757. Thanks to aldeck for
tracking that one down inside Tracker, should fix#4757.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33746 a95241bf-73f2-0310-859d-f6bbb57e9c96
Net_server starts services by invoking fork() followed by exec(). If the latter
fails (for instance because the service isn't installed), the forked child is
invoking exit(). This in turn unloads libbe, triggering static cleanup code in
BMessage, which deletes a couple of message ports that were inherited from the
parent during the fork. After that, net_server was desparately missing those
ports and no longer worked reliably.
* in InitTerminateLibBe, we now register an atfork-(child-)handler, which
takes care to re-initialize the static reply ports used by BMessage code
* added BMessage::Private::StaticReInitForkedChild wrapper and
BMessage::_StaticReInitForkedChild() implementation which overwrites the
inherited port IDs with a set of own ports
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33050 a95241bf-73f2-0310-859d-f6bbb57e9c96
storage.
* Remove reduntant information from the header (field_count vs. fields_size).
* Remove checksums previously used to validate the message passing by area
mechanism.
* Move variables that are purely used by the instance out of the header and into
the BMessage object.
* Use more sensible types for the different message fields.
* Reduce some field sizes to realistic values.
* Make size_t values into uint32 values so the message format will not change
when later moving to 64 bits.
* Pack the structures used for flat message storage so it doesn't change
because of padding.
* Fix message passing by area. It never worked because the created area was
never actually filled with any data!
* Some more allocation checks with graceful fallbacks (should be all now).
* Some more checks for negative index values (should also be all now).
* Make printing more inline with how the rest of the class works and make some
of the output more consistent.
* Also add the new unsigned types to PrintToStream() output.
* Fix printing of unknown types and invalid BMessages, it would always have
printed only the first entry respectively the same error.
* Added some clarifying comments.
* Cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32039 a95241bf-73f2-0310-859d-f6bbb57e9c96
Fixes a crash in WonderBrush when dragging images over the layer list
to import them and triggering auto-scrolling. WonderBrush caches the
drag message and appearantly calls it's own MouseMoved() hood with the
cached message from within the auto-scrolling code.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29464 a95241bf-73f2-0310-859d-f6bbb57e9c96
bool return type. Did I mention this was an unstable Haiku extension?
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28181 a95241bf-73f2-0310-859d-f6bbb57e9c96
the data of two BMessages and allows recursive comparison (BMessage inside
BMessage). Note that using this API might require you to recompile your app in
the path to R1.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28174 a95241bf-73f2-0310-859d-f6bbb57e9c96
was pretty broken. It would store the address of the buffer in the message
instead of the actual data. Funnily, since it was completely stack based before,
it would have stored the address plus the actual data (minus 4 bytes) as that
did reside directly after the address on the stack and the original buffer
length was still used. This corrupted the data, but still "worked" for MDR
because it has fault tolerance, loosing only part of its flattened StringList.
This broke however when switching to a heap allocated buffer for large sizes,
as then the heap address plus some random heap data was added to the message
instead of the actual buffer. Sure interesting that nobody noticed that before...
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26984 a95241bf-73f2-0310-859d-f6bbb57e9c96
passed to the function against NULL.
* Made BMessage::AddFlat() use an optionally heap allocated buffer versus always
a stack allocated one.
* Small simplification in BMessage::AddMessage(), we can simply compare the
buffer with the stack based buffer to know whether we need to free() it.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26943 a95241bf-73f2-0310-859d-f6bbb57e9c96
which does
message->GetInfo("field", &type, &count);
while (message->FindBlah("field", --count, &...) == B_OK)
...;
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26695 a95241bf-73f2-0310-859d-f6bbb57e9c96
defined/undefined to numeric values (0 for undefined). This allows for
trace levels.
* Set SYSCALL_TRACING_IGNORE_KTRACE_OUTPUT default to 1, since this is
what one usually wants.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25213 a95241bf-73f2-0310-859d-f6bbb57e9c96
finally created a solution to avoid that: Header files that contain
configuration settings (and nothing else) go to build/config_headers.
To change settings, create a directory build/user_config_headers (which
is ignored by svn), copy the respective header there and modify it at
your leisure. Currently only tracing_config.h has been moved to the new
location, but more files will follow eventually. It is also recommended
to move optional macro definitions in Jamfile (as for BFS) to a config
header instead; the build system will then automatically rebuild on
changes.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24611 a95241bf-73f2-0310-859d-f6bbb57e9c96
the new sematic of transfer_area so a message area is transfered into the right
teams' address space and it does not need to be cloned there anymore. Passing
by area is only used for messages bigger than a certain size (currently
hardcoded to 40KB) which should be somehow bound to the max port message size.
This makes passing large messages (i.e. > the port limit) possible, so for
example copy&paste of long text, image data, etc. should now work.
Got rid of the fClonedArea member as it is not necessary with the new design,
renamed shared_area to message_area in the private message_header, avoid
an unnecessary allocation of the header for the copy constructors, check
allocations in a few more places and some minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24321 a95241bf-73f2-0310-859d-f6bbb57e9c96
Tracked down the root cause of the issue worked around in r24228.
The behavior that was occurring in this case was as follows:
Vision created a BInvoker off the heap, containing a BMessage likewise
off the heap. It then called BAlert::Go with this invoker. When any of
the buttons in the BAlert were pushed, _SendMessage would enqueue the
message into the target looper (in this case, Vision's window) and wake
it up with write_port_etc. However, in some cases this would result in
a reschedule such that Vision's looper executed before _SendMessage() was
able to complete all its processing. Upon receiving this message, Vision
destroyed the BInvoker in question, which in turn deleted its owned
BMessage -- which was the message that was still in the middle of _SendMessage()!.
Consequently it would crash a few lines later when it hit IsSourceWaiting() and
tried to check flags off its header off the already destroyed this. To fix this,
we now do the AddMessage/write_port_etc last before returning. Also removed the
sanity checks added in r24228 so we don't mask other problems of similar nature.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24297 a95241bf-73f2-0310-859d-f6bbb57e9c96
set up with a null replyTo, sometimes BMessage would crash
calling IsSourceWaiting() after delivering the message. This appeared
to be because fHeader was NULL, but I'm not entirely certain if this
is actually supposed to be possible, so this may be masking a different
bug. This was observable using the multiline paste spam BAlert in
Vision, which would sometimes but not always crash in Haiku after hitting
any of the buttons to dismiss it, though consistently when calling IsSourceWaiting()
from the private BMessage::_SendMessage().
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24228 a95241bf-73f2-0310-859d-f6bbb57e9c96
- no need to initialize the buffer on stack
- no need to initialize "buffer" to NULL
- renamed "buf" to stackBuffer
- enlarged buffer on stack to 16384 bytes (we have a minimum of 192 kB of
stack per thread, anyway).
- check the actual size of the stack buffer against the message's flattened
size instead of the one of its pointer.
- check if the allocation of the helper buffer failed, and return B_NO_MEMORY
in this case.
* Moved static helper functions to the top of the file.
* Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24204 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Allocate the buffer to flatten the message on the heap, if it's size is bigger then a given
buffer on the stack. It seem's to exceed the stack size (this might count for AddFlat() too).
Note: With this change one is able to copy the text into the clipboard (1mb), but it
is still impossible to paste it somewhere, as in BClipboard::_DownloadFromSystem()
SendMessage() fails transferring the data back in the reply msg because of the port size limit...
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24196 a95241bf-73f2-0310-859d-f6bbb57e9c96
destination of the message and it's "what" field are stored. It might be
nice to also get some info about its fields -- maybe as an additional
option.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23810 a95241bf-73f2-0310-859d-f6bbb57e9c96
field was not initialized properly, and the reply target was taken from the
wrong header in this case.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22925 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Always check the return of the resize function
* Handle resize errors gracefully and ensure that the message stays intact
* Don't crash when printing a message that contains a field with no items (unlikely but possible in low memory situations)
* Fixed renaming fields - was completely broken and would have missed up field names and contents
* Some cleanup
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22047 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Removed r5_message.cpp and dano_message.cpp accordingly
* Also moved out KMessage handling from Message.cpp to MessageAdapter.cpp
* Fixed some minor style issues in Message.cpp
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21514 a95241bf-73f2-0310-859d-f6bbb57e9c96
general _Clear(), so that the assignment operator and Flatten() will
trigger such a reply as well.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@20083 a95241bf-73f2-0310-859d-f6bbb57e9c96
header flags after copying the header of the original message. Before
cloning a message that still needed a reply would result in the clone
also needing a reply, which, in the end, led to two replies being sent,
thus screwing up the cached reply ports used for synchronous messaging.
Fixes bug #1008.
BTW, also clearing the reply info breaks Tracker and Deskbar, so I suppose
R5 keeps it and Tracker and Deskbar rely on that behavior.
* Added a bit of debug code removing and printing spurious messages from a
cached reply port.
* Added TODO in BMessage::~BMessage(). ATM, we only send a B_NO_REPLY in the
destructor, but not in case the message is overwritten using the
assignment operator or Flatten().
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@20074 a95241bf-73f2-0310-859d-f6bbb57e9c96
with PostMessage() in case the message queue is full.
Some notes:
* for synchronous replies, we don't use this mechanism yet, but it could be
extended to do that as well.
* the code looks so complicated because we need a way to access the looper's
queue without locking it (to prevent deadlocks); like Dano's solution, I've
abused BTokenSpace to store a BDirectMessageTarget with a BHandler.
* we also need to decouple the lifetime of a looper's queue from its target,
as we cannot lock the looper, and therefore, can't guarantee it stays valid
as long as we're accessing it outside of BLooper.
* init_clipboard() now needs to be done after the global constructors have
been called - since sending messages now needs gDefaultTokens to be initialized.
Since this is done per image, it shouldn't cause any troubles, though.
* some minor cleanup, removed unused _msg_cache_cleanup_() and friends.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19968 a95241bf-73f2-0310-859d-f6bbb57e9c96
again with a synchronous reply - the "reply done" flag wasn't cleared before sending.
* Improved error message in case the buffer for flattening the message couldn't be
allocated. BTW this would be a good place to use the new writev_port() function and
don't do any allocation at all.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17914 a95241bf-73f2-0310-859d-f6bbb57e9c96
This would explain why BLooper::resolve_specifier() was misbehaving before I fixed it
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17834 a95241bf-73f2-0310-859d-f6bbb57e9c96
tested with ForwardMessageTest.cpp
well I hope it doesn't break anything :)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17831 a95241bf-73f2-0310-859d-f6bbb57e9c96
exceptions in case of an error. Therefore, whenever those methods are called,
we need to surround them with a try/catch block.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17766 a95241bf-73f2-0310-859d-f6bbb57e9c96
Stephan Aßmus