We can now use the sets to iterate the bindings in order of their
expiry and exit the loop when we encounter the first one with an expiry
in the future.
Change-Id: I093bea9807b4a432ac965b402394edc43eef5a1f
Add an ordered set of all AorEntry objects, ordered by expire time.
Amend modifier methods to update the set whenever some expire time is
updated or an entry or binding removed.
Change-Id: I2afe5becfa2124dbc8c33b38fe1a3e6d4b8a20d5
Move deletion of bindings into AorHash, similar to how expire times are
set. This makes it possible to manage expiry in a central place.
Change-Id: I76a72881ed202e85fb70e9d8526bac55863d07db
Make AorEntry keep an ordered set of all bindings, ordered by expire
time. This makes it easy to quickly tell when the next binding is set to
expire, and which one. Store iterators to bindings in the set so that
they can be used to directly remove the binding when expired.
Change-Id: Ib63c15d5f6e8990e2f4d237726be54da76194a20
Provide accessor methods. Put the setter into AorHash and require
iterators. All bindings must be in an AorEntry container and all
AorEntry objects must be in the AorHash container. By putting the setter
in AorHash and using iterators we can make sure that the binding is in
the container already and that expiry values are kept consistent.
Change-Id: I1c5140d1a7547059f61e47cd317680fcb6eae9fe
Provide a proper class for key matching/ordering instead of
hand-crafting a key to be used as index in several places. Use a simple
ordered map instead of the bucket-based approach.
Change-Id: I9e1918a48e103b655caac160c125a0393d261fca
Put all AorEntry objects into a single hash map. Eliminates the need for
AorBucket. Use an unorered hash map for performance. Use lock guards
instead of manual lock/unlock. Remove an unused argument to gbc().
Switch to the full tick cycle length instead of the tick-based one as
there are no more buckets to iterate.
Change-Id: I7da3bacaa5a28b7e628695ab879394fc6d56f5d2
Simplify the little-used RegisterCache alias hash by using the new
hash_map template. Eliminates the need for AliasBucket. Use an unordered
hash map for performance. Use lock guards instead of manual lock/unlock.
Manually `delete` contained objects as our table's .erase() doesn't do
that.
Change-Id: Ia9721460ce8264fc0a90184a91beca46012b64dd
Instead of relying on a short tick duration to make sure the thread
exits when requested, add a dedicated condition flag and variable. Use
AmThread on_stop to set it and wake up the thread. Use the condition
variable's timed wait instead of nanosleep for the ticks.
This somewhat duplicates what other threads are doing, and what
stop_requested() is doing, and should be unified into a single mechanism
provided by AmThread in the future.
Change-Id: I364e747b4acc1986fa262588c47d83c063ff2d69
Temporarily releasing the lock only to immediately re-acquire it, just
because the called function does a lock/unlock, without re-checking all
data that is protected by the lock, opens the door to subtle race
conditions. Change the semantics of removeRegistration() to require
being called with the lock held. Update handleRegistrationTimer() to
also not release and re-acquire the lock.
Change-Id: Ib7562ae138b55d0ef83ec142706ec6415037fc2a
Rework the `nbWrite()` function so, that it gets safe
in terms of:
- operating on data written/data remaining integers
- has protection guards for extra big chunks to write at once
Refactor to use size_t/ssize_t instead of primitive int,
where possible (e.g. bytesWritten).
Let the `nbWrite()` use reference instead of pointer
for the bytes written parameter.
Fixes Coverity Scan report:
4. tainted_data_return: Called function write(fd, sp, nToWrite),
and a possible return value may be less than zero.
5. cast_overflow: An assign that casts to a different type,
which might trigger an overflow.
Overflowed integer argument (INTEGER_OVERFLOW)
13. overflow_sink: nToWrite, which might have overflowed,
is passed to write(fd, sp, nToWrite).
7. overflow: The expression nToWrite is considered
to have possibly overflowed.
Change-Id: I32aa6aae5ef5715d61a91714e62b0a094bc03f21
As long as the `TypeInt` is actually a long primitive,
and we tend to switch using the long type everywhere
within xml implementation, then
there is no need to support both `int` and `long`
types, just force users to be on `long` always.
Refactor the user code accordingly.
Change-Id: I5c5b032a824a84f69f99dc7c755d0da745a9a068
As this is only a syntactic sugar there is no difference,
but from the style perspective the project tends to
use C++ style.
Change-Id: I3636c48163b3d262b272b5088e03c074e49a9ec3
Properly select the equal operator based on long instead.
Otherwise it selects the `ValueStruct` as the latest
possible from the variant based `_value` and eventually
it gets stumbled over `[]` operator for
building structs which asserts it.
Change-Id: Ib9a4fe9c66ff02554a84451df8aa940fe766fedc
Don't try to double free the `_ssl_ssl` and `_ssl_ctx`
provided by `XmlRpcSource`, because this job
is already done by `XmlRpcSource::close()`.
Additionally:
- add a separate non-null check for the _ssl_ctx freeing
- move log lines from `XmlRpcClient::close()` to `XmlRpcSource::close()`
Change-Id: Ifef799fa6a0ffa65b60415f39c51f4b54ff4bb14
Additional logic around these two objects to free it during the
destruction phase.
Also add NULLing for the according pointers in other places
which might also free it before the destructor takes place,
which would lead otherwise to a double free attempt.
Change-Id: I046a660e4b1467a7093dbf2c0c3170a950d5f607
Don't appeal to the XmlRpcSource's members (e.g. `_ssl`)
if this has been freed via `this`.
To check this out see controllable delete behavior
via introduced getter `getsDeletedOnClose()`.
Should fix:
freed_arg: close frees this.
CID 542402: (#1 of 1): Read from pointer after free (USE_AFTER_FREE)
use_after_free: Using freed pointer this.
Change-Id: I8f23564957c57491db8cf051e7296671bf2a698a
For newly created call profiles use the move operation
instead of copy one.
Fixes:
CID 542292:Variable copied when it could be moved (COPY_INSTEAD_OF_MOVE)
Change-Id: Ia130fc3e32d23e63f3d41fc6261e90ecf742ecf7
Let STL handle memory management and type safety.
In the future this can be refactored further to eliminate the "Type"
enum and value, as the variant itself keeps this information.
Fix white spacing.
Change-Id: I6e903b51de339b686ba89bb7ae81f4d86eb66bc0
Store integers as long to reduce likelihood of overflows. Retain
overloaded accessors for compatibility.
Use time_t and suseconds_t for timestamps.
Fix up formatting strings to match the new types and to also fix
incorrect formatting of decimal values.
Remove unused methods.
Change-Id: I58bfcd8a34bef2363eac06242dfedf6aae8451e0
Instead of copying the FilterEntry mf, do the move.
Fixes defect:
329 vector<string> elems = explode(value, ",");
330 for (vector<string>::iterator it=elems.begin(); it != elems.end(); it++)
331 {
332 mf.filter_list.insert(*it);
333 }
334 mf.filter_type = Undefined;
>>> CID 550235: Performance inefficiencies (COPY_INSTEAD_OF_MOVE)
>>> "mf" is copied and then passed-by-reference as parameter to STL insertion function
"std::vector<FilterEntry, std::allocator<FilterEntry> >::push_back(
std::vector<FilterEntry, std::allocator<FilterEntry> >::value_type const &)",
when it could be moved instead.
335 profile->messagefilter.push_back(mf);
336 DBG("message_list set to '%s'\n", value.c_str());
337 EXEC_ACTION_STOP;
338 }
339
340 }
Change-Id: If6d04432f7004993fb1c70a0a026799805ed2a1a
Directly contain the ValueArray in the variant instead of manually
managing a pointer. Requires updating the return type of some const
functions to also be const, which in turn requires some of their users
to have the type const qualified as well.
Change-Id: I2ec31d659eef521a3f68d642ee431b5c38f27fdf
Update the AmArg argument to const as these functions never change it,
and it may come from a const source. (The `string` counterpart in the
other overload is already const qualified.)
Change-Id: I1e527425a70e0c6468b8b59de7e4a20fa1f5c218
When dereferencing it (after creating a new context before)
make sure it's not NULL even when just checking for
a connection error, as it might still be NULL and hence
will just seg.fault in this case.
Fixes:
Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking this->redis_context suggests
that it may be null, but it has already been dereferenced on all paths leading to the check.
Change-Id: I67f88f1dfccef66751298b1c26e85efba0458c1f
size_t will always be true against `>=` because
it's a typedef for unsigned long.
Makes no sense to check it, just log what it is.
Fixes: Unsigned compared against 0 (NO_EFFECT)
unsigned_compare: This greater-than-or-equal-to-zero comparison
of an unsigned value is always true. reply->len >= 0UL.
Change-Id: I75803cbe0760a1b5c4a0d056af86bc992ea0a110
Add a default case for `handle_redis_reply()`.
Fixes:
Dereference after null check (FORWARD_NULL)
4. var_deref_op: Dereferencing null pointer reply.
Change-Id: I22edbc3cb14f7e8662efe4c607e1502fbb8c56b6
Instead of using the `tmpnam()` which doesn't exclude
race conditions on the system (same filenames created
simultaneously), because it doesn't open/reserve the file,
just use the `mkstemp()` which actually genereates a filename
and creates the file for the process.
Upon creating the file, close the FD, as not needed atm.
This file can be then later opened using `SCPopenAction`
or anything alike.
Current new implementation does not unlink the file from
the filesystem and a user of it has to explicitly do that
later with for example the `sys.unlink()` from this same
library.
Additionally: to be able to have excplicit file path (so dir),
add a second parameter to this function to provide a directory.
Fixes:
warning: the use of `tmpnam' is dangerous, better use `mkstemp'
Change-Id: Ib5e89461b2d2288be77d5f1a899e5293b76ebf90
Fixes:
CID 542433: (#1 of 1): Uninitialized scalar variable (UNINIT)
34. uninit_use_in_call: Using uninitialized value mf.
Field mf.filter_type is uninitialized when calling push_back.
Change-Id: Ia5c2c2ebb08f1632f973ef0d08d4dca0c5b8934b
Don't do an extra job many times checking
what is the `profile_param`, just catch this once.
Hence conditions must be self-excluding.
Change-Id: I6d98a082e50777c995538fd6303eb0fe527b301b
If a channel has been allocated with
`ConferenceJoinChannel()` then remember to transfer
the ownership over it to the DSM sessions's
"garbage collector" which later takes care to
deallocate it.
Fixes:
CID 542153: (#1 of 1): Resource leak (RESOURCE_LEAK)
10. leaked_storage: Variable chan going out of scope leaks the storage it points to.
Change-Id: Icdc2c67cf2a349ec70fd83f1cf6a43ecfcf69547
Instead of copying string parameter, just pass via reference.
Fixes:
Variable copied when it could be moved (COPY_INSTEAD_OF_MOVE)
Change-Id: I872cd9631bbadc746aea12e7ae5840670fa16e8f
Richard Fuchs
Donat Zenichev