GUIDs are most safer than pointers and in some cases can avoid segmentation faults.
== - ==
Let me show a simple example:
We have a pointer to a creature: Creature *pCreature;
and try to search a creature and store it in the pointer:
pCreature = SearchCreature(ID);
We can check simply if the creature found or not:
if(pCreature)
- if this give true: creature found and we have a memory address which points to the creature
- if this give false: creature not found and this is a NULL pointer
Suppose that we have a valid pointer, so creature found.
We wanna use that pointer somewhere, for example in script phase 3.
But we delete the creature(from world) in phase 1.
In phase 3:
if(pCreature)
pCreature->DoSomeShit();
The if condition gives true, but pCreature points to an _invalid_ address.
In that address maybe have "nothing" or another creature's object or a player or anything else, dont know that.
~~ - ~~
Let me show the correct way:
Creature's GUID: uint64 CreatureGUID;
search the creature and set the GUID:
Creature *pCreature = SearchCreature(ID);
CreatureGUID = pCreature ? pCreature->GetGUID() : 0; // if pCreature is NULL set the GUID to 0
So we have a GUID and not a pointer.
We can get the creature from that GUID:
Creature *pCreature = GetCreature(CreatureGUID);
and we can simply check it is valid or not(found or not):
if(pCreature)
So we deleted the creature in phase 1 and try to use it in phase 3:
Creature *pCreature = GetCreature(CreatureGUID);
And this gives NULL because GetCreature() can't find the creature which GUID is CreatureGUID.
if(pCreature) // pCreature is NULL
pCreature->DoSomeShit(); // DoSomeShit() not called
== - ==
Remove some not used variables.
Some clean.
TODO: Should search/fix these in zones scripts.
--HG--
branch : trunk
- Support for spawning Onyx Lair guardians
- Maximum Whelp count per wave set to 20 as per Blizzlike
- small update to her movement in phase 2 (might need to be overhauled, Breath isn't casted often enough)
--HG--
branch : trunk
* Credits (in no particular order) to:
* n0n4m3, raczman, manuel, Spp, Malcrom, Teacher, QAston, Tartalo,
* thenecromancer, Xanadu, Trazom, Zor, kiper
* Additional credits to:
* TOM_RUS and NoFantasy from MaNGOS
* Thanks for testing Aokromes and XTElite1
* SoTA still needs some work, but is very playable (huge thanks to raczman and
* kiper)
* To upgrade, you need to apply all SQL from sql/updates/3.2.2a_old from the
* last rev you are on
* and then apply all SQL from sql/updates/3.3.2_old to char / realmd / world
* DBs
* Known problem with guild banks.
--HG--
branch : trunk
** It was wasting CPU power as cell-level locking is not needed.
** Our multithreading is on map-level, not cell-level.
** CellLock was just a 'proxy' between Cell and CellPair and in some cases carried redundant data.
** Some minor cleanup in Cell::Visit/Map::Visit.
--HG--
branch : trunk
-added ScriptMgr for loading scripts
-removed bindings
-moved script system to src/game
-moved scripts to src/scripts
-VC project files updated
-cmakes updated (not 100% done yet)
NOTE to Devs:
-file locations changed
-precompiled renamed to ScriptedPch
-ecsort_ai renamed to ScriptedEscortAI
-follower_ai renamed to ScriptedFollowerAI
-guard_ai renamed to ScriptedGuardAI
-simple_ai renamed to ScriptedSimpleAI
-sc_creature renamed to ScriptedCreature
-sc_gossip renamed to ScriptedGossip
-sc_instance renamed to ScriptedInstance
*use the new headers in scripts, thank you
NOTE to ALL:
cmake not fully tested, please report any errors with it
could make creashes, incompability
USE AT YOUR OWN RISK before further tests!!
--HG--
branch : trunk
Spp