Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
commit 3fbadcc Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Tue Dec 31 10:36:02 2019 +1000 Added some more test spells. Now able to test a range of damage and heals. commit c55c53e Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Tue Dec 31 09:30:53 2019 +1000 Removed the source and target entity ids from all the all components and made a single component that holds them. Damage and heals and qi use now need to make sure to also get the SourceAndTarget component to know who to apply their effects onto. This will work as long as damage always applies to targets and qi use applies to the source. It breaks down in two cases: mana burn and final hit spells with damage penalties for not killing the opponent. These can be handled as special cases if they are needed, or two spells could be queued for them. commit 0d7a62e Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Mon Dec 30 01:13:54 2019 +1000 Spells can be cast by name now. commit 45f3f07 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Mon Dec 30 00:53:41 2019 +1000 Simple code to test and use prototype spells is now in place. Added concept of "re-wire" which based on knowledge of the spell type will change spell members after it has copied it to the actor register. Right now, it is just re-writing the source and target entities. In time it will be able to apply spell bonuses, etc. commit 03a340f Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Sun Dec 29 23:08:16 2019 +1000 Removed the qi utilisation type as it makes no sense. commit dbfad7f Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Sun Dec 29 22:32:56 2019 +1000 Started work on adding spells. Added a spell registry for prototypes. Defined a spell component and put a simple attribute in there to get started on testing. Loading and saving to XML. commit d0df047 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Sat Dec 28 19:52:17 2019 +1000 Added one very basic hard coded spell. Any actor can cast it at any other actor. It costs qi and causes damage to the target. Very simple debug output of the health and qi attributes for actors. commit e00ca09 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Sat Dec 28 14:09:24 2019 +1000 We're now destroying the entities once they are no longer needed. It's crude, because we're assuming the entity only contains that one component, but it will do for now. Corrected a small mistake with ticks. It's still just an approximate, but at least it will tick at constant intervals, and catch up if needed. commit 1ea6196 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Sat Dec 28 00:12:55 2019 +1000 The ECS simulation has been moved to it's own class. It's loading test data, and saving results (though the code position for that is terrible and temporary). The simulation is running, but giving excessive results because the test components and entities are never being deleted. commit 09aed03 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Fri Dec 27 17:04:00 2019 +1000 Added test cases. Left DamageType on the Qi component to remind me to change it to something relevant or remove it. commit ca6a548 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Fri Dec 27 13:13:04 2019 +1000 Working on adding Qi as a component with a matching system. Mostly done now, just need some test cases. commit 9d71627 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Thu Dec 26 05:04:41 2019 +1000 First attempt at healing. It's just the opposite of damage for now, except there is no "type" for heals. I suspect I need to add one later - not sure if needed for now. commit ba0b429 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Thu Dec 26 01:14:14 2019 +1000 Adjusted the way DoTs are handled. It's good enough for test purposes but might be brittle for real world handling where ticks are less reliable. commit 7ede6c1 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Thu Dec 26 00:11:55 2019 +1000 Added an ApplyDamage system to take the test damage data and apply it to an entity. It worked pretty well. Only concern really is the lookup times for entities. I'm doing a brute force to get a target entity for now, but there is also a hopefully much faster lookup to get the health component for that entity. I will have to look at using an index to replace the brute force at some point. commit 6a80c6d Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Wed Dec 25 23:21:25 2019 +1000 Starting work on adding a Damage component to be executed at runtime. commit 1ad8f92 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Mon Dec 23 22:17:16 2019 +1000 Adding health and health modifier to the ECS. It's not quite right yet. Really needs a damage type for every health modifier. Added a first pass at the base requirements for item classes. Not sure how to make it work, looking up each one from item instances seems like it might be very slow, especially on loading. commit 53df99e Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Sat Dec 21 22:35:19 2019 +1000 Adding an attribute template type to ease using attributes. Testing it with the Health attribute. commit 606af71 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Fri Dec 20 22:58:12 2019 +1000 Removing some of the kruft from the XML file. No need for the GUIDS right now, so they are suppressed. The properties nodes are overkill. Just putting all the data into the node with the tag for that component. This will make hand editing these a lot easier. commit 8a0e07e Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Fri Dec 20 22:03:34 2019 +1000 Fixed the broken code from the last check in. Components are now correctly being created and returned by reference. Cleaned up the code a little. Moved some code to more sensible places. commit d7e7fdb Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Fri Dec 20 12:52:02 2019 +1000 Removed the template code for assigning to an entity and registry. This is no longer needed since the simpler method of having it done in the constructor is available. We're now ready to start putting in a bunch of components and seeing how well this thing can work. commit 897d48b Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Thu Dec 19 23:24:09 2019 +1000 EnnT test code is now working as expected. Data is being round tripped, although the order of the entities is being reversed - which will make source control of output files a little tricky. Started cleaning up the code, getting ready for testing by adding a slew of components. commit 3ec7aee Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Thu Dec 19 22:02:32 2019 +1000 Removing the submodule for ennt commit aa9e3c9 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Thu Dec 19 21:19:45 2019 +1000 Pushing changes prior to fiddling with the upstream master of the EnnT project. commit 5586c96 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Thu Dec 19 03:02:06 2019 +1000 Test save is operational. Still have an issue with not being able to set up the invoke as a template. Using methods on the component classes as a half measure till I can work that issue out. commit be04af1 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Thu Dec 19 00:28:54 2019 +1000 Testing snapshots, outputting XML, but it's not correct. Assumption about order of calls to the serial object being made was wrong. Need a way to tag each entity, and look it up each time in the XML when writing it's components. This is pretty inefficient, but it's got to be done due to the way snapshots work. commit 7a6039c Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Wed Dec 18 22:29:14 2019 +1000 Experimenting with the snapshot ability as a way to save instead. This time around I have managed to get the operators for the output archive correct, and it runs and calls them. Need to see about writing it to XML now. commit d6d351c Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Tue Dec 17 22:39:34 2019 +1000 Enough test code written to create a new component and add it to an entity using only hashstrings which can be read from an XML file. Everything should be in place now to make the XML loading code work. commit a9630b7 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Tue Dec 3 21:51:34 2019 +1000 Prototyped a lot of the XML reading code. Missing the most important part - instantiating the components. commit 45ee8da Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Tue Dec 3 09:33:25 2019 +1000 Minor cleanups to code. Getting prepared to re-load the XML now. commit 48692b6 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Tue Dec 3 08:54:46 2019 +1000 Test code is creating a decently formatted XML file and using the serialize code to populate the component properties. commit baf003b Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Mon Dec 2 13:34:05 2019 +1000 Successfully pulling data from the ECS and writing the basics out to an XML file. It doesn't have the ability to use the component serialise function yet. Still looking through the code for the best way to handle that. commit 0d2f4e6 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Fri Nov 29 22:10:27 2019 +1000 The basic layout for a component seems to work so far. Next step, saving and loading it. commit 4865a4d Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Sun Nov 17 05:06:58 2019 +1000 Checking in before switching branch. commit 021e7e0 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Mon Nov 4 23:37:17 2019 +1000 Enough code written to support output serialisation of the entities to a null device. Need to actually get it to go somewhere now it compiles. commit cfedb62 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Sat Nov 2 21:57:34 2019 +1000 Attempting to get some serialisation is not going well. Not sure how to make it play nice with anything, let alone YASLI. commit d22a8a4 Author: Ivan Hawkes <ivan.hawkes@gmail.com> Date: Fri Nov 1 22:44:37 2019 +1000 Added in repository for entt, theECS header only code. This needed extra include files which required a change to the makefile. That change will be lost next time it regenerates a soultion so it might be time to figure out how to prevent that happening. Some really dirty test code is in place to make sure the ECS works, and so far it has. Will need to allow it to serialise, which will be the next big challenge.
- Loading branch information
1 parent
0c23a40
commit 9d60e35