| *** tpb has joined #tp | 00:00 | |
| *** ChanServ sets mode: +o tpb | 00:00 | |
| *** StrangerDanger has quit IRC | 00:13 | |
| *** StrangerDanger has joined #tp | 00:27 | |
| *** Agon-laptop has quit IRC | 00:40 | |
| *** verhoevenv has quit IRC | 02:21 | |
| *** ezod has quit IRC | 02:26 | |
| *** ezod has joined #tp | 02:33 | |
| *** nash has joined #tp | 02:46 | |
| *** StupidIncarnate has joined #tp | 03:16 | |
| *** glew has quit IRC | 03:44 | |
| *** mithro has joined #tp | 03:59 | |
| llnz | bbl | 04:32 |
|---|---|---|
| *** llnz has quit IRC | 04:32 | |
| *** bisc has joined #tp | 05:06 | |
| matthewd | StupidIncarnate: hey | 05:35 |
| bisc | tansell: ping | 05:40 |
| tansell | bisc, pong | 05:40 |
| bisc | tansell: I'd like to have your review for http://codereview.mithis.com/85001 | 05:41 |
| tpb | Title: Issue 85001: Halfway patch for right click menu. - Code Review (at codereview.mithis.com) | 05:41 |
| tansell | bisc, you can revert the change to windows/main/panelOrder.py | 05:43 |
| bisc | oops, haven't noticed it. Just a second. | 05:44 |
| StupidIncarnate | tey there matthewd | 05:47 |
| bisc | tansell: updated this issue. | 05:48 |
| tansell | looking at it now | 05:48 |
| tansell | bisc, looking pretty good - some comments added | 05:53 |
| matthewd | StupidIncarnate: So, I agree that you still seem to be a bit ahead of schedule, which is great | 05:58 |
| tansell | matthewd / StupidIncarnate: do you have a demo up somewhere people can test with? | 05:59 |
| StupidIncarnate | nope. if you set me up some server space i can | 06:01 |
| matthewd | tansell: Good idea... I'll set it up on my server | 06:01 |
| matthewd | It'll give me a chance to try out the new from-scratch install instructions, too | 06:01 |
| bisc | tansell: what's the generic way to find object's coordinates? Through objectutils.getPositionList? | 06:04 |
| tansell | bisc, yeah | 06:05 |
| tansell | I'm not sure of the correct way to handle the fact that objects might have multiple positions | 06:05 |
| tansell | (IE a wormhole has two positions) | 06:05 |
| *** cahirwpz has joined #tp | 06:06 | |
| matthewd | Looks like I'm about 0.2ms away from verbal, so I don't foresee latency issues :) | 06:06 |
| bisc | tansell: ok, I think we can postpone decisions about multiple positions now. | 06:06 |
| tansell | cahirwpz, ping? | 06:07 |
| StupidIncarnate | so anything have to happen for midterm matthewd? | 06:08 |
| StupidIncarnate | want anything from me?* | 06:08 |
| bisc | tansell: can you have a look at updated game info window http://codereview.mithis.com/79001 ? | 06:08 |
| tpb | Title: Issue 79001: Game info window. - Code Review (at codereview.mithis.com) | 06:08 |
| matthewd | StupidIncarnate: Don't think so | 06:09 |
| StupidIncarnate | k then just the survey thing | 06:09 |
| matthewd | StupidIncarnate: I guess now's probably a good time to ask, how are you feeling about your overall progress etc? | 06:10 |
| matthewd | I guess the fact you're ahead of schedule suggests you wouldn't feel terrible :) | 06:11 |
| tansell | bisc, LGTM | 06:11 |
| StupidIncarnate | it's been good, being ahead is better than when I was behind at the start | 06:11 |
| matthewd | But yeah... are you enjoying what you're doing, feeling accomplishment, that sort of thing | 06:12 |
| StupidIncarnate | probably when I get the design panel done | 06:12 |
| StupidIncarnate | i mean the redesign felt accomplishmenty I suppose, though it's not completely done | 06:13 |
| cahirwpz | tansell, pong - it's 8 am here, I'm trying to wake up ;) | 06:13 |
| tansell | StupidIncarnate, we should sit down sometime and work through the problems you are having libtpclient-py | 06:13 |
| StupidIncarnate | am I having problems with it? | 06:14 |
| StupidIncarnate | don't remember | 06:14 |
| tansell | StupidIncarnate, with the cache stuff | 06:14 |
| StupidIncarnate | oh ya.... I put that out of mind | 06:14 |
| bisc | tansell: I've got problems with wrapping in Information panel. The problem is that wx.StaticText doesn't want to resize vertically for wrapping in automatic mode. It either takes all size that it can get, or doesn't resize at all. Have you ever confronted such a problem? | 06:19 |
| tansell | bisc, possibly but it was a long time ago | 06:20 |
| bisc | tansell: small patch for moving a method http://codereview.mithis.com/87001. Just though it's better to place it in objectutils. | 06:24 |
| bisc | tansell: what do you think about placing a checkbox 'automatically close' in Update window? I already hate clicking it myself, and any player won't like it as well. | 06:32 |
| cahirwpz | tansell, ping? | 06:45 |
| *** llnz has joined #tp | 06:54 | |
| llnz | hi all | 06:54 |
| cahirwpz | hi llnz :) | 06:56 |
| llnz | hi cahirwpz | 07:02 |
| cahirwpz | llnz, I found very interesting article about protocol serialization / deserialization frameworks - you might want to read it | 07:04 |
| cahirwpz | http://code.google.com/p/thrift-protobuf-compare/wiki/Benchmarking | 07:04 |
| tpb | <http://ln-s.net/7Fk-> (at code.google.com) | 07:05 |
| llnz | cahirwpz: will have a look | 07:06 |
| cahirwpz | llnz, I don't know how exactly protocol handling looks like in C++ tpserver version, but Google Protocol Buffer might be an inspiration if you look to perform some refactoring there | 07:07 |
| cahirwpz | look at the code here: http://code.google.com/p/protobuf/ | 07:07 |
| tpb | Title: protobuf - Project Hosting on Google Code (at code.google.com) | 07:07 |
| llnz | cahirwpz: currently Eypon is working on porting to boost ASIO | 07:10 |
| llnz | as part of GSoC | 07:10 |
| *** StupidIncarnate has quit IRC | 07:22 | |
| *** bisc has quit IRC | 07:31 | |
| *** Greywhind has quit IRC | 07:50 | |
| *** Greywhind has joined #tp | 07:51 | |
| *** bisc has joined #tp | 08:57 | |
| *** bisc has quit IRC | 09:00 | |
| *** bisc has joined #tp | 09:07 | |
| * llnz finds the root cause of a MTSec bug | 09:26 | |
| *** null_000 has joined #tp | 09:26 | |
| tansell | cahirwpz, pong | 09:37 |
| cahirwpz | tansell, hi | 09:37 |
| tansell | so we should continue from our last conversation | 09:37 |
| cahirwpz | tansell, I need some help - I don't exactly understand what was the purpose of 'public', 'protected', 'private' access level in attributes | 09:38 |
| tansell | public == send/receive the raw value straight from the packet | 09:39 |
| tansell | protected == send/recieve a value from a packet after processing | 09:39 |
| tansell | private - only used server side | 09:39 |
| cahirwpz | I almost finished attributes / parameters stuff and I'll move to logging / configuration soon | 09:39 |
| tansell | what is the attributes/parameter stuff? | 09:41 |
| cahirwpz | http://github.com/cahirwpz/tpserver-py/blob/master/tp/server/rules/base/objects/Fleet.py | 09:42 |
| tpb | <http://ln-s.net/7Fn7> (at github.com) | 09:42 |
| cahirwpz | lines 13-28 | 09:42 |
| cahirwpz | it's replacement for SQLTypedBase / Attribute | 09:42 |
| cahirwpz | tansell, why did you introduce public / protected / private access level - what did you want to achieve by that ? | 09:45 |
| tansell | we needed a system which describes what properties an object/order had | 09:48 |
| tansell | which is needed for both the packet serialisation and the database stuff | 09:48 |
| tansell | the | 09:50 |
| tansell | def desc_packet(cls, sequence, typeno): | 09:50 |
| tansell | method uses this information to build a description of an order | 09:50 |
| tansell | the same was planned for the object | 09:50 |
| tansell | when tp04 support was added | 09:50 |
| tansell | understand? | 09:52 |
| cahirwpz | lets say I need to read sources once again | 09:53 |
| cahirwpz | I have some clues, but not the answer | 09:53 |
| tansell | cahirwpz, the functions you want to look at are | 09:54 |
| tansell | http://github.com/thousandparsec/tpserver-py/blob/master/tp/server/bases/SQL.py#L327 | 09:55 |
| tpb | <http://ln-s.net/7FnS> (at github.com) | 09:55 |
| CIA-29 | llnz tpserver-cpp * rda756453eb39 /modules/games/mtsec/buildfleet.cpp: | 09:55 |
| CIA-29 | Fixed BuildFleet in MTSec, now works again. | 09:55 |
| CIA-29 | Over aggressive updating during merge of refactoring of last year's GSoC. | 09:55 |
| tansell | http://github.com/thousandparsec/tpserver-py/blob/master/tp/server/bases/SQL.py#L369 | 09:55 |
| tpb | <http://ln-s.net/7FnT> (at github.com) | 09:55 |
| tansell | cahirwpz, those two functions use the Attribute information to serialise/unserialise the DB | 09:56 |
| cahirwpz | tansell, that has actually been already replaced in my fork | 09:58 |
| tansell | cahirwpz, the other functions which use the information are | 09:58 |
| tansell | http://github.com/thousandparsec/tpserver-py/blob/master/tp/server/bases/SQL.py#L437 | 09:59 |
| tpb | <http://ln-s.net/7FnX> (at github.com) | 09:59 |
| tansell | and | 09:59 |
| tansell | http://github.com/thousandparsec/tpserver-py/blob/master/tp/server/bases/SQL.py#L468 | 09:59 |
| tpb | <http://ln-s.net/7FnY> (at github.com) | 09:59 |
| tansell | which are the to/from packet methods | 09:59 |
| cahirwpz | ... I see | 09:59 |
| cahirwpz | I need to think about this for a while | 10:00 |
| cahirwpz | this might be related to the idea of separating protocol from model implementation | 10:01 |
| tansell | the idea is that they are doing two very similar things | 10:02 |
| tansell | converting from the internal objects to another protocol | 10:02 |
| tansell | (Either SQL or Thousand Parsec protocol) | 10:02 |
| *** cahirwpz_ has joined #tp | 10:04 | |
| *** cahirwpz has quit IRC | 10:05 | |
| *** cahirwpz_ is now known as cahirwpz | 10:05 | |
| cahirwpz | tansell, lets leave the topic for now | 10:05 |
| *** nash has quit IRC | 10:05 | |
| cahirwpz | tansell, what about finishing the discussion from Monday ? | 10:06 |
| tansell | so we where discussing the twisted change and the ORM change | 10:09 |
| tansell | I iterated why I'm concerned with the ORM change | 10:10 |
| tansell | cahirwpz, we also discussed some future things which the ORM would need to support | 10:12 |
| cahirwpz | mhm... I remember now | 10:13 |
| cahirwpz | regarding object versioning I came up with idea of Model View | 10:14 |
| * llnz wanders off | 10:15 | |
| llnz | later all | 10:15 |
| *** llnz has quit IRC | 10:15 | |
| cahirwpz | I still think about it and it's very vague at the moment - but it will be flexible enough to implement fog of war, and other weird ideas | 10:15 |
| cahirwpz | basically - what we want is to create many views on objects in universe, possibly assigning each player a different view | 10:16 |
| cahirwpz | if players are in kind of military union we might need to merge their views | 10:16 |
| cahirwpz | so in fact there should be some mechanism (in ruleset probably) which knows how to generate a model view | 10:18 |
| cahirwpz | each model view should be represented in database by kind of DB snapshot | 10:19 |
| cahirwpz | probably read only | 10:19 |
| cahirwpz | I generally have high-level idea how to do it - I some more time to think about details | 10:20 |
| cahirwpz | sqlalchemy.orm is in fact able to map any class, set of classes, python built-in collection, so we're only limited by the design - ie. we're able to achieve exactly the same results not using DB at all but without persistence | 10:25 |
| cahirwpz | I believe that if we prepare a good design of model (one that avoids O(n log n) or O(n^2) and slower algorithms) we can effectively map in onto database using ORM | 10:28 |
| cahirwpz | hmm... model view can introduce some really interesting ideas like illusionary star systems, space cloning, whatever you can imagine... | 10:33 |
| cahirwpz | tansell, I also noticed that sooner or later we will need some solid vector / matrix library in rulesets - what do you think about that? | 10:35 |
| tansell | cahirwpz, why? | 10:36 |
| cahirwpz | to catch some structures and functions into method and object (vector3d, length, move, etc.) | 10:40 |
| cahirwpz | look at: http://github.com/cahirwpz/tpserver-py/blob/master/tp/server/rules/minisec/orders/Move.py | 10:40 |
| tpb | <http://ln-s.net/7FoO> (at github.com) | 10:40 |
| cahirwpz | for example line 55 | 10:40 |
| cahirwpz | wouldn't be clearer to write "d = self.pos - obj.pos" ? | 10:41 |
| cahirwpz | line 111: | 10:42 |
| cahirwpz | "distance = d.length()" | 10:42 |
| cahirwpz | or even for lines 110-111: "distance= ( self.pos - obj.pos ).length()" | 10:43 |
| tansell | cahirwpz, sure - but a trivial vector class would do - why need a full matrix/vector library? | 10:43 |
| cahirwpz | tansell, I did not suggest any big solution ;-) | 10:44 |
| cahirwpz | tansell, I only suggested that there's need to find a library to shorten such code | 10:44 |
| cahirwpz | maybe something small like that: http://code.google.com/p/pyeuclid/ | 10:45 |
| tpb | Title: pyeuclid - Project Hosting on Google Code (at code.google.com) | 10:45 |
| cahirwpz | it fits in a single file | 10:45 |
| cahirwpz | I would prefer to leave decision up to you | 10:46 |
| cahirwpz | you probably have got a better idea what kind of 3d math library is needed in tpserver-py | 10:47 |
| cahirwpz | tansell, quick question | 10:52 |
| cahirwpz | http://github.com/cahirwpz/tpserver-py/blob/master/tp/server/rules/base/objects/Fleet.py | 10:52 |
| tpb | <http://ln-s.net/7Fn7> (at github.com) | 10:52 |
| cahirwpz | lines 38-42 | 10:52 |
| cahirwpz | is it a list of actions that can be applied to an object (Fleet in this case) ? | 10:53 |
| tansell | cahirwpz, I want something which doesn't require an external dependency | 10:53 |
| tansell | cahirwpz, preferably something which is short too | 10:53 |
| cahirwpz | tansell, euclid looks perfect - single file, imports only standard python modules | 10:54 |
| cahirwpz | http://pyeuclid.googlecode.com/svn/trunk/euclid.py | 10:55 |
| tpb | <http://ln-s.net/7Fo_> (at pyeuclid.googlecode.com) | 10:55 |
| tansell | cahirwpz, it's a list of orders which are valid for an object | 10:56 |
| cahirwpz | we can do it (math library) ourselves but obviously it's reinventing a wheel and a huge waste of time | 10:56 |
| cahirwpz | tansell, valid == applicable ? | 10:57 |
| cahirwpz | tansell, because I cannot say that splitting one ship fleet is valid action | 10:58 |
| cahirwpz | you can apply such action to the fleet, but it should return that request is invalid | 10:59 |
| tansell | cahirwpz, it's not that easy | 10:59 |
| cahirwpz | O.o | 10:59 |
| tansell | there might be a merge fleet order before the split fleet order | 10:59 |
| cahirwpz | aahhh... that's another thing I don't like - Order class responsibilities is mixed with Action class stuff | 11:03 |
| cahirwpz | * are | 11:03 |
| cahirwpz | I'd like to separate classes that represent data (Orders) from classes that operate on data (Actions) | 11:04 |
| cahirwpz | that seems more logical to me | 11:05 |
| tansell | Actions are things which only happen server side | 11:12 |
| tansell | and are generally more general | 11:12 |
| tansell | things like cleaning up dead objects from the universe | 11:13 |
| tansell | and such | 11:13 |
| tansell | orders are specific things that an object does | 11:13 |
| cahirwpz | AFAIK each order is associated with an action | 11:15 |
| cahirwpz | tansell, have a look at http://github.com/cahirwpz/tpserver-py/blob/master/tp/server/rules/base/orders/Colonise.py | 11:16 |
| tpb | <http://ln-s.net/7Fp1> (at github.com) | 11:16 |
| tansell | I'm using Actions/Order in the terms tpserver-py does | 11:16 |
| cahirwpz | IMHO do() method should be put into separate class (ColoniseAction?) and should require a reference to ColoniseOrder when called | 11:17 |
| tansell | cahirwpz, why would that help? | 11:17 |
| tansell | cahirwpz, everything the ColoniseAction needs is in the ColoniseOrder | 11:17 |
| cahirwpz | tansell, separation of concern, cleaning up class hierarchy | 11:18 |
| tansell | there is no point in an Order which has no effect | 11:19 |
| cahirwpz | ColoniseOrder does inherit Object (implicitly) | 11:19 |
| cahirwpz | in old tp-server explicitly | 11:19 |
| cahirwpz | but should also inherit Action | 11:19 |
| cahirwpz | ouch... | 11:20 |
| cahirwpz | ColoniseOrder inherits Order, of course | 11:20 |
| cahirwpz | so in the end ColoniseOrder should inherit Order and Action - possibly having too much of responsibility | 11:21 |
| cahirwpz | "everything the ColoniseAction needs is in the ColoniseOrder" - I'd rather say that some Actions are associated with Orders, and some not - that's exactly the place you should introduce two kind of different Action classes | 11:23 |
| cahirwpz | http://stackoverflow.com/questions/760473/inheritance-vs-composition-for-testability - I think they describe what I really mean | 11:26 |
| tpb | <http://ln-s.net/7FpE> (at stackoverflow.com) | 11:26 |
| cahirwpz | I would avoid deriving ColoniseOrder from Order and Action, and I'd split it in into two classes | 11:27 |
| cahirwpz | ColoniseAction would comprise ColoniseOrder | 11:28 |
| cahirwpz | I'd gladly see most of Orders reorganized in this way | 11:29 |
| cahirwpz | moreover - Actions does not need to be represented in model, where Orders do - that makes them quite different | 11:31 |
| *** null_000 has quit IRC | 11:31 | |
| tansell | well I need to head out now | 11:37 |
| tansell | cahirwpz, it seems to me to add just another level of complexity for no gain | 11:38 |
| cahirwpz | please read stackoverflow link | 11:38 |
| tansell | cahirwpz, I did and it is not very convincing | 11:38 |
| cahirwpz | hmm... | 11:38 |
| cahirwpz | coloniseAction.order = coloniseOrder, and changing few lines in do() body - from self.attribute to self.order.attribute - that's not very complicated | 11:42 |
| cahirwpz | anyway - for now it's not crucial but I'll return to this topic soon | 11:43 |
| tansell | but what is the advantage? | 11:44 |
| *** mithro has quit IRC | 11:44 | |
| cahirwpz | tansell, smaller classes, cleaner responsibility, class hierarchy based on single inheritance | 11:44 |
| cahirwpz | tansell, purely non-functional improvements | 11:45 |
| cahirwpz | tansell, try to think of it in following terms: Order is not an Action, order placing is an action; after order is placed it will be served by some action in near future; depending on order type some action is selected to satisfy it; action may need data contained in order to perform | 11:58 |
| *** Agon has joined #tp | 11:59 | |
| *** mithro has joined #tp | 12:21 | |
| *** Agon has quit IRC | 13:39 | |
| *** null_000 has joined #tp | 14:04 | |
| null_000 | hi everyone | 14:04 |
| null_000 | alanp: ping | 14:04 |
| *** verhoevenv has joined #tp | 14:13 | |
| *** mithro has quit IRC | 14:43 | |
| alanp | null_000: pong | 14:50 |
| null_000 | alanp: hey | 14:56 |
| alanp | what's up, i just saw on twitter that llnz committed something to mtsec | 14:56 |
| null_000 | alanp: have to check it out | 14:56 |
| null_000 | midterm evaluations have started | 14:57 |
| alanp | =D | 14:57 |
| null_000 | is there anything special I need to do? (apart from the questionnaire) | 14:57 |
| alanp | his commit message is fixed buildfleet, i';ll take a look at what happened | 14:58 |
| alanp | no, i don't think so | 14:58 |
| null_000 | I'll compile and test it... hope it works | 15:00 |
| alanp | i think it's on the main branch though | 15:00 |
| alanp | are you still having problems with it? | 15:00 |
| null_000 | build fleet bug was one of the main problems | 15:05 |
| null_000 | I'll have to check if there is anything else preventing me from using it | 15:05 |
| alanp | cool | 15:07 |
| alanp | i'll start killing some of the smaller bugs as well | 15:09 |
| alanp | let me know how the master branch is working for you now | 15:09 |
| *** ezod has quit IRC | 15:25 | |
| *** Agon has joined #tp | 15:49 | |
| *** Epyon has joined #tp | 15:55 | |
| *** bisc has quit IRC | 16:57 | |
| *** Erroneous has joined #tp | 17:06 | |
| *** cahirwpz has quit IRC | 17:24 | |
| *** glew has joined #tp | 17:44 | |
| *** ezod has joined #tp | 17:51 | |
| *** StrangerDanger has quit IRC | 18:25 | |
| *** StrangerDanger has joined #tp | 18:37 | |
| *** StupidIncarnate has joined #tp | 19:16 | |
| *** null_000 has quit IRC | 19:22 | |
| *** llnz has joined #tp | 19:22 | |
| llnz | morning all | 19:22 |
| StupidIncarnate | morning | 19:26 |
| Agon | hello | 19:26 |
| llnz | hi Agon | 19:33 |
| llnz | see my comment on your commit (on github)? | 19:33 |
| * llnz trys to remember who was having trouble with MTSec | 19:37 | |
| Agon | yes, I do. I changed the code. Is it normal that double x = 0.001D; C++ code cant be compiled, because my compiler does? | 19:37 |
| llnz | i don't know | 19:38 |
| llnz | what compiler are you using? | 19:38 |
| Agon | I have problems with minisec, libtpproto-cpp and objects with size of 0 | 19:38 |
| Agon | gcc (SUSE Linux) 4.5.0 20100604 [gcc-4_5-branch revision 160292] | 19:38 |
| llnz | g++ (Debian 4.4.4-6) 4.4.4 here | 19:42 |
| llnz | oh? | 19:43 |
| *** Epyon has quit IRC | 19:46 | |
| Agon | yes, if the getName() method of a ObjectParameter returns "Size" I do a static_cast<SizeObjectParam*> cast on the ObjectParameter and the debuger tells me the value size is 0. Name and description are correct. | 19:47 |
| llnz | odd, but you shouldn't use the parameter name to determin the type | 19:48 |
| llnz | using a dynamic_cast or the ObjectParameterVisitor | 19:49 |
| Agon | I read that dynamic_cast requires some kind of tracking, I dont know much about type casting. | 19:50 |
| *** StrangerDanger has quit IRC | 19:52 | |
| Agon | I will adjust my code and use the ObjectParameterVisitor, its a nice idea. | 19:54 |
| *** epyon-sa1ura has joined #tp | 19:57 | |
| alanp | llnz: we were having issues with submitting orders on a planet once colonized | 20:01 |
| llnz | right, what branch and what client? | 20:02 |
| llnz | (and what dev work is being done?) | 20:02 |
| *** StrangerDanger has joined #tp | 20:09 | |
| alanp | an older mtsec branch | 20:10 |
| alanp | i'll test latest tonight | 20:10 |
| alanp | what do you mean, what dev work is being done? on the problem? | 20:10 |
| alanp | i've had my hands tied with other stuff lately, but some of that is coming to a conclusion now | 20:10 |
| alanp | although i'm going on vaca in a week | 20:10 |
| llnz | so master branch has a number of fixes for mtsec, but a few bugs still | 20:11 |
| llnz | can | 20:11 |
| llnz | can't test some because of libtpclient-py bugs | 20:11 |
| llnz | (in tp04) | 20:11 |
| alanp | which? | 20:12 |
| alanp | which bugs in master, which in tp? | 20:12 |
| alanp | tp04* | 20:12 |
| alanp | i can devote probably ~6 hours this weekend | 20:13 |
| llnz | so i can't test the problem you are having because the tpclient-pywx branch tp04 has had it's reference to libtpclient-py updated for the hack around | 20:15 |
| alanp | hack around? | 20:17 |
| llnz | or not, hold on | 20:17 |
| alanp | i remember there benig some problem with a reference | 20:17 |
| alanp | i think i just had an older tpclient | 20:17 |
| llnz | i just pulled that update, didn't make a difference | 20:23 |
| epyon-sa1ura | dammit | 20:25 |
| llnz | getting "ValueError: The object isn't new enough to update the dictionary with! Current 2010-07-15 08:22:38, update 2010-07-15 08:19:01" when downloading universe info for after when a fleet arrives | 20:25 |
| llnz | hi epyon-sa1ura | 20:25 |
| epyon-sa1ura | llnz: I'll join you later, my main PC just died... | 20:26 |
| llnz | ok | 20:26 |
| epyon-sa1ura | honestly, things can't get much worse dammit | 20:26 |
| Agon | llnz: How is the ObjectParameterVisitor designed to use? I overloaded the class and all its functions, called the visitObjectParameter method with a ObjectParameter value but the compiler doesn't like ambiguous overloaded calls. | 20:33 |
| llnz | Agon: the other way around, pass you ObjectParameterVisitor to the visit method of the ObjectParameter | 20:35 |
| llnz | the objectParameter will then call the correct method from the ObjectParameterVisitor | 20:39 |
| Agon | ah thanks | 20:40 |
| llnz | epyon-sa1ura: just compiled up your code | 20:51 |
| llnz | only thing to fix is the move the mysql includes from mysqlpersistence.cpp to mysqlpersistence.h | 20:52 |
| llnz | *after* the include for boost mutex | 20:52 |
| epyon-sa1ura | llnz: there aren't many changes | 20:52 |
| epyon-sa1ura | main work is on that damn PC I'm trying to bring up now | 20:53 |
| llnz | i'm happy with you committing non-working code to your repo | 20:55 |
| llnz | if you want | 20:55 |
| llnz | might be a good idea | 20:55 |
| epyon-sa1ura | I'll try to recover the files at least | 20:55 |
| epyon-sa1ura | llnz: how about I setup a separate folder for the new core? | 20:55 |
| epyon-sa1ura | it will be temporary -- afterwards I'll move the files into the normal folder | 20:56 |
| llnz | that's a good idea, but don't move it back afterwards | 20:57 |
| llnz | split the core of the server into logical directories | 20:58 |
| llnz | (or modules, if you will, but not loadable modules) | 20:58 |
| llnz | we already have tinyxml and battlexml directories | 20:59 |
| llnz | having a server directory is a good move | 21:00 |
| epyon-sa1ura | hmm | 21:08 |
| Agon | llnz: I am using the ObjectParameterVisitor, now, great idea. But I am still getting object sizes of 0 and ResourceLists with 0 items. | 21:13 |
| llnz | Agon: what object? | 21:14 |
| llnz | oh, must be a planet | 21:14 |
| llnz | which planet? | 21:15 |
| Agon | all objects I receive which have a Position3dObjectParam | 21:17 |
| Agon | and for the size of 0: all objects which have a SizeObjectParam | 21:18 |
| Agon | And I receive same positions for multiple objects | 21:19 |
| llnz | ok, will look into it | 21:21 |
| llnz | it is possible for multiple objects to be at the same position | 21:21 |
| llnz | Agon: what ruleset? | 21:25 |
| *** StupidIncarnate has quit IRC | 21:25 | |
| Agon | llnz: quickstart-minisec.conf | 21:26 |
| llnz | the universe should have a size of 1000000000000 | 21:30 |
| llnz | planets have a size of 2 | 21:30 |
| llnz | and star systems around 30000-90000 | 21:31 |
| Agon | oh so I cant use the size to draw different sized planets :( | 21:33 |
| llnz | well, the size could change | 22:00 |
| llnz | be different for different planets | 22:01 |
| *** Agon has quit IRC | 22:08 | |
| *** Agon-laptop has joined #tp | 22:09 | |
| llnz | bbs, going to work | 22:22 |
| *** llnz has quit IRC | 22:22 | |
| *** StrangerDanger has quit IRC | 22:52 | |
| *** StrangerDanger has joined #tp | 22:54 | |
| *** llnz has joined #tp | 23:37 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!