| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 | ||
| 11:50 | FIXED |   |
| 11:49 | STUPIDITY |  The Dijkstra out-of-bounds protector needed a "return" command, since it was missing the code causing the error was still executed... |
| 11:48 | COCKROACH |  All fixes ignored??? |
| 11:45 | FIXED | A little NIL translation issue |
| 11:44 | CONFIRMED |  Indeed, Dijkstra and Kthura were not in one mind when it came to how to use booleans for passible or impassible areas.... I made an extra routine in Kthura to make the reverse, but I also kept the original.... This to prevent compatibility issues with other stuff.... |
| 11:42 | NOTE |  In the meantime my backup has finished so I can get back to "business as usual" mode |
| 11:28 | FIXED | What I thought is true, Dijkstra did have an out-of-bounds protection during the route calculation, but not during the check of the starting and end coordinates. If you request starting and ending coordinates outside the bouds of the blockmap, the path calculation will now always result into a failed empty route |
| 11:19 | BUG |   A few things are clear, though
|
| 11:09 | DEBUG |  An extra debug line will have to provide me some information on the reason WHY! |
| 11:06 | NOTE | This is a very common bug I had with the pathfinder in the original Dyrt as well, also when I was on Star Story, and my 63 fires of lung.... Only TFT REVAMPED didn't suffer here, but that was only because TFT REVAMPED and Star Story use the same engine, so the issue was fixed before that project began XD |
| 11:05 | BUG | Wherever I click NOTHING happens at all... |
| 11:03 | TEST |  Take II |
| 11:03 | FIXED | Illegal group call |
| 11:01 | TEST | Let's see what happens.. Take I |
| 11:00 | LINK |  And now the command to make Irravonia actually walk has been given, meaning she should walk now to the location the player directs her to, *if* in fact she can reach it... Please note, due to the way the blockmap has been set up, Little Irravonia will keep herself to a 32x32 grid |
| 10:44 | TEST | So I can start my test anyway, as I don't wanna waste time waiting for my backupper to finish its job (and I really need to fix some issues with that tool asap). |
| 10:43 | POWERSHELL |  Well luckily Powershell grants me a few quick commands to "void" that.... |
| 10:41 | FAILURE |  Unfortunately my backup tool got a bit stuck in the same window from where I run my building scripts... I hate it when that happens.... :-/ |
| 10:40 | TEST | And let's see what that'll do! |
| 10:40 | DEBUG | For now I've limited myself to making sure the system understand where you want the hero to walk to |
| 10:40 | DONE |  I've Made a "dummy" field... It does nothing yet, but at least this makes it easier on me to make 100% sure I can easily do alternate movements, like clickable objects, and that is very important for NPCs, savespots, signs and other things you can find in your everyday RPG dungeon or city. |
| 10:39 | DONE | I'll make a few worksouts in preparation to use Dijkstra though.... I got some time before I really must quit my session today |
| - = 10 Aug 2019 = - | ||
| 23:01 | DONE | I needed a little break! Since the walking around routine is also more complex than just setting up a Dijkstra calculation whenever the player clicks something (as there are also clickable objects in the field), this may be something that won't be done until after the weekend, but when I say such things, you never know... You should all know by know how quickly I can change my mind when it comes to these things.... |
| 22:56 | TEST | I did a test and no parse errors came up, which is so far a good thing |
| 22:51 | STATUS |  All i can do now is to make sure all parse errors are taken care of... As the game doesn't yet have any controles allowing you to move, the Dijkstra routines are not called or anything yet. |
| 22:50 | TODO |  #27 (And it doesn't look important, as this causes under-the-hood-issues, the player may never see, but memory leaks (even when they are resolved as soon as the game ends) are always a bad thing) |
| 22:41 | LINK |  And the NIL link code has been created |
| 22:35 | LINK | Or rather a 3rd as a WalkToSpot was also there |
| 22:34 | LINK | The C# WalkToCoords has been written.... That is half of the contract |
| 22:22 | TODO | Of course, now I need to write the proper link code for Bubble, which is in this case for NALA |
| 22:16 | KTHURA |  Moving in general and also in accordance with Dijkstra |
| 21:13 | KTHURA | Kthura is now able to calculate the route to walk over the Kthura Map, thanks to the underlying Dijkstra unit.... I must note though, that the route will be calculated, but the result will be ignored as I do not have th proper walking update stuff available yet.... (can't do all in once you know). |
| 19:37 | NOTE | (This will require me to include Dijkstra also in the Kthura Map Editor source code, but that is a later concern). |
| 19:36 | KTHURA | I've now set the path finding variables to work with my own Dijkstra module |
| 19:34 | LINK | Dijkstra Has now been linked to NALA, so I can use it in Kthura |
| 18:40 | CONFIRMED | I'v deliberately created a track that was impossible to get from start from finish to see if this would indeed get a 0 path and a "false" for success rate... This happens to be the case |
| 17:04 | STATUS | Time is however not on my side to actually attach this code to Kthura and to Bubble, but that doesn't matter right now... It works, and that's important ;) |
| 16:56 | STUPIDITY | And how come I never realized how simple this algorithm was..... :-/ |
| 16:56 | FIXED | I've been working things out on the Nodes parents, and NOW it works.... And both the demo map and the breakdown confirm that.... WOOOHOOOO! |
| 16:52 | REMOVED |  .git folder form JCR6 in the test project |
| 16:52 | BUG | But seeing this mapped out, with a complete breakdown list, this no longer makes sense.... Some nodes have been correctly generated, yes, but been linked to the wrong parent.... |
| 16:35 | CONFIRMED | I could not yet see the actual route, but.... What I did see was that the route was calculated to be 24 units long, and guess what, when I walk the route manually as a human, I come to the same calculation.... This could already be a good indication, the Dijkstra routine works.... |
| 16:33 | TEST | Testing stuff |
| 16:26 | LINK | But as I'm now just in testing stuff, not gonna do that, and let's just add JCR6 so I can get this over with! |
| 16:26 | NOTE | Perhaps I need to revise qstr, as I wonder why a quickstring routine needs JCR6 support ;_/ |
| 16:24 | FIXED | Error fixing |
| 16:24 | VISUALSTUDIO |  Quick Test Program set up |
| 16:13 | NOTE | Incoming phone call delayed me |
| 14:58 | TODO | Before I can test this I need to write some test code though..... |
| 14:58 | DONE | Frankly speaking I SHOULD have set up a Dijkstra based pathfinder now, however if it's robuust or working at all.... I can't tell, so a test is in order. |
| 13:51 | POWERSHELL | Created empty Dijkstra.cs file |
| 13:50 | STATUS |  And with that I think I removed everything that can cause me some trouble |
| 13:49 | REMOVED | Quick form calls |
| 13:49 | REMOVED | Backups also not needed |
| 13:48 | REMOVED | I won't use GTK in these tests |
| 13:48 | REMOVED | Unneeded documents |
| 13:48 | REMOVED | So I removed at least the link the git-crap |
| 13:47 | LINK | Now I did link in the entire TrickyUnits folder, but I surely won't be needing that XD |
| 13:46 | VISUALSTUDIO | For now I will focus on the pathfinder alone as a separate project.... If succesful I got my own pathfinder which I can later on also export to other languages, so there's a lot to win for me here :P |
| 13:32 | STUDY |  Wait a minute.... I think I can create a kind of Dijkstra based pathfinder myself now.... The step by step description I see now is the first one set up in a way that maks sense.... Perfect.... |
| 13:25 | GITHUB |  I've made a request in AnnSerba/A-star#1 ... This can prevent that I may have to remove the code Ann Serba provided has to be removed in the future, wasting a lot of my time in the process... |
| 12:45 | STUDY | I've studied the pathfinder routine for C# I found... They do not look too difficult to understand and use, however they came without any license terms, and that could be a bit of a bummer. Now NALA has been released under a GPL license, but Kthura as a library has not been so, I really need to work this out in a way licenses cannot be an issue. |
| 11:44 | DUMMIED |  The debug line for faders.... I don't think I need that anymore, as it appears to be working |
| 11:42 | JUDGMENT |  APpears to work |
| 11:40 | TEST | Let's try that again, shall we? |
| 11:40 | FIXED | SASKIA typo |
| 11:39 | SASKIA |  Of course, I gotta try if this all works |
| 11:32 | ENHANCEMENT |  I've made it possible for the autoscroller to force being executed. This will allow me to make sure SASKIA scripts can force the camera to the correct position if needed |
| 0:25 | VISUALSTUDIO | I've created a QuickNIL tool which can help me to make prototyping possible ;) |
| - = 9 Aug 2019 = - | ||
| 23:13 | FIXED | Okay, I guess it was either the visible parameter or the alpha parameter, anyway, I altered them both, and that fixed the issue, case closed :P |
| 23:09 | EXPERIMENT |  Let's see.... |
| 22:44 | MYSTERY |  It will still be a great mystery to sort out why Irravonia doesn't show.... Perhaps the core, perhaps the MonoGame driver.... It's always a bit of an issue |
| 22:39 | TEST | And take XXIX |
| 22:39 | POWERSHELL | Run build script |
| 22:39 | DONE | And the Kthura core will also copy the spawn points dominance value into the actor... That is very extremely important, and could also be the main reason why little Irravonia didn't appear |
| 22:38 | DONE | Made sure the SASKIA script will remap |
| 22:28 | BUG |  Almost there, but the autoscroll had to activate immediately and not after Irravonia speaking + Irravonia isn't seen.... The latter could just be a remap issue |
| 22:25 | TEST | Take XXVIII |
| 22:24 | FIXED | Fine tuning |
| 22:19 | TEST | Take XXVII |
| 22:19 | SOLVED |  Ah.... That explains.... I think |
| 22:15 | HUH |  The error indicates something that is NOT right.... |
| 21:30 | TEST | Take XXVI |
| 21:30 | FIXED | Did I fix this? |
| 21:25 | FUCKYOU |  It appears the leader isn't recognized for some odd reason |
| 21:24 | TEST | Take XXV |
| 21:24 | DEBUG | extra debug must confirm this and give more info |
| 21:23 | HUH | No frames loaded |
| 21:16 | TEST | Take XXVI |
| 21:16 | DEBUG | More info |
| 21:13 | RESULT |   |
| 21:11 | FUCKYOU | I hate the NIL bug that ignores the keyword "nil" when it's the final word of the line |
| 21:11 | TEST | And Take XXV |
| 21:09 | TEST | Take XXIII and XXIV |
| 21:05 | FIXED | I keep forgetting that "each" requires the tables library, which is not loaded when internal scripts are being processed :( |
| 21:04 | DEBUG | Another vs sync and debug run, then ,eh? |
| 20:56 | BUG | Index out of array range, eh? (C# error). |
| 20:52 | TEST | Take XXII |
| 20:52 | FIXED | At least I fixed another case error, so I guess that other fix is also a real fix |
| 20:51 | FIXED | I think I fixed that.... (or not) |
| 20:48 | TEST | Take XXI |
| 20:47 | EXPERIMENT |  Maybe this fixes it? |
| 20:36 | SOLVED | Right, the bug is that the picture file to be attached to the player sprite is never properly defined.... Good to know.... |
| 20:34 | TEST | Take XX |
| 20:32 | DEBUG | Will this answer a few questions? |
| 20:26 | TECHNO |  This all looks a bit obscure, so I hope I can get to the bottom of this... At least I know that the texture loader is to blame |
| 20:24 | RESULT |  |
| 20:24 | TEST | Take XIX |
| 20:22 | DEBUG | I've added a way to traceback C# as I need to look at that.... |
| 20:16 | DONE | Let's prepare for running in VisualStudio |
| 19:45 | STATUS | I'm gonna need VisualStudio to nail this one... But since this is affecting the core of Kthura, I guess I'll need VisualStudio to fix all this anyway |
| 19:43 | DEBUG | I've added an extra line which must tell me |
| 19:43 | UNDESIREABLE |  I need to sort out if older Kthura maps do fall through the garbage collector or not... Some information the debugger gave me, make me think not.... |
| 19:32 | TEST | Take XVIII |
| 19:32 | CONFIRMED | It does appear to confirm this takes place in the kthura drawing algorithms, and that will make it a rough ride to pinpoint the source of the bug.... |
| 19:31 | FIXED | Of course you realize I fixed this now! |
| 19:31 | STUPIDITY | Stacking overflow due to bad C# overloading.... Yes, I'm a first class moron! |
| 19:30 | FAILURE | Not quite sure what is happening, but everything is freezing, even the music freezes... This is really really bad! |
| 19:26 | TEST | Take XVII |
| 19:26 | DEBUG | I've expanded some data... hopefully this gives some answers (I'm afraid not, but hey, whatever I can narrow down is good, right?) |
| 19:22 | DEBUG | Unfortunately the traceback information is too vague |
| 19:16 | TECHNO | This will be one of the harder errors to track down.... This information is too minimal to do something with, however the traceback can at least show me which call from NIL is responsible, and hopefully this can narrow down my options. |
| 19:13 | BUG |  |
| 19:07 | TEST | Take XVI |
| 19:07 | FIXED | I fixed that |
| 19:07 | STUPIDITY |  SAKSKIA requires strings to be quoted! Just like normal NIL and Lua, dumbo! |
| 19:03 | TEST | Take XV |
| 19:03 | FIXED | Well I fixed that |
| 19:03 | STUPIDITY | Because I forgot to set the "spot" parameter in the SASKIA Script |
| 19:02 | COCKROACH | What I need to figure out is why the "spot" parameter is 'nil'.... |
| 19:00 | COSMETIC |   And I made sure the mouse pointer wouldn't be darker in the field. Not that important, but hey I saw that this was not right, so why not set it right before I forget it! ;) |
| 18:59 | STATUS | What I can at least tell is that the part without player actors appears to work again |
| 18:55 | TEST | Take XVI |
| 18:55 | FIXED | Did I fix it now? |
| 18:55 | FUCKYOU | Oh, crap! |
| 18:52 | TEST | Take XV |
| 18:52 | FIXED | I think I fixed it (hard to tell) |
| 18:51 | BUG | The game demands a player, which the prison can (initally) not give |
| 18:47 | TEST | Take XIV |
| 18:47 | FIXED | Case errors |
| 18:40 | TEST | Take XIII |
| 18:40 | FIXED | Or at least, I think so.... |
| 18:39 | BUG | But to the Field flower.... |
| 18:39 | STATUS | Well the error I get now does at least no longer refer to the Kthura link code.... |
| 18:37 | OFFTOPIC |  Well we get to take MCMLXXV or not? |
| 18:36 | TEST | Take XII |
| 18:36 | FIXED | A leftover from old code |
| 18:35 | BUG | The trouble doesn't end there... of course |
| 18:35 | CONFIRMED | Well at least that one is fixed |
| 18:32 | TEST |  Take XI will tell! |
| 18:32 | FIXED | Did I fix it? |
| 18:32 | NOTE | Or not? |
| 18:31 | BUG | right "nil" appears to be the parent, which I actually already expected, but at least that's confirmed now.... This is gonna be some hell to debug, I'm afraid |
| 18:28 | TEST | Take X |
| 18:28 | DEBUG | Expanded error message so I can (hopefully) get a better view on what exactly goes wrong |
| 18:27 | KTHURA | Link script chunk properly named |
| 17:08 | TODO | A break... I need to clear my brain a little |
| 17:07 | BUG | "None or incorrect parent given!" Now this will be a best to debug |
| 17:05 | TEST | Take IX |
| 17:05 | FIXED | Indexing error |
| 17:04 | FUCKYOU | GRRRRR! |
| 17:03 | TEST | Take VIII |
| 17:02 | FIXED | A few random occuranaces |
| 17:00 | TEST | Take VII |
| 17:00 | FIXED | So this time it DOES count as a fix |
| 17:00 | CONFIRMED | The "Unknown term" error that pops up thanks to this message confirmed my earlier statement... however this was a var I forgot to rename |
| 16:57 | TEST | VI |
| 16:57 | NOTE | THAT is NOT a fix... something that ain't broken, can't be fixed.... At least I hope to prevent the false crash errors this way |
| 16:56 | DONE | I've wrote some voidance code which hopefully gets me around the (incorrect) dupe identifier error |
| 16:52 | STATUS | It appears all link code errors have been accounted for |
| 16:49 | TEST | Take V |
| 16:49 | FIXED | Bad function type |
| 16:46 | STATUS | (And I've only just begun, I'm afraid)..... |
| 16:45 | TEST | Take IV |
| 16:45 | FIXED |  Another wrong variable |
| 16:41 | TEST | Take III |
| 16:41 | FIXED | Right-o |
| 16:38 | TEST | And onto TAKE II |
| 16:38 | NALA |  Recompiling |
| 16:38 | FIXED | Yup... First issue coming in! |
| 16:36 | TEST | TAKE I (of the many I may require, I'm afraid, as this was pretty complicated) |
| 16:36 | DONE | Mesa Back |
| 15:55 | FAILURE | I only discovered that I ran out of bread, so I must make a run to the store first, to make sure I won't starve soon :P |
| 15:55 | STATUS | I have no clue at all if this all works (I expect loads of bugs and cockroaches tbh), but this is the first moment on which testing is possible, at all... |
| 15:17 | NIL |  Automated updater for the actor texture |
| 15:13 | TECHNO | And I was glad I found this one in time, as the cockroaches this one could have caused are beyond imagination! |
| 15:12 | FIXED | Code in the incorrect file moved to the correct one |
| 15:11 | LINK | And the linkup code has also been written |
| 15:10 | C# |  Texture configuration possible through the API now |
| 14:42 | LINK | Linkup code done |
| 13:16 | C# | I've done the cores of getting and setting textures.... For now I'll only link this for the actors |
| 13:04 | DONE | I've set up the Leader property. This will be used now to determine whose sprites will appear as the Player Actor... VERRRRY important! |
| 12:56 | CLOSED |  #25 (I should have done that before) |
| 12:54 | UPDATED | |
| 12:34 | STATUS | This can take awhile, as this is hard, boring and totally unsatisfying work, but unfortunately, it must be done, or the game cannot work in any way.... |
| 12:33 | NIL | Main code for "SpawnPlayer" has been set up... Well, most of it, as some important parts yet remain to be done... |
| 9:24 | LINK | Linking code for "SpawnPlayer" |
| 0:38 | STATUS | A full workout will be done later, as I'm tired now |
| 0:36 | ENHANCEMENT | I've enhanceed the flow manager of BUBBLE with a way to transfer simplisstic data between states and flows without having to rely on superglobals |
| - = 8 Aug 2019 = - | ||
| 23:57 | FIXED | So I rewrote that part |
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 | ||
