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
11:50FIXED
FIXEDRebecca
11:49STUPIDITY
STUPIDITYThe Dijkstra out-of-bounds protector needed a "return" command, since it was missing the code causing the error was still executed...
11:48COCKROACH
COCKROACHAll fixes ignored???
11:45FIXED
FIXEDA little NIL translation issue
11:44CONFIRMED
CONFIRMEDIndeed, 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....
11:42NOTE
NOTEIn the meantime my backup has finished so I can get back to "business as usual" mode
11:28FIXED
FIXEDWhat 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

11:19BUG
BUG

A few things are clear, though

  1. The coordinates are correctly adapted to the blockmap grid, so at least that is not the issue
  2. It is also clear that the protection for walking out of bounds is not really operative... The test program did make the route touch the edges (on purpose) to see if that would properly block things.... So I must fix Dijkstra itself to always reject out-of-bounds walking requests
  3. I do not know if Irravonia not walking is an actual failure in the route calculation, or the blockmap, but I think I got an idea on this though, as I think Dijkstra and blockmap have reversed meanings on True and False, but I need to verify that one
11:09DEBUG
DEBUGAn extra debug line will have to provide me some information on the reason WHY!
11:06NOTE
NOTEThis 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
11:05BUG
BUGWherever I click NOTHING happens at all...
11:03TEST
TESTTake II
11:03FIXED
FIXEDIllegal group call
11:01TEST
TESTLet's see what happens.. Take I
11:00LINK
LINKAnd 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
10:44TEST
TESTSo 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).
10:43POWERSHELL
POWERSHELLWell luckily Powershell grants me a few quick commands to "void" that....
10:41FAILURE
FAILUREUnfortunately my backup tool got a bit stuck in the same window from where I run my building scripts... I hate it when that happens.... :-/
10:40TEST
TESTAnd let's see what that'll do!
10:40DEBUG
DEBUGFor now I've limited myself to making sure the system understand where you want the hero to walk to
10:40DONE
DONEI'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.
10:39DONE
DONEI'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 = -
23:01DONE
DONEI 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....

22:56TEST
TESTI did a test and no parse errors came up, which is so far a good thing
22:51STATUS
STATUSAll 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.
22:50TODO
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)
22:41LINK
LINKScyndi And the NIL link code has been created
22:35LINK
LINKOr rather a 3rd as a WalkToSpot was also there
22:34LINK
LINKThe C# WalkToCoords has been written.... That is half of the contract
22:22TODO
TODOOf course, now I need to write the proper link code for Bubble, which is in this case for NALA
22:16KTHURA
KTHURAMoving in general and also in accordance with Dijkstra
21:13KTHURA
KTHURAKthura 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).
19:37NOTE
NOTE(This will require me to include Dijkstra also in the Kthura Map Editor source code, but that is a later concern).
19:36KTHURA
KTHURAI've now set the path finding variables to work with my own Dijkstra module
19:34LINK
LINKDijkstra Has now been linked to NALA, so I can use it in Kthura
18:40CONFIRMED
CONFIRMEDI'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
17:04STATUS
STATUSTime 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 ;)
16:56STUPIDITY
STUPIDITYAnd how come I never realized how simple this algorithm was..... :-/
16:56FIXED
FIXEDI've been working things out on the Nodes parents, and NOW it works.... And both the demo map and the breakdown confirm that.... WOOOHOOOO!
16:52REMOVED
REMOVED.git folder form JCR6 in the test project
16:52BUG
BUGBut 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....
16:35CONFIRMED
CONFIRMEDI 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....
16:33TEST
TESTTesting stuff
16:26LINK
LINKBut 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!
16:26NOTE
NOTEPerhaps I need to revise qstr, as I wonder why a quickstring routine needs JCR6 support ;_/
16:24FIXED
FIXEDError fixing
16:24VISUALSTUDIO
VISUALSTUDIOQuick Test Program set up
16:13NOTE
NOTEIncoming phone call delayed me
14:58TODO
TODOBefore I can test this I need to write some test code though.....
14:58DONE
DONEFrankly 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.
13:51POWERSHELL
POWERSHELLCreated empty Dijkstra.cs file
13:50STATUS
STATUSBrendor And with that I think I removed everything that can cause me some trouble
13:49REMOVED
REMOVEDQuick form calls
13:49REMOVED
REMOVEDBackups also not needed
13:48REMOVED
REMOVEDI won't use GTK in these tests
13:48REMOVED
REMOVEDUnneeded documents
13:48REMOVED
REMOVEDSo I removed at least the link the git-crap
13:47LINK
LINKNow I did link in the entire TrickyUnits folder, but I surely won't be needing that XD
13:46VISUALSTUDIO
VISUALSTUDIOFor 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
13:32STUDY
STUDYWait 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....
13:25GITHUB
GITHUBI'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...
12:45STUDY
STUDYI'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.
11:44DUMMIED
DUMMIEDThe debug line for faders.... I don't think I need that anymore, as it appears to be working
11:42JUDGMENT
JUDGMENTAPpears to work
11:40TEST
TESTLet's try that again, shall we?
11:40FIXED
FIXEDSASKIA typo
11:39SASKIA
SASKIAOf course, I gotta try if this all works
11:32ENHANCEMENT
ENHANCEMENTI'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
0:25VISUALSTUDIO
VISUALSTUDIOI've created a QuickNIL tool which can help me to make prototyping possible ;)
- = 9 Aug 2019 = -
23:13FIXED
FIXEDOkay, 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
23:09EXPERIMENT
EXPERIMENTLet's see....
22:44MYSTERY
MYSTERYIt 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
22:39TEST
TESTAnd take XXIX
22:39POWERSHELL
POWERSHELLRun build script
22:39DONE
DONEAnd 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
22:38DONE
DONEMade sure the SASKIA script will remap
22:28BUG
BUGIrravonia 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
22:25TEST
TESTTake XXVIII
22:24FIXED
FIXEDFine tuning
22:19TEST
TESTTake XXVII
22:19SOLVED
SOLVEDAh.... That explains.... I think
22:15HUH
HUHThe error indicates something that is NOT right....
21:30TEST
TESTTake XXVI
21:30FIXED
FIXEDDid I fix this?
21:25FUCKYOU
FUCKYOUIt appears the leader isn't recognized for some odd reason
21:24TEST
TESTTake XXV
21:24DEBUG
DEBUGextra debug must confirm this and give more info
21:23HUH
HUHNo frames loaded
21:16TEST
TESTTake XXVI
21:16DEBUG
DEBUGMore info
21:13RESULT
RESULT
21:11FUCKYOU
FUCKYOUI hate the NIL bug that ignores the keyword "nil" when it's the final word of the line
21:11TEST
TESTAnd Take XXV
21:09TEST
TESTTake XXIII and XXIV
21:05FIXED
FIXEDI keep forgetting that "each" requires the tables library, which is not loaded when internal scripts are being processed :(
21:04DEBUG
DEBUGAnother vs sync and debug run, then ,eh?
20:56BUG
BUGIndex out of array range, eh? (C# error).
20:52TEST
TESTTake XXII
20:52FIXED
FIXEDAt least I fixed another case error, so I guess that other fix is also a real fix
20:51FIXED
FIXEDI think I fixed that.... (or not)
20:48TEST
TESTTake XXI
20:47EXPERIMENT
EXPERIMENTEric Maybe this fixes it?
20:36SOLVED
SOLVEDRight, the bug is that the picture file to be attached to the player sprite is never properly defined.... Good to know....
20:34TEST
TESTTake XX
20:32DEBUG
DEBUGWill this answer a few questions?
20:26TECHNO
TECHNOThis 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
20:24RESULT
RESULT
20:24TEST
TESTTake XIX
20:22DEBUG
DEBUGI've added a way to traceback C# as I need to look at that....
20:16DONE
DONELet's prepare for running in VisualStudio
19:45STATUS
STATUSI'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
19:43DEBUG
DEBUGI've added an extra line which must tell me
19:43UNDESIREABLE
UNDESIREABLEI 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....
19:32TEST
TESTTake XVIII
19:32CONFIRMED
CONFIRMEDIt 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....
19:31FIXED
FIXEDOf course you realize I fixed this now!
19:31STUPIDITY
STUPIDITYStacking overflow due to bad C# overloading.... Yes, I'm a first class moron!
19:30FAILURE
FAILURENot quite sure what is happening, but everything is freezing, even the music freezes... This is really really bad!
19:26TEST
TESTTake XVII
19:26DEBUG
DEBUGI've expanded some data... hopefully this gives some answers (I'm afraid not, but hey, whatever I can narrow down is good, right?)
19:22DEBUG
DEBUGUnfortunately the traceback information is too vague
19:16TECHNO
TECHNOThis 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.
19:13BUG
BUG
19:07TEST
TESTTake XVI
19:07FIXED
FIXEDI fixed that
19:07STUPIDITY
STUPIDITYAziella SAKSKIA requires strings to be quoted! Just like normal NIL and Lua, dumbo!
19:03TEST
TESTTake XV
19:03FIXED
FIXEDWell I fixed that
19:03STUPIDITY
STUPIDITYBecause I forgot to set the "spot" parameter in the SASKIA Script
19:02COCKROACH
COCKROACHWhat I need to figure out is why the "spot" parameter is 'nil'....
19:00COSMETIC
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! ;)
18:59STATUS
STATUSWhat I can at least tell is that the part without player actors appears to work again
18:55TEST
TESTTake XVI
18:55FIXED
FIXEDDid I fix it now?
18:55FUCKYOU
FUCKYOUOh, crap!
18:52TEST
TESTTake XV
18:52FIXED
FIXEDI think I fixed it (hard to tell)
18:51BUG
BUGThe game demands a player, which the prison can (initally) not give
18:47TEST
TESTTake XIV
18:47FIXED
FIXEDCase errors
18:40TEST
TESTTake XIII
18:40FIXED
FIXEDOr at least, I think so....
18:39BUG
BUGBut to the Field flower....
18:39STATUS
STATUSWell the error I get now does at least no longer refer to the Kthura link code....
18:37OFFTOPIC
OFFTOPICWell we get to take MCMLXXV or not?
18:36TEST
TESTTake XII
18:36FIXED
FIXEDA leftover from old code
18:35BUG
BUGThe trouble doesn't end there... of course
18:35CONFIRMED
CONFIRMEDWell at least that one is fixed
18:32TEST
TESTMerya Take XI will tell!
18:32FIXED
FIXEDDid I fix it?
18:32NOTE
NOTEOr not?
18:31BUG
BUGright "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
18:28TEST
TESTTake X
18:28DEBUG
DEBUGExpanded error message so I can (hopefully) get a better view on what exactly goes wrong
18:27KTHURA
KTHURALink script chunk properly named
17:08TODO
TODOA break... I need to clear my brain a little
17:07BUG
BUG"None or incorrect parent given!" Now this will be a best to debug
17:05TEST
TESTTake IX
17:05FIXED
FIXEDIndexing error
17:04FUCKYOU
FUCKYOUGRRRRR!
17:03TEST
TESTTake VIII
17:02FIXED
FIXEDA few random occuranaces
17:00TEST
TESTTake VII
17:00FIXED
FIXEDSo this time it DOES count as a fix
17:00CONFIRMED
CONFIRMEDThe "Unknown term" error that pops up thanks to this message confirmed my earlier statement... however this was a var I forgot to rename
16:57TEST
TESTVI
16:57NOTE
NOTETHAT 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
16:56DONE
DONEI've wrote some voidance code which hopefully gets me around the (incorrect) dupe identifier error
16:52STATUS
STATUSIt appears all link code errors have been accounted for
16:49TEST
TESTTake V
16:49FIXED
FIXEDBad function type
16:46STATUS
STATUS(And I've only just begun, I'm afraid).....
16:45TEST
TESTTake IV
16:45FIXED
FIXEDDernor Another wrong variable
16:41TEST
TESTTake III
16:41FIXED
FIXEDRight-o
16:38TEST
TESTAnd onto TAKE II
16:38NALA
NALARecompiling
16:38FIXED
FIXEDYup... First issue coming in!
16:36TEST
TESTTAKE I (of the many I may require, I'm afraid, as this was pretty complicated)
16:36DONE
DONEMesa Back
15:55FAILURE
FAILUREI 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
15:55STATUS
STATUSI 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...
15:17NIL
NILAutomated updater for the actor texture
15:13TECHNO
TECHNOAnd I was glad I found this one in time, as the cockroaches this one could have caused are beyond imagination!
15:12FIXED
FIXEDCode in the incorrect file moved to the correct one
15:11LINK
LINKAnd the linkup code has also been written
15:10C#
C#Texture configuration possible through the API now
14:42LINK
LINKLinkup code done
13:16C#
C#I've done the cores of getting and setting textures.... For now I'll only link this for the actors
13:04DONE
DONEI've set up the Leader property. This will be used now to determine whose sprites will appear as the Player Actor... VERRRRY important!
12:56CLOSED
CLOSED#25 (I should have done that before)
12:54UPDATED
12:34STATUS
STATUSThis 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....
12:33NIL
NILMain code for "SpawnPlayer" has been set up... Well, most of it, as some important parts yet remain to be done...
9:24LINK
LINKLinking code for "SpawnPlayer"
0:38STATUS
STATUSA full workout will be done later, as I'm tired now
0:36ENHANCEMENT
ENHANCEMENTRebecca 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 = -
23:57FIXED
FIXEDSo 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