| 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 |
| 18:55 | VOID |  Don't ask! |
| 18:52 | TEST |  Take CMLVII |
| 18:52 | LINK |  Link up code... Far from beautiful, but IT WORKS (at least, it should) and that's all that should matter! |
| 18:48 | C++ |  compiling |
| 18:48 | C++ |  StopWalking API |
| 18:39 | TEST |  Take CMLVI |
| 18:39 | FIXED |  Syntax error x2 |
| 18:36 | TEST |  Take CMLV |
| 18:36 | FIXED |  The first required #use clause caused an error due to being missing |
| 18:34 | TEST |  Take CMLIV |
| 18:34 | SCRIPT |   NewEnemy property |
| 18:28 | CONFIRMED |  Très bien! That's what I wanted to see |
| 18:27 | TEST |  Take CMLIII |
| 18:26 | VOID |  Whatever |
| 18:24 | TEST |  Take CMLII |
| 18:24 | FIXED |  Another. How did I manage to do that? |
| 18:22 | TEST |  Take CMLI |
| 18:22 | FIXED |  Wrong operator |
| 18:15 | TEST |  Take CML |
| 18:15 | DONE |  Other approach |
| 18:13 | TEST |  Take CMXLIX |
| 18:13 | VOID |  Fuck it |
| 18:10 | TEST |  Take CMXLVIII |
| 18:10 | DONE |  added some required fields |
| 18:10 | SOLVED |  There was nothing wrong... Silly me |
| 18:06 | TEST |  Take CMXLVII |
| 18:06 | VOID |  In order to not spook this up any more than I have to let's do this with a different approach |
| 18:05 | BUG |  Now I am a bit flabbergasted by the claim that the function I created would not exist |
| 18:01 | TEST |  Take CMXLVI |
| 18:01 | VOID |  Set up start function (big moment in which little happens, but what happens is vital) |
| 17:39 | TEST |  Take CMXLV |
| 17:39 | STATUS |  Well, in order to resume I need a take to get my persoanl update ... Nothing serious! Just a memory refreshment. Don we all need that sometimes? |
| 12:05 | STATUS |  FINALLY the error message I wanted to see |
| 12:03 | TEST |  Take CMXLIV |
| 12:03 | FUCKYOU |  Got it? |
| 12:00 | TEST |  Take CMXLIII |
| 12:00 | VOID |  More lies covered? |
| 11:56 | TEST |  Take CMXLII |
| 11:56 | FIXED |  Syntax error |
| 11:55 | TEST |  Take CMXLI |
| 11:55 | VOID |  Ah, yeah, that way |
| 11:52 | TEST |  Take CMXL |
| 11:52 | DEBUG |  An impossible error message. I need to figure out why this happens. A few more debug lines must guide me here |
| 11:47 | TEST |  Take CMXXXIX |
| 11:47 | FIXED |  C-string format error |
| 11:43 | TEST |  Take CMXXXVIII |
| 11:43 | FIXED |  Missing definition |
| 11:41 | TEST |  Take CMXXXVII |
| 11:41 | FUCKYOU |  Problem voided and I'm very pissed about having to do this |
| 11:39 | MYSTERY |  It appears to go wrong sooner than I could possibly imagine |
| 11:37 | TECHNO |  My debug log should tell me more about what's happening now |
| 11:36 | TEST |  Take CMXXXVI |
| 11:35 | BUG |  The error I get makes no sense... at all! |
| 11:01 | TEST |  Take CMXXXV |
| 11:01 | FIXED |  404 (well the errors don't have numbers here, but you know what I mean) XD |
| 10:50 | TEST |  Take CMXXXIV |
| 10:50 | TECHNO |  What I am going to do now may make perfect sense to fellow coders, but if you ain't it won't. I'm deliberately gonna cause errors, but the error messages will tell me if some underlying setups are corect. Yeah, you follow? |
| 10:42 | FIXED |  A few data errors in this devlog after some saved data was lost due to a reasonless Blue Screen Of Death |
| - = 22 Oct 2023 = - |
| 23:41 | CONFIRMED |  All compilation error accounted for... (for now) |
| 23:40 | TEST |  Take CMXXXIII |
| 23:40 | FUCKYOU |  Not gonna mention this |
| 23:39 | TEST |  Take CMXXXII |
| 23:39 | FIXED |  Faulty definition |
| 23:38 | TEST |  Take CMXXXI |
| 23:38 | FIXED |  Syntax error |
| 23:37 | TEST |  Take CMXXX |
| 23:37 | DONE |  Ah yeah, just copy in all #use clauses, I don't care. It's clear I'm gonna need them all anyway |
| 23:36 | TEST |  Take CMXXIX |
| 23:36 | NOTE |  All I care about now in tests if it this all compiles |
| 23:32 | SCRIPT |  Not much I can do today, but a small template has been set up (as I could just copy and adept the template for the field flow. So far the easy part) |
| 9:58 | STATUS |  Today I have no time at all to work on the game. And especially since the combat engine is the most complicated part of the game which I do NOT want to mess up, I guess it's wiser to leave that for either tomorrow or perhaps tonight, but no sooner than that. |
| - = 21 Oct 2023 = - |
| 23:08 | STATUS |  FINALLY the error I wanted to see (namely the error indicating the combat flow scripts do not yet exist), but that is no more concern for today. I'm tired and I need rest now. See ya another time! |
| 23:06 | TEST |  Take CMXXVIII |
| 23:06 | FIXED |  Syntax error |
| 23:04 | TEST |  Take CMXXVII |
| 23:04 | FIXED |  Conflict |
| 23:02 | TEST |  Take CMXXVI |
| 23:02 | VOID |  Oh well |
| 23:00 | TEST |  Take CMXXV |
| 23:00 | FIXED |  Interfacing case error (luckily these interfacings are part of the core and once they work, they work) |
| 22:58 | VISUALSTUDIO |  Take CMXXIV |
| 22:58 | FUCKYOU |  Da fuck? |
| 22:56 | TEST |  Take CMXXIII |
| 22:56 | FUCKYOU |  GRRR! |
| 22:55 | TEST |  Take CMXXII |
| 22:55 | TRANSFER |  Big fonts |
| 22:50 | TEST |  Take CMXXI |
| 22:50 | FIXED |  Link up code typo |
| 22:48 | TEST |  Take CMXX |
| 22:48 | NOTE |  Now the game should give the combat intro screen, but it should surely crash as soon as the combat itself is to be loaded. For multiple reasons. For starters the combat code itself does not yet exist, duh! Yet another should should do it. |
| 22:43 | DONE |  All direct compilation errors accounted for |
| 22:41 | TEST |  Take CMXIX |
| 22:41 | FIXED |  Hmmm |
| 22:41 | TEST |  Take CMXVIII |
| 22:41 | FIXED |  Anything but the kitchen sink |
| 22:40 | TEST |  Take CMXVII |
| 22:04 | SCRIPT |  Effect for encounter (will be the same as TFT. Is less resource and memory consuming than what I originally had in mind. And perhaps better looking too). |
| 22:02 | TRANSFER |  Boss intro tune (although not yet needed, it would be safer to have it present already) |
| 21:40 | FIXED |  A bug in the encounter editor giving odd values |
| 21:36 | CONFIRMED |  Link to worlmap works! Good! |
| 21:34 | TEST |  Take CMXVI |
| 21:34 | FUCKYOU |  Fuck it! |
| 21:32 | TEST |  Take CMXV |
| 21:32 | FIXED |  Exit didn't work |
| 21:30 | SCRIPT |  Template for combat link up |
| 20:15 | STATUS |  FINALLY the system tries to call the not yet existent combat routine giving me the error message I wanted to see. |
| 20:13 | TEST |  Take CMXIV |
| 20:13 | VOID |  I think I must do things differently. I guess Lua had it's original "len" function removed. |
| 20:09 | TEST |  Take CMXIII |
| 20:09 | FIXED |  When interfeacing directly with Lua, case sensitivity is always an issue |
| 20:06 | CONFIRMED |  I see... what I wanted to see. Good! |
| 20:04 | TEST |  Take CMXII |
| 20:04 | VOID |  Let's have a look at this |
| 20:03 | FUCKYOU |  Line! Good one. Well, perhaps, I can do something better than a line |
| 20:01 | TEST |  Take CMXI |
| 20:01 | FIXED |  I think I fixed that |
| 19:57 | INVESTIGATION |  Just as I though.... MaxEnc contains value zero. Now the question is why |
| 19:55 | TEST |  Take CMX |
| 19:55 | DEBUG |  Let's find out why the guage bar doesn't appear |
| 19:02 | FIXED |  At least I've put the gauge a bit higher |
| 19:01 | VOID |  Static locals. I really must see to it that that feature gets to work properly, as this is driving me NUTS! |
| 19:01 | FAILURE |  Devlog tool crashed |
| 18:59 | CONFIRMED |  As I do see a crash caused by the combat starter (not quite the error I expected) I guess the encounters themselves will fire up. |
| 18:58 | INVESTIGATION |  I also need to sort out if that means if the encounters don't start at all |
| 18:57 | BUG |  Well the gauge appears... I do need to put it a bit higher on screen but that's only a minor issue. The major issue is why the gauge is empty |
| 18:55 | TEST |  Take CMIX |
| 18:55 | FIXED |  Did I fix the hotspot of the guage? |
| 18:55 | CONFIRMED |  Well, that void did the trick |
| 18:51 | TEST |  Take CMVIII |
| 18:51 | FIXED |  Tag? |
| 18:51 | HUH |  Something happens |
| 18:50 | TEST |  Take CMVII |
| 18:50 | FUCKYOU |  I don't know if this does anything, but I gotta try |
| 18:45 | FUCKYOU |  Take CMVI |
| 18:45 | FUCKYOU |  Fucking fuck fuck! |
| 18:43 | TEST |  Take CMV |
| 18:43 | DEBUG |  let's fuck up things even further! |
| 18:42 | INVESTIGATION |  How could that happen? |
| 18:42 | HUH |  In stead of the countless lines I should see in my debug log, none appear, which means that the routines in question are not called at all. |
| 18:41 | TEST |  Take CMIV |
| 18:41 | DEBUG |  Let's fuck in some more logs, shall we? |
| 18:38 | INVESTIGATION |  It's absolutely clear to me that the RENC routine is for some reason still thinking no encounters should take place.... But why? |
| 18:36 | HUH |  Nothing appears to happen. Or is that my imagination? |
| 18:34 | TEST |  Take CMIII |
| 18:34 | LINK |  And that should make the linking complete |
| 18:34 | LINK |  The update part has now also been linked |
| 18:34 | LINK |  The draw part has been linked now, but that's only part of the work |
| 18:32 | DUMMIED |  The debug code has been dummied, as I no longer need it. |
| 18:32 | SITE |  Added tag DUMMIED |
| 18:31 | NOTE |  I do see that the number of foes is properly loaded, but a few things are not in order.... But that was PURE STUPIDITY... I forgot to link the Renc routines to the main flow. |
| 18:29 | TEST |  Take CMII |
| 18:29 | FUCKYOU |  Got it now? |
| 18:28 | COCKROACH |  Still no fucking good? |
| 18:25 | TEST |  Take CMI |
| 18:25 | FIXED |  I think I found the little critter |
| 18:21 | TEST |  Take CM |
| 18:21 | DEBUG |  More work out here! |
| 18:16 | BUG |  Well, the GINIE does at least report 0 for the list of foes. No idea why, so that needs to be sorted out |
| 18:14 | TEST |  Take DCCCXCIX |
| 18:14 | DEBUG |  A little check I need to do in the GINIE link up code |
| 18:08 | INVESTIGATION |  Hmm, still no data in the log, which indicates that the foes are not being processed AT ALL. Now we got ourselves a serious case to solve, my dear Watson. |
| 18:06 | TEST |  Take DCCCXCVIII |
| 18:06 | DEBUG |  I've put in an extra line that should tell me how this could happen |
| 18:03 | INVESTIGATION |  Right, just as I thought! The enemies are not properly taken into account, leading to no enemies tied to the map, and thus the rest of the Renc does what it's supposed to do, declaring the entire area enemy-free. |
| 18:01 | TEST |  Take DCCCXCVII |
| 18:01 | DEBUG |  Another debug line that should give me more clarity |
| 17:58 | DEBUG | Layer FRENDORBUSHES has 35 encounters Layer __BASE has no encounters! That's at least the result I wanted to see, but that makes things only more mysterious. But wait there's more to check. |
| 17:56 | TEST |  Take DCCCXCVI |
| 17:56 | DEBUG |  Let's first figure out if the number of encounters is loaded properly |
| 17:54 | FIXED |  A little issue with this devlog |
| 17:49 | INVESTIGATION |  The question is of course WHY |
| 17:49 | HUH |  WTF! After all this work NOTHING HAPPENS AT ALL! |
| 17:45 | TEST |  Take DCCCXCV |
| 17:45 | FIXED |  Code typo |
| 17:32 | TEST |  Take DCCCXCIV |
| 17:32 | FIXED |  Property definition error |
| 17:29 | TEST |  Take DCCCXCIII |
| 17:29 | STATUS |  Time now to put to the test wether or not the random encounter routine will respond at all (yes, it will crash... at least when it works the way it's supposed to work it WILL crash!) |
| 17:18 | TODO | |
| 13:15 | SCYNDI |  All compilation errors accounted for (at last). Well I only expected this to take time. |
| 13:14 | SCYNDI |  Take DCCCXCII |
| 13:14 | VOID |  I really need to implement += and -= into Scyndi |
| 13:12 | SCYNDI |  Take DCCCXCI |
| 13:12 | FIXED |  Mistake in varname |
| 13:10 | SCYNDI |  Take DCCCXC |
| 13:10 | FIXED |  Code typo |
| 13:09 | SCYNDI |  Take DCCCLXXXIX |
| 13:09 | FIXED |  Haste makes waste |
| 13:06 | SCYNDI |  Take DCCCLXXXVIII |
| 13:06 | SCYNDI |  Well let's implement it then |
| 13:06 | HUH |  Did I really forget the SUFFIXED function in the Scyndi Core? |
| 13:03 | TEST |  Take DCCCLXXXVII |
| 13:03 | FIXED |  Keyword "and" is not existent in Scyndi (it was in NIL and Neil) |
| 13:02 | SCYNDI |  Take DCCCLXXXVI |
| 13:02 | NOTE |  But more testing for compilation crap is still required |
| 13:02 | TECHNO |  This will at least disable to compiler to generated any errors for non-existent combat linkups. When a combat starts, the game will now DEFINITELY crash due to non-existent content, but for now that's fine... actually even the desired behavior. |
| 13:00 | SCRIPT |  A (for now) empty script created for combat linkups |
| 12:59 | STATUS |  And now we come to the error I was waiting for. The missing of the Combat linkup code |
| 12:57 | SCYNDI |  Take DCCCLXXXV |
| 12:57 | SCYNDI |  I wonder how many more #use clauses I'll need that Neil never needed XD |
| 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 |