(I did make sure NALA was recompiled (and a good thing as that didn't happen) so the faulty versions are now whoop! Gone!)

Time for a little test!

Since Feenalaria, the same woman you could meet in The Fairy Tale, is now Fairy Elder, it's only fitting I put her face into the textboxes in which she speaks.

Typos and stuff in the scenario at Irravonia's home

Ate my dinner

Still an error, but that error was to be expected as it links to part not yet being coded/scripted, so all-in-all... Me happy!

And that fixes the chain to Irravonia's house as a whole... Whoopdy-freakin'-doo!

An the evil was that NIL was a bit picky in its use request about .nil not being fully lowercase, which should frankly not matter at all. That has been taken care of!

An now the proper error message shows in stead of a "not enough memory" error, and the stack overflow has been eliminated... Am I cool or what :P

I found the source of these out of memory errors... I think

Now what I see is that there's an 'inproper library name' error thrown, and that this is causing a cycle of function calls leading to an infinite cyclic-function call, in the end leading to a stack overflow.... I need to sort out why this happens, but at least I got SOMETHING to rely on

Here goes:

Well this has been confirmed by the output thrown on the Visual Studio output console

Some analysis of the memory meter in Visual Studio shows the clear mark of a stack overflow.... That could explain the out of memory errors... Question is what causes those overflows in the first place....

A few extra SAY tags to find out what the hell is going on there....

I wonder.... It goes wrong when the new mapscript is about to be loaded.... :-/

Another try{..}catch thing

Apparently not!

I've put in a proper error catcher in the State Creation routine... Hopefully this will rid me of those fucking false "out of memory" errors, and make sure the system makes sense of a change!

When a standard mapscript does not exist, the system will load the MapScript "none.nil" in stead, which only covers the required callback functions, and nothing more than that.

Music also starts correctly

Chain to Irravonia's home

Well, now I fixed the portraits so the correct ones should appear

Right-o! The main issue's been fixed....

SIGH!

See I *knew* the fix was merely ignored, now that I had an error in my fix (causing the game to crash immediately, which was in this case a good thing), I could see that the modified file was not in the resource I had for testing.

To ensure the modified BoxPics.nil file will be loaded I've added a #say tag, so the console will always confirm things are in order!

On the next instance for Brendor the crash comes back... Now I'll leave that one bugged on purpose I can test why the code fix in being ignored

At leat I fixed Textbox Irravonia, but ....

sprintf in the BoxPic routine and no unwanted whitespaces added? I guess I'll never understand this...

I've set the BoxPic routine to show a plaque with a question mark on it when no BoxPic picture with the required name was found.

KA-BLAMMO! Just as I thought... Yet something that was not allowed to happen due to BoxPic pictures being optional, so a fix is definitely required

Written a proper error catcher in the LoadImage API control

Wait a minute... I think I've seen the little bugger

Since the debugger indicates the textbox itself is being generated, I need to narrow things down.... I'll now take a look if the textbox height is properly being calculated....

Study was a success, but please note that I still don't have a clue where the issue is caused, so I'm still pretty much in the dark here...

Alternate indexing in C#

Yup, the exception is there, claiming a key not pressent in a dictionary is present... In other words the dictionary gets a key request for an non-existent record... that is pretty serious....

Gonna run the game in Visual Studio and hopefully some exception will be noted in the debug output, helping me a bit further on the road

I've adepted by sync script to also copy the music files if they are changed, which is possible thanks to RoboCopy

I've re-read the "evil" text where it appears to go wrong, and I cannot find a real reason to error at all, and since the error is beyond vague, this is a very serious issue (if I cannot find why this happens the project is over, so I must find out what the hell is going on here).

Aha.... It's not even the map chainer were it goes wrong.... The opening story doesn't even finish.... but why....?

Let's try THIS:

Not even a CSAY message I put in comes through, and that really indicates something's wrong here

An extra marker placed in the SASKIA script

I see the mapname was incorrect, however given the debug codes I added this could not cause this to happen without notices

Which doesn't appear to be the case

Extra marker placed to make sure the request itself even done

At least I can see that no map is being loaded on the moment it's requested, which is already nice to know

Marker text for Kthura map placed... This can at least confirm the map itself has been loaded properly, so I can be sure the "evil" doesn't lie there....

That didn't produce any usable results

Since this is looking for a needle in a haystack I've set up an active check for finding Kthura maps within a JCR6 resource within NALA, and hopefully that gives me some errors that actually PROVIDE INFORMATION!

So this will take some time to find out why this happens

Well that doesn't appear to be the case

But let's first re-run the general building script to make sure no conflicts can be there

.NET error in user code -- The typical useless error message

Running

Music linked to the map

I did however make a little preparation to more easily link "Calm Indoors" by Widzy to Irravonia's house

No time yet to test, and since I'm tired now, and in need of rest, that will come later.

A script has been set up to show little Irravonia, while her parents are discussing Agon's disease

They are there on this backside, though

Not really a bug, but I see now that Tephondar's wings are missing on his front side...

Tephondar sprite

Yannee Sprite

Feenalaria Sprite

Set up a living room in Irravonia's house

Running my PowerShell Kthura Building script so Kthura is fully up-to-date now

Deleting objects works

A total remap was desirable

A remove feature has been added, but not yet tested

A quick map created for testing a feature I must add to the Kthura editor (which I forgot to implement, silly me)

Got rid of data I no longer need

I've designed the corridor of Irravonia's house. This corridor is important because this is where Irravonia is eavesdropping when the Fairy Elder (Feenalaria) tells them of Irravonia's condition

Imported textures needed to recreate Irravonia's house

Although it's certain that nearly all levels will be recreated with Kthura, the teddybear transfer tool can still help me to me sure all textures will be correctly in place

The other text all seems to be in order, or at least according to the editor it is

I need to investigate #23 once more

Xtra test, but it should be alright

And while I was at it I also fixed #23

I fixed an order issue in Scenario block C of the prologue block

Fix confirmed

Let's do it!

I think I fixed the language ignore issue, but I gotta test to be sure

I REFUSE to call this 'fixed' (nothing was broken in the first place), but at least we can move on now

Let's try to "force" things

Let's see where it goes "wrong"

Which is bullshit

Out of memory

Well at least I won't use string.format anymore for strings where spaces matter, that's for sure....

It DOES seem that string.format() is to blame for all that leading spaces mess

And how could code cause an out of memory, stop doing it after dummying it, and not resuming to do it aftr undummying it, without modifications?

Okay, that trim not wanting a boolean was at least an error that made sense

???

I wish I knew why that FALSE out of memory report comes from...

(This should not make any difference, but hey... who knows)....

Let's see..... :-/

Although it's still beyond me why it keeps adding spaces... EVEN WITH TRIM() FUNCTIONS!

The string issue is indeed fixed

Let's have another go, Butterfly!

So by doing that now, the not available strings have been fixed

String was not properly linked to the metatable...

As far as I can judge so far, NIL is not responsible for those unwanted leading spaces

Add some Trims

I guess I gotta investigate... for some reason!

And NONE of the strings defined, for some reason

Spaces added to the var names for some reason

AGAIN!

Taken care of!

Missed a few spots

Append is not + but .. in Lua and thus also in NIL

To work, or not to work... that's the question

I've added a VARS overview

Code revealing the state of superglobals, as this would only spook up future debugging, and it's no longer needed anyway as SuperGlobals have been confirmed to work

But at least I know for sure that the names are now properly set

THAT still didn't do what I wanted to see....

a test to confirm this

That could explain why this didn't work, but first

Aha... The name variables are not even defined....

Let's see why the names are not set....

The margin issue does solve the text out of screen issue

I've gone for a wider margin, and let's see what that does....

The width's too great which can cause text to be generated outisde the screenborders, becomeing impossible to see, let alone, read.

The language setting is ignored, so I always get English even when I asked for Dutch

Variables are NOT being substituted

the basics work, but....

Not sure if it works, so here goes Take A

I've tried to set up a chain to the next box of text, or going back to the original flow if needed....

IT WORKSSS!

Illegal Function Call

Again!

Syntax error

Let's see what happens

Put that on screen

Set out what to put on screen and what not

X coordinate text change

Illegal Function Call

Another test

Forgotten "self"

#22

Okay, I got hungry, so I ate some eggs for breakfast first....

I've set the header to appear... ONLY the header for now... The scenario text comes later... Since people love the text to come in letter by letter (for reasons beyond me), that will take some extra effort....

Well, let's investigate what I can do to make text appear in the textboxes.... I do have the feeling (or false hope, you name it) that I might have done all the hard work now....

And now I'm really gonna call it a day!

PERFECT! That will do!

First act of stupidity

THAT should do it!

the box itself looks better, but the picture does not

Dang.... An entire page full on one day.... I really did a lot today

I need to see how that looks though

Made sure the height is always 150 min.... This would come out better with the portraits anyway, which are by default height 96

The outcome appears in order based on the calculation, eh?

77 looks very strange to me.... How could that height come out, I wonder....

DEBUG: y:1003; h: 77; c:1041

Temp line to get some more data, as I think there's something fishy going on here!

No difference.... I guess that was to be expected, but still

Let's try this....

Not yet sure....

Take Five, let's see what happens, eh?

I've set the y coordinate of the picture to the center of the textbox too

For starters i made sure the color is set to "white", as that may save me a lot of trouble

of course I should have checked that earlier

Oh, I made a vertical auto-center, eh?

It.... works... but I'm NOT happy, meow!

Take Four!

Which didn't end up in an out of memory this time (not that there was any reason to, but when it did throw that error, that reason was non-existent, as well... I I wonder)

And forgotten end

Another forgotten self

Take Three!

Okay NOW I should have that!

I fixed something else.... Now it had to be fixed anyway, but it was not what I though I had fixed.... (Stooooopid)

Let's hope I won't surpass 40 this time :-/

Take Two!

Cow! Use "self"... dork!

Take One

Forgive me my skepsis given what happened today... :-P

Does BoxPic what it has to do here?

Copyright note error

I need a break... Ok?

Calling it... for now.....

Several more repositories updated

Running

It appears to be working... again

Connection failure with GitHub

Note to self about BoxPics.nil -- I forgot about it, but it WILL save me a lot of trouble, so why not make use of that little file :)

Copyright and license blocks

Updated

and requiring 47 takes to get it done, is WAAAAY to much....

And tomorrow I've little time at my hands (which is normal on Sunday for me), so I'm not sure I'll be able to get much done then....

Of course you realize, that due to the big amount of time that was now wasted, I can't get to putting the actual text in

FINALLY

At least I see what I wanted to see... for now...

Take XLVII

Missing field

Take XLVI

Case Error

Take XLV

I've made sure that an empty string is also not accepted by the alternate font loader

Test XLIV

(All this trouble all to get a texbox on screen... And then to think I chose this road to SAVE time)