Now I did forget to recompile NALA, and a new feature that's added to TQMG (which NALA uses) is required to make this work... That can however not be reason for an "out of memory" kind of error, but hey, it needs to be done anyway before I can fix anything, so here goes....

That error is NEVER right, but it means that something else went wrong... And what it is.... A needle in a haystack

"Not enough memory"

And used it to display the points

A "DarkText" library created

Some images of Aziëlla are faulty tagged with alternate text "Scyndi".... This will not happen again, but existing occurances cannot be fixed at the present time, and even if I could... It's not worth my time to do so....

Configuration changes to this DevLog!

Although you may initally not see any changes

Well, at least that part works now

A sense of "self"

Illegal object reference

Case error

I see.... I must make sure that NALA crashes out when assets are not found

Comma parse errors that came in this way.

Re-written as small part of code and see what happens

I wonder!

Variable misreference

Code typo

Case error

Easy fix, but still.... It was expected

"tag" unknown... I KNEW it!

I don't know why I'm sure they won't work,though....

I've set up the base model for the combat bars....

Done

Before this can be applied a small study is needed... Nothing serious, as the font will work as normal, but I need to check out some things, first.

Well, font is a great word, but the font engine will use it to show the HP, AP and VITALITY (if appliccable)

All bugs accounted for... Or at least... so it seems....

That stupid thing I do not wish to talk about!

OPLETTEN!

Illegal call to NALA API

I think I fixed it... I just made the way tables are interally called in a crappy way that was bound to crash

As far as I can judge now, all stat parsing is now working the way it should.... This has not yet caused me the end of my total suffering, but at least progress was booked

Another test coming up!

I think that the MaxCopy value was null, and maybe that's what caused this error... I've put in an concstructor in the C# code that must make sure that if not set MaxCopy is just an empty string

So I've set a line that must tell me WHY it thinks that AP has a MaxCopy setting (which it hasn't).

There's bullshit going on about MaxCopy....

Another debug line required I'm afraid (I'm getting tired of this).

But what will happen next?

Case conflicts with the SuperGlobal.VarCommunicate property

I think I fixed the issue at hand, yet I cannot be sure, so another test is in order

Fix confirmed, however another bug yet remains, and that is a pretty odd one too

A calculation notation fault in the underlying C# API that caused SUM and AVG scripts to return 0 always unless the last element has a higher value

A no longer needed debug line

Another debug line to be sure....

Hmmm.... closing in... I think

Unfinished string

More debug lines added so I can (hopefully) pin this down a bit better

Let's test this out now

Furthermore, the max of AP is the same for all characters, however determined by level and chosen difficulty setting

AP is not a stat in Dyrt, but determined by level

Well That has also been fixed now!

GRRRR!

The cause of the cyclic error should now also be fixed

that works

Let's find out!

Does, or doesn't it work?

Recompile NALA

Alright, I've set up a proper error throwing system for this, so if this happens again, a proper error should be immediately thrown, so we cannot keep on stacking up until a stack overflow exception is thrown by .NET

Aha... A cyclic infinite loop eventually leading to a stack overflow.... I can fix that, but first let's build in some protection inside NALA to make sure this cannot happen again, and that a proper error message will be thrown.

Let's check this in VS itself, shall we?

Congratulations, NALA crashes completely!

But lets run another test to have THAT ONE confirmed...

Illegal function call and I do think that is also what causes #20 to happen

Let's see how the stat script looks.... (It's there were it appears to go wrong).

Case issues, but that is not what casues #20 to happen

Let's see who this could happen, as I don't understand it anymore... :-/

#19

#20

Another parse error

Forgotten "end"

I've recoded the SubCollecto generator, and this should fix the issue at hand!

A fix is a fix, but is it THE fix for #18.... Another test will tell me

Not sure if this fixed the issue at hand, but at least one reference to a static that is no static has been fixed

Not completely good, but the error has changed, so the fix took effect.... Pity another bug popped up now

Issued as #18

Which is now fixed

That is of course when you declare it at the wrong spot so it resets all the time

What shows is that the variable 'chklevel' is always 0, even though that should NOT be the case...

Another checkup

I've set in some more anti-whitespace-protection

And they replaced that for me... Good

Bringing my scooter to a repairman as the battery was as good as dead

But I'll have to look into that later....

Which despite all that still refuses to work

At least the corect data is now generated for the parser

I forgot to save.... (oh crap!)

All Lua can make out of this table is {1} WTF?????

Written a slower, yet more accurate routine, to read JCR lines.... Hopefully this will help to get things done properly

MOAR debug crap

Well, the debug data shows indeed that the lines could not be parsed... As a matter of fact, I've NEVER seen such bullshit in debuglines as I did see here today.
Further investigation is now required to find out how this bullcrap could come to be...

I really have to do this... A break down debug line

At least it's been confirmed that the level data itself has been read..... 9009 of code could not have been the outcome by that debugline otherwise...

Another line added to find out more, as I get the feeling something funny's going on here

It looks like the level stats are not even read at all... :-/

let's sort out this non-existent stat....

Another issue

I think I fixed the trouble now....

Parse errors

Moar info then?

Nope.....

As the bug was NOT fixed despite my hopes.... :-/

Let's hope this little enhancement will find me answers

And I think I fixed the bug that bugged me before I went to sleep

I've slept like a baby

Code still bugged, but I really must go to bed now... So that will be for tomorrow.

"nil"

I think I'm just getting tired :-/

MOAR stupidity

Bug in built in NALA line splitter

I fixed it now!

You shouldn't type "==" in C# when you mean "is not".... In C# (and all other C variants) that is "!="

That action != clever!

And let's see what it returns

Now I wanna know what that "script" is, so I added that to the traceback shit!

The errors get stranger and stranger

#16

Trust me, if you have coded with either Lua or NIL long enough you surely want to "end" it all (DRUM FILL!)

Forgotten "end"

MOAR fixing... I think....

I think I fixed the issue

And Let's see if that actually fixes stuff

Recompile NALA

A bug inside NALA

And I think I've prevented it from happening again... At least in this particular instance.

I think I fixed some code typos

RIght-o

Let's fire up a test then....

Will nothing happen? Then I must take it as "good"

If I got an error, then I guess something's not right....

Help me, Weniaria!

Please note that I can only trust on it that everything works the way it's intended if no crashes take place.... Other than that, I got nothing...

I've linked the character data for Young Irravonia to the New Game initiator...

I made changes based on these calculations, and tell ya what it DOES look better

I made a few recalculations hoping I can make things look esthetically better.

Ah, it was nothing

HMMMMM.....

I have forgotten my self

(that space is not a typo)

Code typo

Forgotten )

Case Error

Irravonia's face should now appear in the first box

A quick linkup routine to quickly access BoxPics

#15

Irravonia's portrait aliased for Young Irravonia... This is a dirty method and totally not friendly on the memory I know, but this was the easiest way to go and in the longer run also more efficient

And Kirana too, also a playable character despite her being an important enemy.

Shanda has been done too as she is also playable

Added tag SPOILER

And the other main heroes too

Brendor Main Potrait

Irravonia main portrait

Party member wrong tag

Time for a break!

Fixed tiling bug?

Nope that didn't do anything

Swapping some parameters, and see what happens :-/

Well that fixes a part of the problem

So that's fixed

Of coruse I should not forget that the back boxes need to be drawn as well...

Coordinate editting

In game screensize checker

it was stupid.. that suffices!

I don't wanna talk about it

A forgotten ) (sigh)

This debug line should answer a few things....

The width of each box should be fixed now....

there's a lot NOT right, but let's first get into a the coordinates of the boxes

Illegal function calls

I should be almost there now... At least when it comes to parse errors

Link errors

Code typo

I think I fixed the unclosed group

Well let's get back to business

I'll sort this out later, first to the store to make sure I've enough bread to get through the upcoming days.

A group scope not properly ended?

"." and ":" on arrays and function followup crashes fixed... well... sort of!

NIL didn't know the empower operator

And again

Not a fix, but a bug in NIL voided (for now).

Another test!

And now that everything is properly embedded the error (which was also stupidity, first class) became visible, and it has been fixed.

Running compile script

The embedding issue

Something went wrong in embedding the SysFont after I relinked that, I see

The latter I understand least of all

The game does start, but the status bar does not show, and trying to open the debug console even ends the game abruptly

And let's resume the test, shall we?

Running the compilation script

I did add a "lines" function to JCR6, but I shouldn't have forgotten I need to recompile NALA due to that! (sigh)

Another missing declaration

Not really a fix, but worked around a bug that is still being nasty in NIL

Missing declaration

MOAR of those

Code typo

Forgoteen "self"s

Not sure what I did to fix this, but mysteriously the computer has enough memory now (it was never clear why it hadn't in the first place)

More #say tags must provide an answer

?

I wonder if qff is to blame... But I also wonder why it was imported in the first place

The code crashes almost immediately after the first #say marker... This is really odd

One code typo

Placed in some #say markers so I can see what is happening

The error pops up while NIL is still translating the party sub

"Out of memory"?

Let's see what happens now

Regenerated Anyway.nil

Conflict prevention