I still didn't hide what should be hidden, this is because I need Zone Actions for that, but the zone actions require the walking routines to be fully operational before they can be properly tested, so that's why I'll focuss on walking around first!

And the blockmap I see is the blockmap I WANTED to see

This check is needed to make sure the blockmap will NOT be the factor that messes up walking around, you nsee

I've done a blockmap check

Running!

But first a little break!

And in the Yaqirpa (and only there) a kind of "follow-the-leader" system will be in place

The next step is to make walking around possible...

Problem solved, hahaha!

Not too much bother though... It only requires me to resort to a bit of dirty code, that's all....

Right.... How could I forget.... Neil only executes group constructors the first time a group is being accessed.....

...

Well?

Wait a minute!

I've added one more clause to definitely seal this....

The mapscript is not loaded.... The constructor in the MapScript group would have been triggered if it was....

An extra line the system cannot refuse (no, I'm not the Godfather), should give me an answer. If the system refuses that confirms only more that stuff is not loaded the way it should be.

The debug log shows everything SHOULD have worked.... Since it didn't more investigation is needed

And well?

One Debug line added I now do need

Debug line I no longer need

And now to find the answer the the question "Why!"

It appears (so the logs imply) the mapscript is not loaded when loading the map from a savegame file....

....

Let's see why the transpad isn't removed!

I need a break, now!

That is a later concern, though

Except for the transporter pad that should not be there during the prologue

Most what I see is what I want to see

Let's see then....

That should cover that issue

Not the most elegant method, but hey, what works, that works!

This was indeed the reason

Although I do have a feeling I know what's happening here!

Briggs being level 20 (which is a stat) shows the stats apparently load properly... No errors due to divisions by zero seem to indicate the point maxima are loaded properly, but the amount of points the characters have is another story... Or so it seems....

And another test is in order!

Unwanted fallthrough

Seems to be so!

But does that work, now?

Well that's been fixed now of course!

Pointer declared and defined as NULL *INSIDE* the loop...... Fuck you, Jeroen!

This is BEYOND stupidity!

Now this pointer should never possible be NULL, so let's see what's happening here.

The question is now how this can happen!

Just as I thought... Not the string was the issue, but the pointer!

Well?

Let's see.....

Well?

What does this do?

Well let's see....

Fixed?

I see that the directory stripper is bugged..... It puts in the last slash, and it shouldn't be there

Moar debug stuff

I've dummed the Kthura object killer debug output.... At the present time I don't need that anymore!

So far everything in the logs looks like the way it should look

Well, all I can do now is test

I've set up a quick macro which should get up some more clarity on this one!

Now the question is WHY this happens

The only thing that could cause this is when maxima are all set to 0.... And that means the party data is not properly loaded!

I see.....

Huh?

Let's see

Speed is the reason I chose the way I did

Party Sub auto-load

Well, if this works, we can really get into the game itself

And now the setup has been done to enter the FIELD flow once loading a savegame is succesful

Well?

DORK!

Well, let's see!

Voided that issue!

With this all set, I can make sure the game links through to the Field Flow

Please note that during the prologue, only Wendicka can lead the group, but after the prologue all playable characters can be put in the lead

The setup to automatically get the player sprite to have the correct texture based on who is the leader

A way to check stuff

I did realize that the leader was not yet set... This is a bit of an error on my side, but not one that is impossible to fix, or to void, without having to create new savegames files... Of course, I have not yet a file that can fully login to Game Jolt...

The effect I was hoping for as what I get now

Take V

Protection can sometimes work against you, and being a bit unsafer can sometimes be the better call, I suppose

DAMN!

Take IV

BARF = Build Apollo Resource File
What else did you expect?

Let's run BARF again

Fix ignored!

Take III

Barf script

Illegal API function call

Take II

Invalid method called

Take I

Wish me luck!

Some things are not yet saved though, like achievements and stuff... I simply can't do all at once, you know

The most notable effect will now be that "LOADING" appears on screen, and that the music changes to the background music of the Yaqirpa. This does not yet do much in sight otherwise, but if all that happens without crashing, it's only a matter of linking things up further

I've linked that now the the save game loader

That is the code for that has been written now... There's still a lot more to do to make it actually work!

The map stuff should now also be taken in order

Stuff done!

I'll first get into some tests in Dyrt.... So be back later!

I *THINK* I fixed #210 for real now, but I'm not sure

And more work next time!

Running!

Well at least the GameVars appear to be working now

Taek XVIII

Oh, not a property, just an undentified metatable not linked

And after my break in general

After dinner

I really must get the bug fixed about static properties, so that will be my next thing to get into....

Take XVII

Missing State

Now THAT seems more like it!

Take XVI

ANd I think I fixed Neil while I was resting from entering my house

Okay, real life stuff got a bit on things

I'll investigate this later though... I need a break, and I need some food...

I do not know if the GameVars script is executed or not, but it looks like it's not.... None of the Game Vars have their specific value, so that reeks like a "no"

Take XV

So I dummied the code causing that for now, so i can focus on that first

Before all that I need to do some additional tests

Ah, the error that should come....

Take XIV

Right that was also not even what happened.... It was just a mispelled idientifier in Neil itself, and since Lua doesn't need declarations (hence while I needed Neil), unexplainable bugs happen (the sucky thing about Lua)

The debug log even confirms this impossibility as the string is just returned as is should be

Take XIII

Extra debug lines must provide me some answers

Checked the code, and given the way C++ works and the code around it this cannot be possible

Nil returned by API

Take XII

That should do it!

FINALLY A SCRIPT BUG!

Take XI

Truly fixed now then?

The compiler hates me for this fix

Take X

Fixed that!

DOH!

Ignored?

Take IX

I need to force a few things or the loader will give me one hell of a problem!

And now it works.... Odd!

Take VIII

So let's see inside Visual Studio

Time to find out why

Fix ignored!

Take VII

I had to recompile for that!

Fixed that idiocy

Forgot to configure the path in the C++ code for the savegames, and as a result it cannot find them!

IDIOT!

Take VI

Recompile

FIxed Missing link

Ah right.... missing link

Okay.... I had to dummy a variable that got undeclared in the process

Take V

I'll turn the try blocks out, but this will produce crashes on corrupted data (unfortinately)

Take IV

Bad C-string-format expression

Take III

WTF is happening here?

Take II

Adding a few debug lines

The error presented is useless to me, so this is gonna be hard!

Ah, try and catch were not accepted yet as keywords, eh?

Take II

Neil proved its worth here!

Dupe identifier

Here goes!

In the way the game is now, the Game Vars are loaded, and it will look like nothing more happens (unless there's an error, as the system should then complain about a corrupted savegame). Game Vars can however be checked in my debug console, so I guess that is the best way to go now then.

That is a bit of THE oil behind the game scripting, so very verrrry important

The script should at least be able now to parse and read the GameVars

A few enhancements to the Neil core library

Fantastico! It all compiles

Well that should cover the C++ part of the story, I guess

The mechanism which loads the general data and the party has been set... The general data needs to be parsed and processed by the scripts though, so there's still a way to go!

The plan for today is to get some stuff done for loading savegames. And also to test some stuff in Dyrt... The latter will be limited to testing though... Although I must say that one of the New Game+ dungeons is also soon to be added there....

That worked

Let's test what I got anyway... Take I

At least the game will buzz if you try to load an empty slot.... Actually loading a game is a bit more delicate, and will require me more time to do just right

Basis done for loading

The buzzer will be used if you try to load an empty slot

Loaded by the MAIN script

Transferred the buzzer

Take IV

Taken care of that

No party bar in load screen... remember, there is no party yet!

Take III (I think)

Wrong class

File reference incorrect

Only way to find out is by testing this, of course.... Take I

It's important to note that loading itself doesn't work... the game will crash as soon as you try, however it should at least be possible to select a game to load

Linked this to the menu

Function to go to the Load Game Screen

The load game will always use this screen, yes, also when you use the computers in Nizozemska (which are merely XTs) to save before.... Should be obvious, but never hurts to make sure to tell, right?

Activator to set script to "Load", which is (of course) needed to make loading games possible

stuff is not yet done, and I don't even expect things to be done today... I hope I can get a few more things done on Star Story or otherwise testing in Dyrt, but it might be better if I skip that until the maintenance programs are done

I could still get some Dyrt stuff done, but it took longer than normal

System maintenance done

I see now... My dark mode addon's settings were reset and influenced FireFox where it shouldn't have

That still doesn't answer the question why Firefox fails to show my devlog in accordance to the css file and Chrome does, though

Ah, my internet appears to be running normally again.... (At last)

At least google chrome does show the devlog page the way it should