One more debug line should now solve it all?

Well, the logs still give me data that make this outcome impossible, so I guess I need to search on

Take V

I can't see the trouble, but let's move on...

So where does it go wrong then?????

I see the chopping also goes correct

Let's see now!

MOAR debug lines

The debug stuff shows, that the computer did detect the header... Now why didn't it load?

Take IV

Let's see if the claim is correct

And even after a restart it keeps persistent on that, so this is gonna be a challenge

Take III

It did seem, the computer didn't detect the savegame file exists now, and that's something I need to sort out

Save game file itself appears to be saved properly

Didn't have to click... Just overing activated the savegame routine

Take II

It's always so irritating parsers do easily give the wrong line when the error is either at the start or the end of a line

Syntax error

Take I

Saving should (assuming I didn't make any mistakes) now be possible

And now the compiler DOES agree! :)

Ah, function name typo

However the compiler doesn't agree with me

That should link it

Neil part done

Turned that into API functions

Quick Stretch Draw routines

Save Icons

Let's see

For now this is only cosmetic as nothing happens when you click, but at least I can see if stuff works this way

Hover effect

I see now... existing entry or non-existent entry was messed up

Doesn't look too good

The header does appear though

No empty slots show

Invalid tag

Illegal Function call

Take IV

Macro error

Take III

"Empty Slot"

Take II

Asset load parameter error

Let's see how that looks

I've set up the Screenbox

The audio will be important later, but the font, that's a different story, that is important now!

And I linked the big font to the loader

Linked the audio to the main loader

A 35 pixel version of the Robotica font.... I needed that anyway, and this was the proper moment to come up with it, I guess

Puzzle solved music box jingle

Level up - Music Box jingle (yes, that will remain the same)

Take V

Missing var initiation

Take IV

PAth error

Take III

File name typo

Take II

Syntax error

(although all the savegame UI will show now is a black screen, the character bar and the mouse pointer).

To make my life easier, I've set the debug command to easily get into the save screen

And from there the moment has arrived that I can up the UI (at last).

I've made a header readout script for the savegame manager (the UI flow that is)

And that automatically links stuff together

Added them to the API list

I did make some functions that can both check the existence of a savegame, as well as limit itself to only reading out its header

I have therefore decided to make a capped system, where only 1000 savegames in total can be made and no more. Now I didn't even create that much in one folder during my produciton of earlier titles, so I don't expect anything like that to happen during the production of this game either...

The fact that C++ does NOT have a default library to load file lists, except for one which is Windows only, does complicate things a bit. I am therefore a bit reliant on a bit of dirty ways to get things done

And not only that, it also contains what I wanted to see, which is even better!

Header is saved.... Oh goodie!

Sigh!

Forgot to call the function compiling the header

And that should do it, for now

Header compiler

Quick study about some things in the map data

Now there's only one more thing to do, before I can create the entire savegame menu up.

NEXT!

MOAR CRAP!

NEXT!

I think I fixed a 'count' issue

Let's go!

The basic lists should be created now

#14

List creation should take place now

#13 can be closed now!

Well?

And now the core thing behind #13 should be done (at last!)

Which are the same as in C and C++

Replaced old BlitzMax escapes with their respective C# variants

And due to that I can finally get onto working out #13

AT LAST THE ERROR I WAS HOPING TO SEE!

Next!

Reference error

I hate you!

And again!

Removed a file that only exists due to an earlier typo

Name file directory error

Well, how many tests did this take?

Missing closure, and thus this didn't get added

That cannot be

C++ doesn't save the lists at all????

Fixes ignored

Another test is in order!

I think

"Character 'Briggs' misses the data entry 'LISTS'. Either this file is incomplete or the engine that wrote this file is outdated!"

Another test is in order

Should be fixed now

Fixing

#13 I see.... Directory issues in my C++ code... No C# code's fault

No errors, but it does get no furhter than loading that party members set up I see

Hopefully that's been fixed

As expected though, the links entry DOES indeed only contain the closure tag

I see what causes Stach to suffer, and also to cause the RPG char viewer to malfunction

Stach crashes upon analysing the "Links"

Let's use Stach to see what is happening

It stays with no effect at all, though

(errors are expected as a new tag, yet unknown to the C# code is present)

Right, a few new savegames were created, now to see for #13 if the RPG reader picks them up

Again

Hopefully "end" will no longer be confused with end (string!=keyword)

I do need to create my savegames anew in order to see if stuff really works now

Links will be saved now!

Technically the same goes for the link detector, but as C++ is more delicate than C# I hope things go well here (in this stage of the game's development it will go no further than creating an empty link file, maybe 1 byte indicating the closure tag, but still)

That was just copying from C# into C++ and doing a few small adaptions

Function that saves individual link

This complicates things as I didn't want to go for the links until i could check their full workings, but I'm afraid I got no other choice now

Ah, BINGO! The RPGChar routine also checks the links....

No odd things pop in the debug logs, but perhaps I need to sort things out more into detail then

A few config fixes, but those are only cosmetic fixes for my own interest

And when it comes to #13 , things turn out worse than I thought, as the char viewer ignores all party data inside the savegame files of Star Story, which is, since they use the same encoding as my earlier games, quite odd... I need to find out why!

Init block set of Map's data as otherwise Neil will throw an error

So far so good, but I see one important thing

And Let's see how things go!

Undummied Wind Saving

Not everything is done now, though!

Compile

Oh, stupid! There was a better method indeed, how could I forget?

Glue code

That was the C++ part

API added

Wind can be set and be retrieved from KthuraActor*

But now is the time to call it a day!

#13 It seems that RPGCharView doesn't even detect the party data, while it should, so this is something to sort out!

Well?

Forgotten link

Moar stuff in savegames

So far so good.... Not there yet.... nooooo nooo, not nearly there, but the start is made!

Issued as NeilProject/Neil#21

Voided that for now, but I must make a not of that in Neil

I must make sure that end is ONLY recognized as a key word when not a string

It's probably more as I didn't count a few

Take XVII

One more fix?

It's good to see tha map file and the layer are stored in the Game Vars... That saves me a LOT of trouble!

"end" in gv missing

The fix was not in C++ but in the game script.... moron!

Let's see then!

The GameVars script was not correctly generated

Moar debug!

Nothing should have made it possible for that NULL value to end up in that map spot, yet it happened.... The map is an protected variable, so it can only be defined from inside the class

I already suspected that, but how this could happen is now the qustion

For some reason, the data inside a Data field is NULL

Not a clue what is going on

Take XVII

I've avoided direct vector creation calls, to make sure the foreach does not have to face a recreation all the time, but is sure to use the same data sequence all the time

Take XVI

Next stop fighting the issues with the save game manager itself

The Illegal Directory that was produced in the process

Finally I see the directory data I wanted to see

Take XV

I think I fixed all this crap now

Not there yet, as for some reason Windows does not give its environment data to C++, but at least the replacer appears to work now and that's half the battle

XIV

I hope I fixed the TReplace routine

Take XIII

???

Clearly NONE of the Dirry values appear set?

What is clear though, is that stuff is not operating the way it should....

At least the infinite loop is broken

CRASH!

Take XII

I think I tackled that one!

I think I fixed the infinite loop, as the replacer didn't like empty strings for substitution... not to mention there was more wrong there... However that does not explain why was an empty string in the first place!

even more interestingly, the setting appear to be where the crap freezes

Interestingly, the results Dirry produces are not what I wanted to see...

Take XI

This line should provide some answers on that

I wonder if Dirry is to blame... The routine has not yet seen that much action after all

Does not get any furhter than receiving the request.... But where does it go wrong then?

Take X

Extra lines added

No console output generated... I do fear I will have to analyse this in Visual Studio

But now the system is freezing, and I wonder why

fix confirmed

Take IX