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 Either this file is incomplete or the engine that wrote this file is outdated!"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

?

It is strongly implied by the log that the routine which has to read out integers is not called at all!

Take VIII

Some extra debug lines must tell me wtf is going on as never in the world should a "nil" have been possible here!

WTF?

Take VII

Code typo producing a syntax error

Take VI

Syntax error

Take V

Busted #use clause

Take IV

Syntax error

TakE III then

But it compiles now

More issues here

Take II

I think I fixed that....

A very old code error.... never produced errors before, and now suddenly it does.... Odd!

Well, let's see what this does, eh?

My debug console now supports it!

This is not much, but this will allow me already to set up a few debug test routines

SaveGames should now at least store the party data, and the gamevars

Since this is a JCR6 file, remember that I can easily check things with tools such as Stach

Given the current timing I don't think I can get the savegames fully done today, but that doesn't matter too much.... If I can at least get a few things on the road, then it's fine!

Some needed real-life stuff

And what I said is only the core stuff.... The actual data is compiled by the game scripts, and the scripts doing that have not yet been written!

Technically saving should now work... that is if the game asks for that... Data cannot yet be loaded though, so we still gotta work on that.

For good reasons!

So I had to do this with my Go app, and I hate that!

This is Microsoft for ya

Looking for a good directory creation routine - Failure C: The follow up-method though is not compatible with my C++ version, which is with the old deemed deprecated even stranger

Looking for a good directory creation routine - Failure B: Second one is official, but deemed "deprecated"

Looking for a good directory creation routine - Failure A: One method is Windows only, and does therefore not fit in my Cross-Platform approach of matters!

SaveGame Manager link up code

Like In all RPG games I created before (well, the most of them at least, as my very first will likely be different), the savegame routine will be using JCR6 to bundle all data, but still keep it seperate.

Here I'm going to create the save game routine, now it should be noted that I'll work out the save game routine in several portions....

First of all, to make sure the Nizozemska version of the savegame routine remains 100% compatible with the regular way, and also to make sure all everything can work on its own.

In order to make this all possible, I will need to write hugh poritons of C++ code, however that should not be too much of an issue... I think.... ;)

Like I said before today is entirely reserved for Dyrt...

And will be it for today

Due to some complexities I was facting today I could not get things done on the monster encyclopedia for Dyrt, and since that is something that requires my full attention, I'll focus on that tomorrow first, and now that the savegame routine is the next step in Star Story which is equally as much of a hassle I cannot really combine the two projects until both are done I'm afraid.

Since Dyrt is more pressing at this moment I'll get on the encyclopedia first, and depending on how things in Dyrt are after that, I'll get back to Star Story to get the savegame routine done, but if Dyrt is in a state I need to put emphasis on that game first, Star Story will have to wait.

But when I think about it, it'll be alright.

Well, let's see!

And THAT should do it!

Kill

Kill

Let's see how things come out!

Inventivity is required for the job, y'know!

However, they would cause needless confusion if they were used now, so that's why I chose to remove them from the prologue, but not from the later instance

This also allowed me to make the mapscript "shamelessly" remove the transporter pad... They were only present in the map for later access by the Hawk

So does the JCR6 entry overview output

The builder output seems to spell good news

One of the reasons I loathed the "existing" engine as it was next to impossible to implement it there... Well that an many other reasons.

The first dungeon is a bit different in code than the others, and I wanted to prevent conflict, just duping the map wasn't the right way to go, so naturally aliases had to do, and fortunately JCR6 supports that

this is something I've done with the original game as well, so that's nothing new. I only did it immediately now where I did it later for the original.

I've aliased the Yaqirpa for later access with the Hawk

#11

The base stuff for the Yaqirpa

In a basic sense the save game routine can be set up from here, however that won't be for today, and I need to set up a few things first

So far all is cool

And does it work now.... please?

#use clause

Yawn