15 replies to this topic

[huhn_m]

Sorry for the long time without updates.
Here is the new changelog. The release will be done by wendesday I think since I want to include one more feature maybe:

0.3 (Current Version)
Added For-Loops (buggy, see comment below)
Added String Pointers
Added PIXL command
Added PIXT command (without function since it must be tested on calc)
Added LINE command
Added RECT command
Added CSTR command
Renamed PAUS to SKEY (as originally intended for making the language compatible to new keyboard system to be released in 0.4)
Changed welcome screen
Added Modulo Operator (
Added Inverse Mode to pixel oriented commands
Inital pointer support for PIXT command and string conversions
Changed String format to support variables and "added" strings (ccst)
Added String "adding" (concat)
Added check for open things on compile end (e.g. functions not terminated with FEND)
Added String segmentation.
Added String-Tree Model
Added new command LDS command to make interpreter / bytecode more versatile.
Added configuration through config.h. You can manually edit it or use the supplied editor (that hopefully makes it until release!)
Added initial mouse support
Added support for buttons (same problem as rect!)
Extended example to include ALL features the system currently supports.

Fixed improper termination because of halt command not leading to end of program but to intermediate termination.
Fixed Bug in debugApplication procedure, now terminates correctly after call.
Fixed ErrorMSG handler not terminating properly. It now terminates with the debug message and then restores interrupts and returns to dos.
Fixed Serious Bug where a-b was interpreted as b-a

Known Bugs:

*On Exit the program leaves an non-empty keyboard buffer and calls another program because of this! (not verified, use emReboot in config.h)
*FOR Loops have no entry check yet. If the start value is > the end value then the condition will never be fulfilled but the function is called once though.
*Rect does not resort coordinates (x1 must be < x2 etc!)

quite a long log compared to the previous versions ...

[Deimos]

Wow, that looks impressive I think i should even be able to write a game using this version of MLC

[SRM]

Yes, it looks like Deimos is right !! It's a great job !!

You're saying you added inverse mode for pixel oriented commands, so are the seven colours operationnal ?

I don't remember, is there a step value for the FOR loop ? And can this step be negative ?

Two interesting functions to see in 0.4 are input and random. How do you intend to make a random function ?

Try to solve known bugs before releasing.


Testers, prepare to have some work from now !!

[huhn_m]

Random: I don't know ... I'll have to read through various ideas.
For LOOPS: Negative steps are not yet allowed since NO negative numbers are allowed now at all. This will be done later.

Input: Yes. The keyboard system will be next. Do you want a string based input as well?
Function parameters will not be in 0.3 since they require some change about how the compiler stores
function information. Will be in 0.4 and then also 100% working FOR loops

Pixel Commands: No, I'm sorry. No colors yet. However, there are white, black and inverse.
As for the bug with #RECT and the button command. I'll look into this tomorrow since it can be quite annoying.

I think you will especially be impressed with the included demo app since it is capable of this MLC1 wasn't.
Though the function parameters are desperately needed for all of the show features to work more efficient and "more correct"

[Deimos]

Well, we somehow managed to do some cool stuff in MLC1 without function parameters However i agree that they will be very useful and will make things much easier.
As for input, i think that string based input isn't that important and can wait. I currently can't see any use for it except inputing highscores.

[SRM]

About the input command, as it's a basic stuff, I think it's important to have this in the earlier versions.

Negative numbers are absolutely needed. I don't understand how we could make any games without it.

[huhn_m]

I'm sorry. I deleted them temporarily. However, I'll up them again in about one hour. Sorry for the inconvinience.

But you could also wait till about tomorrow since I hope to have the beta3 ready be then and this has some bug fixes and is much more versatile (and has big parts of the graphics engine implemented)

[huhn_m]

Updated! (not version 3 only the old versions are up again.)

Sorry for the inconvenience (sp? strange word )

[samy]

thank you. And now I'm waiting for tomorrow....

[samy]

hey the beta 3 is out, it seems very good, but I don't know how to install it on calc!!!
If I understand what I've been reading: we must do the program on computer and then put it in the afx?
... but I don't manage it!

otherwise I have one suggestion: in the future (or maybe it's already possible) I think it would be a good thing to be able of going from a program to another as it is in basic with the command 'Prog', because it would allow to make great rpg for example with different part (moves on a map, fight, talk....). It is just a suggestion

[huhn_m]

for the installation wait some day or so. I'll make a wiki page about it.

As for PRGM I won't do it since it would imply a new compile pass during runtime and would be quite memory costly.

however, there will be a include functionality in later versions so that you can include other functions but this is a VERRY minor goal but should also help in managing big programs.

Thanks for you interrest in the program. Maybe I'll even include an "installer" in the next version. Who knows ...

[samy]

ok thanks a lot for all!
But instead of waitng I'll will try to install it on calc by my own: I've just discovered that the os.exe was the one to install on calc I don't know what is the next step.. but I'll find it by myslf...; or by your documentation! lol

[huhn_m]

better install osu.exe

It is the UPX compressed version of the program.

[huhn_m]

sorry. I found some new major bugs (one of them beeing caused by a test program SRM sent me, so thank you!) that need fixing before I'll release Beta3R2 since I want this release to be as bug free as possible.
One of the mistakes is a buffer underrun 8) don't know where this comes from I'll try to do more debugging this night and tomorrow but I had quite a busy week at work so no guarantees on a release date.

Also the bug that showed when clicking the abort button about 10 times is not yet resolved but I expect this to be directly related to the problem with the virtual stack.

Thanks for your patience

[SRM]

Prior to quality !! I like the way you work.