NonstickAtom785 c384674a13 Update readme.md | 4 vuotta sitten | |
---|---|---|
.. | ||
tutorials | 4 vuotta sitten | |
data-types.md | 4 vuotta sitten | |
drawing.md | 4 vuotta sitten | |
fonts.md | 4 vuotta sitten | |
grammer2.5.inc | 4 vuotta sitten | |
math.md | 4 vuotta sitten | |
modules.md | 4 vuotta sitten | |
readme.md | 4 vuotta sitten |
Author........Zeda Thomas
[email protected]
Project.......Grammer
Version.......2.51.0.0 (I will probably forget to change this :( )
Last Update...15 February 2020
Language......English
Programming...Assembly
Size..........2-Page app
Hey! You! This app is still in development! If you have issues, please send me an email, post in the forums, or, post an issue on GitHub! Thanks, you're the best :)
To follow the progress of Grammer, check out Omnimaga, Cemetech or GitHub, where development is most active.
If you have questions or suggestions, feel free to email me or post in the forums!
Grammer is a powerful programming language for the TI-83+/84+/SE calculators. Unlike TI-BASIC, it is not designed to do math and as such, Grammer math is fairly limited in some respects. This also means, however, that it uses a math system with its own tricks and optimizations. If you are going to learn how to effectively use Grammer, you will first need the Grammer interpreter on your calculator (this document assumes you have the latest version of the Grammer App). After that, you should become familiar with Grammer's:
First, send Grammer 2 to your calculator. If you have this document, I assume you have the App. Next, run the app on your calc. Grammer is now installed until your next RAM clear. The app's menu controls are as follows:
Gram
filter. This shows only files with a Grammer header.AppV
filter. Toggled on, this will only show appvars, off will show progams and appvars.Asm
filter. Toggled on, this will only show assembly programs.If you want to make a program, you have to remember two very important things:
.0:
Now, before I explain all the technical stuff, could you to create this program and run it?
:.0:Return
:ClrDraw
:Text(0,0,"YUM
:DispGraph
:Stop
It should look like this:
To understand pointers, you have to understand how memory works. First, every byte of memory has what is called an address. An address, is a pointer to that byte. For example, the first byte of memory is at address 0, the second byte is at address 1, et cetera. On these calcs, there are 65536 bytes of memory addressed at a time. The last 32768 bytes are RAM. This is where your program and everything in it is stored. This has some powerful implications. If you have a pointer to a section of code and you tell Grammer to start executing there, it can jump there immediately. If you have a pointer to a string, you can use that pointer to draw the string or use it. This means you don't have to create any external variables for your strings or sprites. If you want to create an appvar for save data, having the pointer to that lets you edit the data, save to it, or read from it. Pointers are powerful. As such, you will probably be using them frequently. It is not coincidence that Grammer vars hold 16 bits and pointers are 16-bit.
Vars are two byte values. These can hold a 16-bit number, so these are well suited to holding pointers. These are all the letters A
to Z
and θ
and A'
to Z'
and θ'
. For readability, you can use the lowercase letters instead of A'
, for example.
How do you store to these? Like all BASIC programmers know, use →
. For example: 3→A
Likewise, for a string: "Hello World→A
Don't be fooled, that does not store the string anywhere. It just stores where the string is located in A.
So if you change the byte at A, you will change the "H" in your program. If you want to try it, run this program and then check the source again.
:.0:
:"HELLO WORLD→A
:WriteB A,47
:Stop
You will see that after running the program, it now says "QuartReg ELLO WORLD"
! That is because we changed the first byte of the string to the value 47
, which corresponds to the token QuartReg
. Be careful! Not all tokens are one byte-- lowercase letters are a notorious example of two-byte tokens.
Pointers are also useful with labels or finding programs. Anything that requires searching, actually. For example, to goto a label, you would do something like Goto Lbl "HI
and that would jump to the label named .HI
. You can also get a pointer to this label. If you need to jump to that label often, this saves lots of time by not having to search for the label every time. Remember, everything is math in Grammer:
:.0:Return
:Lbl "HI→A ;Locates the label .HI and stores its pointer to A.
:Repeat getKey(15 ;`getKey(15` returns 1 if clear is being pressed
:call A ;This calls the subroutine pointed to by A
:End
:Stop
:.HI
:B+1→B
:Text '0,0,B ;Displays the number B at (0,0)
:DispBuf
:End ;End the subroutine
Remember to always End
your subroutines! Now I want to take time to finally start explaining some code. Labels can be a bit more descriptive than in BASIC. You can use up to 42 bytes for a name, technically, but try to maintain readability. You can also use pretty much whatever tokens you want in a label name. For example, I had a label named ICircle(
that I had draw inverted filled circles. Lbl
takes a string argument where the string is the name of a label. It returns the pointer to the line after the label.
Goto
and call
let you redirect your program. Goto
jumps to the code and prgm
will execute the code and return once it gets to End
. Both of these take a pointer to figure out where to go.
Repeat
works like it does in TI-BASIC. It first executes the code between Repeat
...End
and then it tests the condition. If the result is 0, it repeats the procedure. If it is anything else, it stops looping and continues after the End
.
getKey
gives you two ways to read keys. Key codes are different in Grammer from TI-BASIC. The first way is to use it like you would in TI-BASIC. For example, getKey→A
stores the current keypress in A. You can also use something like getKey(15
to quickly test a specific key. This is great if you want to use multiple key presses. I shamelessly stole this wonderful idea from Axe, I think. Or maybe somebody suggested it to me? I can't remember.
See the Math documentation.
See the Drawing documentation and the Drawing Commands.
You can use DrDnar's Monochrome Font Editor to generate fonts! It supports many kinds of font formats, including Grammer fonts.
If you want to create a font from scratch, you can check the technical documentation here.
Here are a bunch of commands that do not fit in the drawing or math category. This is a whole lot of info (about 9 pages), but I made some examples later on so that you can figure out how to use these better :)
Example | Token | Grammer | Description / Example |
---|---|---|---|
a→b | → | → | Stores a value to a var. Ex. Return→A' will store the value output from Return to A' . |
a→ib | →i | →i | Stores a Grammer variable to an OS Real variable. For example, 3→iA will store to the OS Real var A . |
a→Str1 | →Str* | →Str* | Stores a Grammer string to an OS string variable. For example, "Hello→Str2 will store to Str2, and "Hello→Str20 will store to the hacked string 20. Keep in mind that the Str0 token corresponds to string 10. |
a→bc | → | → | Stores a 32-bit value in two variables. →AB will store Ɵ' in A and Ans in B. So to store a 32-bit result from multiplication: A*D→AB . |
// | // | // | This is used to start a comment. The comment goes to the end of the line. A commented line is skipped. As a note, the user can include a comment after code. |
" | " | " | This starts a string. The output is a pointer to the string that can be used to reference it later. |
π | π | π | If you put a π symbol before a number, the number is read as hexadecimal. For example, π3F would be read as 63 . If what follows is a non-hexadecimal character, this returns a pointer to the float value of pi. |
! | ! | ! | This has several uses. The first is to work like the not() token in TI-BASIC. So for example, 3=4 would return 0 because it is not true. However, !3=4 would return 1 . Likewise, !3=3 would return 0 . The other use is with loops and If . For example, If A=3 will test if A is 3 and if it is, it executes the code. However, !If A=3 will execute the code if A is not 3 . See If , If-Then, If-Then-Else, While, Repeat, and Pause If. |
i | i | i | This is the imaginary i. Use this to access OS real vars. For example, to read OS var A and store it to Grammer var A : iA→A . To store a Grammer var to an OS var: B'→iA . This does not read decimals or imaginary parts of a number. |
E | E | E | This indicates the start of a binary string. This is the exponential E . |
Name | Description |
---|---|
If <expr> |
If the expression following is not 0 , the line following it will be executed. The line is skipped if it is 0 . Conditions are computed to the next newline. For example, you can have → and : in an If statement in Grammer. |
If-Then |
This is similar to If except if the statement results in 0 , any code between and including Then and End will be skipped. This works like the TI-BASIC command. |
If-Then-Else |
This is similar to If-Then , except if the condition is non-zero, only the code between Then and Else is executed, otherwise the code between Else and End is executed. |
!If <expr> |
Like If , but executes if the condition is false. |
!If-Then |
Like If-Then , but inverts the condition. |
!If-Then-Else |
Like If-Then-Else , but inverts the condition. |
For For( |
The arguments for this are For(Var,Start,End . Var is the name of a var, Start is the starting value to load to the var. End is the max value to load to the var. Executes the loop, incrementing the variable each time until Var=End |
For <<expr>> |
The arguments for this are For(#loops . #loops is how many times the For loop will... loop. This is useful if you need it to loop without actually affecting a variable. |
Pause If |
This will pause so long as the condition is true for example, to pause until a key is pressed, Pause If !getKey . |
!Pause If |
This will pause while the condition is false. So to pause until enter is pressed, do !Pause If 9=getKey . |
While <condition> |
Executes what is between While...End so long as the condition is non-zero. |
!While <condition> |
Executes what is between !While...End so long as the condition is zero. |
Repeat <condition> |
Executes what is between Repeat...End and continues to do so as long as the condition is zero. |
!Repeat <condition> |
Executes what is between Repeat...End and continues to do so as long as the condition is non-zero. |
Local x,y,z ▶Nom( |
This starts a block in which a defined list of variables is preserved. For example, in Local A,B,C: <<code>> :End , no matter what <<code>> does to A,B, and C, they will be restored when the End is parsed. |
:If A=B ;Since A=B is false, the following line is skipped
:9→A
Or also:
:If 3=B→A:*14:-14 ;This is the full statement
:Text 0,0,"Yay!
:If 3=4
:Then
:3→A
:9→B
:16→C
:End
:If 3=4
:Then
:3→A
:9→B
:Else
:16→C ;This is the code that gets executed
:End
:Repeat getKey=15
:End
!Repeat
checks if the statement is false in order to end. For example,
to remain in the loop while Enter is being pressed:
:!Repeat getkey=9
:End
:For R,0,48
:Circle(32,48,R,1
:DispBuf
:End
:Stop
:0→A→B
:While getKey≠15
:A+1→A
:B-1→B
:End ;This tells the While loop to End / restart!
An example with Local
:ClrDraw
:Local A,B
:0→B
:For A,0,9
:B+A→B
:Text '0,0,A
:Text '6,0,B
:DispBuf
:Pause 33
:End
:End
Name | Description | |
---|---|---|
nextPntr Return |
This returns a pointer to the next line of code. | |
Goto |
This is unlike the BASIC Goto command. This jumps to a pointer as opposed to a label. |
|
Lbl |
This returns the pointer of a label. The argument is a pointer to the label name. For example, Lbl "HI will search for .HI in the program code. Also, you can specify which variable the label is in. For example, if you wanted to jump to a label in another program, you can add a second argument as the name of the var. For example, to find the label HI in prgmBYE: Lbl "HI","BYE Note: If your label moves, you'll need to manually update the pointer! For example, when creating a variable, or resizing one, or archiving/unarchiving. I recommend doing all of that first, then finding labels. If your label moves, then you might accidentally execute garbage data. |
|
Pause xx |
This will pause for approximately xx/100 seconds. So Pause 66 will pause for about .66 seconds. |
|
Pause |
This will wait for [Enter] to be pressed and released. | |
call prgm |
This is used to execute a sub routine. | |
call ( prgm |
This executes a subroutine, and sets ? var to to point to the arguments. For example, prgm(Z,1,2,3 will call the routine that Z points to, and sets ? to point to the 1 . |
|
] |
This takes a pvar as an argument. It parses the code at the pointer as an argument, then updates the pointer to point to the next argument. This is useful for parsing arguments passed to a subroutine. Take the example above, if the subroutine that Z points to has the code ]? , then the 1 will be parsed. The next ]? will parse the 2 , and so on. |
|
setUpInt Func |
The arguments are setUpInt Pointer[,Counter . This will automatically execute the subroutine pointed to by <> based on Counter. Counter is based on an internal counter, not based on actual timings like seconds or milliseconds. The default is 128. See the examples below.
|
|
Asm Asm( |
This can be used to run an assembly program. | |
AsmHex AsmPrgm |
This allows you to input an asm code in hex. (C9 is needed) | |
jmp ln( |
This will let you jump forward or backward a given number of lines. | |
L | The list L. Arguments are L line#,[start,[size,[EOL This let's you execute a specific line number. By default, it starts the line count within the main program, but you can pass an optional start value, an optional size value (default is 32768 bytes long), and an optional End-Of-Line argument (default is 63, the newline token). |
|
StackOp Param |
? points to parameters, this stores those parameter values to variables. For example, if ? points to 1,2,3,4 , then StackOP A,B,C,D will store 1 to A , 2 to B , 3 to C , and 4 to D . This updates ? . This is useful for subroutines that take parameters! See the example below. |
|
StackOp ' Param |
This pushes values to the parameter stack. For example, Param'A,0,1,B+2 pushes the value of A, 0, 1, and B+2 to the stack. |
|
StackOp ° Param |
This pops values off the parameter stack into a var. For example, using the previous push sequence, Param°A,B,C,D would store the original B +2 to A , 1 to B , 0 to C , and the original A to D . |
|
StackS Pmt_Bgn |
This token is located at [Apps][Finance][Up]. This is a var that holds the base location for the parameter stack. Changing this value automatically resets the parameter stack pointer. | |
StackE Pmt_End |
This token is located at [Apps][Finance][Up][Up]. This is a var that holds the end location for the parameter stack. | |
Local x,y,z ▶Nom( |
This starts a block in which a defined list of variables is preserved. For example, in Local A,B,C: <<code>> :End , no matter what <<code>> does to A,B, and C, they will be restored when the End is parsed. |
Name | Description |
---|---|
getKey |
This returns a value from 0 to 56 that is the current key press. You can use this chart for values. |
getKey( |
getKey( will allow you to see if a key is being pressed. For example, getKey(9 will return 1 if enter is pressed |
Input |
This allows you to input a string. The pointer to the string is returned. (this is not a permanent location, the data will be overwritten the next time Input is used). To get a value input from the user, you can use expr( : expr(Input →A . This will store the result to A. Input can also take an optional string input. The input string will be displayed after what the user is typing. If you execute this code, I think it'll explain it better. It's honestly pretty cool for a calculator. See below for information on the Input vars! |
Menu( |
Syntax is, Menu(y,x,w,"Title","Item0","Item1","Item2","Exit . It basically makes a pop-up style menu, returning the number of the selected item. Returns 0 if exited due to [CLEAR] or [ON] . *Note: you can append an optional last argument that starts with ' to specify a default option. For example, if the last argument is '3 then the third option will be selected by default.* *NOTE: You can't use font mode 0 for this; instead the menu routine will convert to font mode 2.* |
Menu(' |
Draws a menu that queries Grammer subroutines for the content to render. Syntax is Menu('"Title",y,x,height,width,GET_ELEMENT_ptr,SELECT_ELEMENT_ptr . The subroutine for GET_ELEMENT will receive the index in Ans. Return 0 if out-of-bounds, else return a pointer to the string to draw. The subroutine for SELECT_ELEMENT will receive the index in Ans, and the header number in Ɵ' . Modify this as you want, the result will be returned as the result of the menu. Returns 0 if exited due to [CLEAR] or [ON] . *NOTE: You can't use font mode 0 for this; instead the menu routine will convert to font mode 2.* |
Menu('' |
This works like Menu(' , but when you put ' in front of the header argument, Menu(' will instead interpret it as a pointer to a routine. The routine will take a value in Ans, the current header, and that routine should return the appropriate string. Return 0 if out-of-bounds. |
Ans |
This will return the value of the previous line. |
eval expr( |
This will compute a string as a line of code (useful with Input ). See below for more info on eval ! |
inStr inString |
This is similar to the TI-BASIC command. This will return the location of a sub-string. The inputs are where to start searching and the string to search for: inStr SearchStart,SearchString[,maxlength] . The size of the input string is returned in Ɵ' and if there was no match found, 0 is returned. |
Search length( |
This will return the size of a variable (in RAM or Archive) as well as the pointer to the data in Ɵ' . For example, to get the size of the appvar named Data : Search "UData→A . If the var is not found, -1 is returned. |
Search ' length( |
This is used to search for a line. For example, if you want to find a specific line number in a program, this is what you would use. The syntax: Search 'StartSearch,Size,LineNumber,[LineByte , StartSearch is where to begin the search Size is how many bytes to search in. 0 will search all RAM. LineNumber is the line number you are looking for. LineByte is an optional argument for what byte is considered a new line. The output is the location of the string and Ɵ' has the size of the string. If the line is not found, the last line is returned instead. |
Here is an example with the basic menu:
Menu(1,1,16,"Title","ITEM 1","ITEM 2","ITEM 3→M
And here is an example using callbacks:
:Lbl "GET→A
:Lbl "SEL→B
:Menu('"HEAD",2,33,59,30,A,B→M
:Text '0,0,M
:Stop
:
:.GET
:→X<26
:If !
:End
:"ITEM A→Z
:WriteB Z+5,X+65
:Z
:End
:
:.SEL
:+1
:End
Or a more complicated menu that has three headers (use left/right):
:Lbl "GET→A
:Lbl "SEL→B
:Lbl "HEAD→C
:Menu(''C,2,28,59,40,A,B→M
:Text('0,0,M
:Stop
:
:
:.HEAD
:→X<3
:If !
:End
:If X=0
:" STRING>
:If X=1
:"<PIC>
:If X=2
:"<GDB
:End
:
:
:.GET
:→X<10
:If !
:End
:Ɵ'
:If =0
:"Str
:If =1
:"Pic0
:If =2
:"GDB0
:→Z
:WriteB +3,X+48
:Z
:End
:
:.SEL
:+1
:End
There are two Input variables that you can store to:
x→Input This sets the address of the Input buffer.
x→Input' This sets the size of the Input buffer. Remember, 1 byte is used to mark the end of the string!
expr(
can be used to evaluate a line of Grammer code, usually from a string. For example:
eval "2+3→X
And this will set X
to 5.
That's pretty boring, but it is handy if you want to parse user input:
eval Input →X
*NOTE: This will evaluate up to a newline (Input
appends a newline to the user's input).
This means a user could enter any valid Grammer code, even if it messes with your program's variables!*
This note also means you might have some unexpected behavior. For example, let's try this:
"Rect X,Y,15,15,2→Z
eval Z
DispBuf
That inverts a 15-by-15 rectangle at (x,y), as you might have expected.
Now let's try to invert that twice:
"Rect X,Y,15,15,2→Z
eval Z
DispBuf
eval Z
DispBuf
An error is thrown! Why is this? Well, this will teach something about the internals of the
Grammer parser. The "
token basically returns a pointer to the next token. So the first line
causes Z
to point to the Rect
token:
"Rect X,Y,15,15,2→Z
^
Z literally points here
Since eval
parses up to a newline, and we give it the pointer Z
, eval
executes:
Rect X,Y,15,15,2→Z
instead of Rect X,Y,15,15,2
as you might expect. So this means
Z
is modified by the first eval
, so the second eval
is trying to read some random
piece of memory as Grammer code. Ouch!
So the simple fix is to put the →Z
on the next line:
"Rect X,Y,15,15,2
→Z
eval Z
DispBuf
eval Z
DispBuf
And now it inverts and then reinverts. Ta-da!
Here is a useful trick with eval
. Suppose we want to invert a sprite on two buffers (grayscale?)
display it, then re-invert it (this is useful for showing a sprite, without destroying the graphics
buffer).
We could do something like:
Sprite 2,S,Y,X
Sprite 2,S,Y,X,1,8,B
DispBuf
Sprite 2,S,Y,X
Sprite 2,S,Y,X,1,8,B
But by combining the Return
command with expr(
, we can do this:
nextPntr→Z
Sprite 2,S,Y,X:Pt-Off(2,S,Y,X,1,8,B
DispBuf
eval Z
.0:nextPntr
ClrDraw
Text °"(x,y)=( ;ClrDraw sets the cursor to (0,0), so I can use °
eval Input ",)→X ;I get the next input here. The string is ,)
Text ,+1 ;This increments the X coordinate.
eval Input ")→Y ;This gets the Y value.
Pxl-On(Y,X ;Or whatever you want to do with the coordinates.
DispBuf
Stop
inStr
example.
:Lbl "DATA→A
:inStr A,"How→B
:.DATA
:HELLOHowdyWoRlD!
This is a command subset. Commands start as misc #,
. The TI-OS Token equivalent is solve(
.
# |
Name | Description |
---|---|---|
0 | CopyVar | misc 0,"VarName1","VarName2"[,size[,offset . This will copy the program named by VarName1 from RAM or archive to a new program named by VarName2. If Varname2 already exists, it will be overwritten. So for example, to copy Str6 to Str7: misc 0,"DStr6","DStr7 . This returns the pointer to the new var and the size of the var is in Ɵ'. The last arguments are optional. Size lets you choose how many bytes are copied (instead of just copying the whole var). You can also add an offset argument to tell where to start reading from. |
1 | CopyDataI | misc 1,loc i ,loc f ,size . This copies data from loc i to loc f . (Forward direction) |
2 | CopyDataD | misc 2,loc i ,loc f ,size . This copies data from loc i to loc f . (Backward direction) |
3 | ErrorHandle | misc 3,Pointer . This will allow your program to have a custom error handler. Pointer is 0 by default (meaning Grammer will handle it). Otherwise, set it to another value and Grammer will redirect the program to that location. The error code is returned in Ans. Ans and Ɵ' are put back to normal when the error handler completes. Errors: 0=ON , 1=Memory |
4 | CallError | misc 4,Error# . This will execute the error code of a Grammer error. For example, to make a Memory error: solve(4,1 . Using Error 2, you can input a string for a custom error: misc 4,2,"Uh-Oh! |
5 | PortWrite( | misc 5,port#,value . This writes to a port. Ports give information about peripherals. You can find ports documentation at WikiTi |
6 | PortRead( | misc 6,port# . Reads a byte from the port. |
7 | CopyVars | misc 7,addr . Copies the pointer vars to some other location. Currently this requires 108 bytes of space in the new buffer. Good for backing up vars. |
8 | OverwriteVar | misc 8,addr . Overwrites pointer variables with new data. |
Here is an error handler example
:misc 3,Lbl "ERR
:<<code>>
:.ERR
:If =1 ;Means there was a memory error
:Stop
:End
Name | Description |
---|---|
clrPart R►Pr( |
This will clear the particle buffer. |
runPart R►PΘ( |
This will recalculate the particle positions and draw them. If you want to change the particle buffer, just add a pointer argument. FindVar "EBUF→A:runPart A-2 |
addPart P►Rx( |
This will add a particle to the buffer. Just use the pixel coordinate position. For example: addPart 2,2 |
partType P▶Ry( |
This will change the particle effect. 0 is normal sand, 1 is boiling, 2 lets you put in a basic custom rule set. If you want it to check Down, then Left/Right, then Up, use the following pattern: 0000 1000 0110 0001 2. That makes it first check down, and if it cannot go down, it then checks left or right, and if it cannot go left or right, it tests up. In decimal, that is 2145, so you would do: partType 2,2145 . To make things easier, though, you can just use a string. This will achieve the same effect: partType 2,"D,LR,U . Note that you do need the actual string, not a pointer. |
addPart ' P►Rx(' |
This will convert a rectangular region of the screen to particles. The inputs are addPart 'Y,X,Height,Width . This scans the area for pixels that are turned on and adds them to the current particle buffer. |
Name | Description |
---|---|
$ ▶DMS ▶DMS |
Found in the angle menu, this is the "module" token. Modules allow you to extend Grammer's functionality. Grammer comes with a default module which must be included to use some functions (like the Menu command). Currently, you can have up to five other modules. For example, if you have a module packaged as an appvar called MyModule : "5MyModule→▶DMS . In order to execute a function MyFunc( from one of the modules, use : ▶DMSMyFunc . If you have the token hook enabled (from Grammer's main menu), it looks a little cleaner: "5MyModule→$ and $MyFunc , respectively. |
Play conj( |
Warning: I have no knowledge of musical jargon, so excuse my mistakes. This is a sound command with three inputs. The syntax is Play Note,Octave,Duration . See below for notes. Octave is 0 to 6. Duration is in 64th notes. So for example, a 32nd dot note uses 3/64th time. Duration is thus 3. |
conj(' conj( |
This sound routine has two different functions Play 'Duration,'Period or Play 'Duration,DataLoc,Size . This reads data for the period directly to save time (intead of converting numbers on the fly). Size is the size of the data in words, not bytes. |
| # | Note1 | Note 2 |:------ |:------| | 0 | C | C | 1 | C# | Db | 2 | D | D | 3 | D# | Eb | 4 | E | E | 5 | F | F | 6 | F# | Gb | 7 | G | G | 8 | G# | Ab | 9 | A | A | 10 | A# | Bb | 11 | B | B
Grammer doesn't really have any data structures which is both good and bad. Bad because it makes you have to think a little more about how to approach a problem, but good in that it allows you to create precisely what you need. This is where you will need commands to create variables, insert or remove data, and edit the data. I will also try to explain how to create some basic data structures like arrays and matrices. First, here are the commands you have to work with:
Name | Description |
---|---|
FindVar Get( |
This uses a string for the name of an OS var and returns a pointer to its data. If the variable does not exist, this returns 0. If it is archived, the value returned will be less than 32768. Ɵ' contains the flash page the variable is on, if it is archived, otherwise Ɵ' is 0. As an example, FindVar "ESPRITES→A' would return a pointer to the data of prgmSPRITES in A' . |
( |
Use this to read a byte of data from RAM |
{ |
Use this to read a two byte value from RAM (little endian) |
WriteB int( |
Use this to write a byte of data to RAM. |
WriteW iPart |
Use this to write a word of data to RAM, little-endian (a word is 2 bytes). For example, to set the first two bytes to 0 in prgmHI: FindVar "EHI→A:WriteW A,0 |
MakeVar Send( |
Use this to create Appvars or programs of any size and filled with zeros (so long as there is enough memory). For example, to create prgmHI with 768 bytes: MakeVar 768,"EHI . Programs must be prefixed with "E" , protected programs "F" and appvars "U" . Lowercase letters are allowed! :) |
[ |
Store a sequence of bytes to a given location. For example, A[1,2,3,4 will store 4 bytes at A. You can also store some elements as two-byte words by using the ° token. A[1,2,3°,4 |
[[ |
Stores a sequence of 2-byte words. A[[1,2,3,4 |
[( |
Stores hexadecimal input as raw data. A[(3C7EFFFFFFFF7E3C |
GetInc IS>( |
This is used to read memory. The argument is one of the pointer vars. It reads the byte pointed to by the pvar and then the pvar is incremented (so consecutive uses will read consecutive bytes). |
GetDec DS<( |
This is used to read memory. The argument is one of the pointer vars. It reads the byte pointed to by the pvar and then the pvar is decremented (see note on GetInc ) |
Arch Archive |
Follow this with a var name to archive the var. For example, to archive prgmPROG , do this: Arch "EPROG . |
UnArch UnArchive |
Use this like Arch , except this unarchives the var |
DelVar |
Use this like Arch , except this will delete a var |
del sub( |
Use this to remove data from a variable. the syntax is: del #ofBytes,Offset,"Varname . For example, to delete the first 4 bytes of program Alpha: del 4,0,"EAlpha . |
ins augment | This is used to insert data into a var. The syntax is: ins #ofbytes,Offset,"VarName . For example, to insert 4 bytes at the beginning of appvar Hello : ins 4,0,"UHello . |
Display the first 4 bytes of prgmPROG using GetInc
:FindVar "EPROG→Z
:Text '0,0,GetInc Z,16
:Text '°GetInc Z,16
:Text '°GetInc Z,16
:Text '°GetInc Z,16
Now let's make an array! First you need to know what you want. Do you want to have 2-byte pieces of data or 1-byte? I like using one byte, so here is what we do:
:.0:Return
:MakeVar 256,"VDat→Z ;We create a TempProg with 256 bytes of data called Dat.
:Z[rand,rand,rand ;write 3 random bytes.
:ClrDraw
:For 3
:Text '°GetInc Z ;Display the value at byte Z. Also increments Z.
:Text '°",
:DispBuf
:End
:Stop
That didn't really need a 256-byte variable, but I figured I would show how to
make one. Anyways, what that did was make a 256-byte tempprog (which the OS
automatically deletes once control is returned to the OS and you are on the
homescreen). Then, we stored 3 random values to the first three bytes, then we
displayed those values with commas after each number. If you want to use that 256
bytes for a matrix, instead, you can make it a 16x16 matrix and access elements
using a formula. For example, to read (Y,X): (Z+X+Y*16
. That means that the data is stored in rows. That is why we take the row number and
multiply by 16 (that is the number of elements per row). This happens to be the
syntax that tilemaps are stored (stored in rows).
Name | Description |
---|---|
Mode Text( Fix |
Use this to set the typewriter delay. The larger the number, the slower the typewriter text is displayed. |
Mode Fix |
See description below. |
Full |
This is used to set 15MHz mode. Alternatively, if you add a number to the end Full0 sets 6MHz, Full1 sets 15MHz, Full2 toggles the speed. 15MHz is only set if it is possible for the calc. This returns 0 if the previous speed setting was 6MHz, 1 if it was 15MHz. |
setFont Output( |
See description below. |
Use this to set certain modes. For all the modes that you want to set, add
the corresponding values together. For example, to enable inverse text
and inverse pixels, use Mode 1+2
or simply Mode 3
Here are the modes:
2-Inverse pixels. Now, on pixels mean white and off means black. In assembly terms, it reads from the buffer, inverts the data and sends it to the LCD.
4-Disable ON key. This will allow ON to be detected as a key, too
8-Hexadecimal Mode. (Numbers are read as hexadecimal)
16-PixelTestOOB. Returns 1 for out of bounds pixel tests
32 - Display numbers as signed values.
If you want to use bit logic to set or obtain specific bits or info about the current modes, you can do things like this:
Stores the current mode value:
:Mode →A
Set the first three modes without changing the rest:
:Mode or 7
Toggles mode 4 (enable/disable [ON] key)
:Mode xor 4
This is used to change the font. The syntax is:
setFont 0
will change to the default 4x6 font.setFont 1
will change to the variable width font.setFont 2
will allow you to use the 4x6 font at pixel coordinatessetFont 3,font
will let use Omnicalc or BatLib styled fonts.setFont 4
will use the OS small font.setFont 5
will use the OS large font.setFont °#
will change the draw logic (except for the OS fonts).
The output is a pointer to the fontset (custom or standard set), which could be useful if you wanted to read or process the built-in set.
For setFont 3,
remember that Omnicalc's font data starts at an offset of
BOLD
, you would do:
setFont 3,11+FindVar "EBOLD
Using the output of Full
, we can test if a calculator allows 15MHz.
:Full ;Sets 15MHz if possible
:If Full ;Sets 15MHz if possible, but also returns 1 if the previous `Full` was able to set 15MHz
:Then
:Text 0,0,"15MHz possible!
:Else
:Text 0,0,"6MHz only :(
:End
You can use this as a guide to the key values output by getKey example, Clear=15
Also, there are the diagonal directions:
6=Down+Right
7=Up+Left
8=Up+Right
16=All directions mashed
Here is code that changes X and Y based on key presses.
:.0:
:0→X→Y
:Repeat getKey(15
:runPart
:If getKey(9
:addPart Y,X
:X+getKey(3
:min -getKey(2,95→X
:Y+getKey(1
:min -getKey(4,63→Y
:End
:Stop
I have to give special thanks to Yeongjin Nam for their work on writing a better tutorial for Grammer and as well Louis Becquey (persalteas) for their work on writing a french readme/tutorial. Both of them have also made many valuable suggestions that have helped make Grammer what it is right now. Thanks much! Thanks to @NonstickAtom785 for the suggestions and (many) bug reports that I would not have found otherwise!
I also thank Hans Burch for reconstructing Grammer 2 after I lost my work. It must have been a tremendous amount of effort and tedium, and I greatly appreciate it. They've continued to provide valuable feedback about bugs and it has been extremely helpful.
Finally, I would like to thank the sites that have let me get the word out about this project, especially Omnimaga and Cemetech.