Author Topic: Axe Parser (Read 553474 times)

Quigibo · « **Reply #1755 on:** May 09, 2011, 11:36:35 pm »

I'm planning a major feature addition soon which I am dubbing "Axsembly" (Please excuse my lack of creativity with portmanteaus

).

Basically, it will allow direct assembly commands to be integrated directly into Axe... but while keeping the syntax looking almost exactly like regular Axe. I will be using the lowercase letters for the registers and will allow a wide range of arithmetic instructions. The major difference with Normal Axe syntax is that assignments are done with the equal sign instead of the store arrow (like in most programming languages) and only a limited range of instructions are allowed. Both of these are to make Parsing on my side possible. Everything else is basically the same as you're used to already. There will also be a simple interface to transfer an Axe variable into a register for manipulation and then bring it back up.

For instance, the instruction "a=a+a" or "a=a*2" will both translate to the assembly instruction "add a,a". However, "a=a*3" does not correspond to a single instruction, and therefore must be made by hand using 3 instructions like "b=a : a=a*2 : a=a+b". There will also be some new operators like ++ which acts as an "inc" when used in a unary operation or an "add with carry" in a binary operation. Different kinds of rotation will also be supported. All these commands can be interwoven with current syntax, but I wouldn't rely on abusing the side effects because they can change from version to version.

This does have one conflict with Axe's current syntax however: The binary number prefix which is currently "b" will have to be changed to something else, but I think that command is used infrequently enough to not anger anyone who would be affected by this change.

I am hoping this will be a great introduction to assembly for new programmers who want to play around with it while staying in Axe's feature-rich environment. This was a feature I was panning to add at the beginning, but then gave up, but just now thought of a way I could do it efficiently.

Michael_Lee · « **Reply #1756 on:** May 09, 2011, 11:42:11 pm »

So basically, is this equivalent to inserting hex into the program, except easier to read?

Runer112 · « **Reply #1757 on:** May 09, 2011, 11:44:56 pm »

This sounds pretty awesome, although I am a little worried that this will make programming in Axe less "authentic," if you know what I mean. Although I do see the merits as well. But is it even necessary to change the b prefix for binary to something else? When parsing, if the b is followed by a 1 or 0, treat it as an indicator of binary data. Otherwise, treat it as the b register.

Freyaday · « **Reply #1758 on:** May 09, 2011, 11:45:12 pm »

Wait--Axe now has side effects? O.o

Quigibo · « **Reply #1759 on:** May 09, 2011, 11:53:39 pm »

Quote from: Runer112 on May 09, 2011, 11:44:56 pm

This sounds pretty awesome, although I am a little worried that this will make programming in Axe less "authentic," if you know what I mean. Although I do see the merits as well. But is it even necessary to change the b prefix for binary to something else? When parsing, if the b is followed by a 1 or 0, treat it as an indicator of binary data. Otherwise, treat it as the b register.

That's true, I guess I could do that.

Axe itself doesn't have side effects, but at the assembly level, the registers change around every time Axe does a native command. Sometimes you can use these values for optimization. Like if you knew the b register was always 0 after a Pt-On() for instance, you could use this knowledge to save some bytes. However, I am simply warning against this because I could change the built-in routines at any time and an optimization like that might not work anymore.

Hot_Dog · « **Reply #1760 on:** May 10, 2011, 11:53:30 am »

Will Axe create ASM programs even if the programs turn out to exceed 8 KBs?

Builderboy · « **Reply #1761 on:** May 10, 2011, 03:38:05 pm »

It will

ztrumpet · « **Reply #1762 on:** May 10, 2011, 04:37:21 pm »

Axsembly sounds pretty cool. Good luck getting it to run smoothly.

DJ Omnimaga · « **Reply #1763 on:** May 11, 2011, 02:43:26 pm »

That sounds nice, but then for the contest somebody will have to help me judging entries, because I haven't coded Axe in almost one year and I'll need to scan every entry to make sure only a certain percentage of Asm() and your new ASM feature is used, and sadly I might not make the difference between Axsembly stuff and regular Axe code.

ikemike · « **Reply #1764 on:** May 18, 2011, 08:13:54 pm »

Can somebody take a look at this code and see why everything up to "rand ^999→Q" repeats itself?

Code: [Select]

.KBYCRACK Crackity Hack!

prgmDEFINE
DiagnosticOff
sub(RCL)
sub(SSP,3,35,Pic1)
sub(SSP,25,30,Pic2)
sub(SSP,47,30,Pic3)
sub(SSP,69,35,Pic4)
sub(SSP,3,48,Pic5)
sub(SSP,25,43,Pic5)
sub(SSP,47,43,Pic8)
sub(SSP,69,48,Pic8)
sub(SET)
Repeat getKey(54)
DispGraph
End
sub(W)
0→T→P
Repeat T=4
40→X
1→A
Repeat getKey(54) or (X=40 and (A=2))
If A=1
Line(2,X/2+2,4,X/2+2)
ElseIf A=2
ref(2,X/2+2,3,1)
rref(2,X/2+2,3,1)
End
DispGraph
If X=0:2→A:End
If A=1:X-1→X:End
If A=2:X+1→X:End
End
If getKey(54)
sub(W)
40-X→X
X*X*X/371+P→P
Else
sub(W)
End
T+1→T
sub(SET)
End
20→X
Repeat getKey(54) or (X=100)
sub(RCL)
sub(FSP,5,30,Pic63)
sub(FSP,5,X/2,Pic6)
X+1→X
DispGraph
End
20-abs(60-X)→X
If X<0:0→X:End
X*X*31/40+P→P
sub(W)
conj(Pic01,L₃,768)
conj(Pic02,L₆,768)
sub(Q)
ClrDraw
sub(Q)
ClrDraw
sub(W)
rand ^999→Q
rand ^999→R
rand ^999→S
If P>999:999→P:End
P→Z
If Q>Z
Q→Z
ElseIf R>Z
R→Z
ElseIf S>Z
S→Z
End
0→X→Y
Fix 5
Repeat Z=X
If P≥X:sub(TSP,4,40,Pic7+Y):End
If Q≥X:sub(TSP,26,40,Pic7+Y):End
If R≥X:sub(TSP,48,40,Pic7+Y):End
If S≥X:sub(TSP,70,40,Pic7+Y):End
If Y=0
32→Y
Else
0→Y
End
Vertical -
Vertical -
Vertical -
ref(0,48,96,10)
rref(0,48,96,10)
If X≤P:Text(4,55,XDec):End
If X≤Q:Text(26,55,XDec):End
If X≤R:Text(48,55,XDec):End
If X≤S:Text(70,55,XDec):End
If X>P:Text(4,55,PDec):End
If X>Q:Text(26,55,QDec):End
If X>R:Text(48,55,RDec):End
If X>S:Text(70,55,SDec):End
If X<(Z-25)
X+10→X
ElseIf X<(Z-10)
X+1→X
Pause 100
Else
X+1→X
Pause 200
End
DispGraph
End
DispGraph
Return


Lbl Q
0→A
Repeat A=8
DispGraph
Horizontal +
Horizontal +
Horizontal +
Horizontal +
Vertical +
Vertical +
Vertical +
Vertical +
DispGraph
Horizontal -
Horizontal -
Horizontal -
Horizontal -
Vertical -
Vertical -
Vertical -
Vertical -
A+1→A
End
Return

Lbl SET
ref(1,1,5,22)
ref(1,1,5,22)
rref(2,2,3,20)
rref(2,2,3,20)
Return

Lbl W
0→A
Repeat A=40
DispGraph
A+1→A
End
Return

Lbl TSP
Plot1(r₁,r₂,r₃)
Plot1(r₁+8,r₂,r₃+16)
Return

Lbl FSP
Plot1(r₁,r₂,r₃)
Plot1(r₁+8,r₂,r₃+16)
Plot1(r₁,r₂+8,r₃+32)
Plot1(r₁+8,r₂+8,r₃+48)
Return

Lbl SSP
Plot1(r₁,r₂,r₃)
Plot1(r₁+8,r₂,r₃+16)
Plot1(r₁+16,r₂,r₃+32)
Plot1(r₁,r₂+8,r₃+48)
Plot1(r₁+8,r₂+8,r₃+64)
Plot1(r₁+16,r₂+8,r₃+80)
Return

Lbl RCL
conj(Pic0,L₃,768)
conj(Pic00,L₆,768)
Return

calc84maniac · « **Reply #1765 on:** May 18, 2011, 08:24:25 pm »

I don't know about your problem, but I do have an optimization:

Code: [Select]

If X≤P:Text(4,55,XDec):End
If X≤Q:Text(26,55,XDec):End
If X≤R:Text(48,55,XDec):End
If X≤S:Text(70,55,XDec):End
If X>P:Text(4,55,PDec):End
If X>Q:Text(26,55,QDec):End
If X>R:Text(48,55,RDec):End
If X>S:Text(70,55,SDec):End

becomes

Code: [Select]

Text(4,55,min(X,P)Dec)
Text(26,55,min(X,Q)Dec)
Text(48,55,min(X,R)Dec)
Text(70,55,min(X,S)Dec)

ralphdspam · « **Reply #1766 on:** May 18, 2011, 10:52:37 pm »

Quote from: DJ_O on May 11, 2011, 02:43:26 pm

That sounds nice, but then for the contest somebody will have to help me judging entries, because I haven't coded Axe in almost one year and I'll need to scan every entry to make sure only a certain percentage of Asm() and your new ASM feature is used, and sadly I might not make the difference between Axsembly stuff and regular Axe code.

So, if there are ASM commands, will they count as an Axe command? Hehe

Deep Toaster · « **Reply #1767 on:** May 19, 2011, 11:09:07 am »

Quote from: DJ_O on December 31, 1969, 07:00:00 pm

2: ASM code must not compose more than 10% of your program code and the only allowed way to use assembly is via the Asm() command. The remaining 90% must be Axe or Axiom code. (See previous rule for Axiom information). Other languages such as TI-BASIC are NOT allowed.

Taken care of already

Quigibo · « **Reply #1768 on:** May 22, 2011, 11:27:05 pm »

Hmm... the problem with the new constants runs deeper than I thought. The normal static pointers optimize with the assumption they will eventually be replaced by a value within the normal pointer range during the second pass since their actual value is unknown during the first pass. Constants on the other hand ARE known on the first pass and this replacement model doesn't work in that case since the value is never updated during the next pass. Therefore, this requires me to modify some very core routines.

So because I'll be rewriting this part anyway, I might finally add a request from a while ago which was to allow static variables to be used before they're defined. So you can put data at the end of your program instead of only the beginning. This has some drawbacks however: First, it will be harder to debug errors. Since the compiler won't know if a symbol is defined until the end of the source, it would not be able to scroll to the specific place it was used and would instead have to just display "Missing: Str1A" for example, just like labels. Secondly, it uses up an extra symbol in the symbol table for each use until its finally defined. As an example of this, here is the symbol count for this program:

Code: [Select]

:.TEST
:Str1       ;1 Symbol
:Str1       ;2 Symbols
:Str1       ;3 Symbols
:"H"->Str1  ;Now defined, so 1 symbol from now on
:Str1       ;1 Symbol
:Str1       ;1 Symbol
:Str1       ;1 Symbol

I'm guessing this will result in a lot of "MAX SYMBOLS" errors if people start using this too much since you only get 153 symbols to use. What are everyone's thoughts about this? Is it worth the extra effort? I feel like most people who would use this would want to use it on large programs in which case they would probably run out of symbols anyway and be forced to move them back on top, which is a very annoying process.

Runer112 · « **Reply #1769 on:** May 22, 2011, 11:29:08 pm »

I personally vote for another pass. That would solve the problem, wouldn't it? For me, even massive programs still compile in only a few seconds. I think giving up an additional second or two would be acceptable to allow for more free program structure.

Author Topic: Axe Parser (Read 553474 times)

Quigibo

Re: Axe Parser

Michael_Lee

Re: Axe Parser

Runer112

Re: Axe Parser

Freyaday

Re: Axe Parser

Quigibo

Re: Axe Parser

Hot_Dog

Re: Axe Parser

Builderboy

Re: Axe Parser

ztrumpet

Re: Axe Parser

DJ Omnimaga

Re: Axe Parser

ikemike

Re: Axe Parser

calc84maniac

Re: Axe Parser

ralphdspam

Re: Axe Parser

Deep Toaster

Re: Axe Parser

Quigibo

Re: Axe Parser

Runer112

Re: Axe Parser