Amoszine 9/Amoszine 9 (Disk 3 of 3).adf/Steves_ALister_Source.lha/AMOSLIST_0595 2

Text File | 1992-03-06 | 92KB | 2,719 lines

X-Status:

[e-mail] (Paul Hickman) wrote on 11.05.1995 some
text under the subject Re: Software Failure. I can't leave this
uncommentated ;-)

PH> If it crashes at the end of a loop / if block, the probable cause is
PH> this:  When an AMOS command is executed, it receives its parameters
PH> on a stack using register A3. If must remove all these from the stack
PH> by adjusting A3 before it returns. The address to loop/jump to for
PH> if/loop commands is also on A3, so if a command does in the loop does
PH> not remove all of its parameters correctly, your program will jump to
PH> the wrong place. Can you mail me the section of code in question?

That's what I wanted so say, too %-)))

Bye, Chris Hodges <[e-mail]>       __    __
A500/290/CD/'30er; German Extension-Coder   __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Ardcrony (n.)
  A remote acquaintance passed off as 'a very good friend of mine' by
  someone trying to impress people.
(from: "The Deeper Meaning of Liff")

F
X-Status:

[e-mail] (Greg Cox) wrote on 10.05.1995 some text under the
subject List Dead or ?. I can't leave this uncommentated ;-)

GC> Haven't seen any post on the list for almost a week now is the list:

Nice to hear that somebody else had got this problems... from 3.5.95
to 16.5.95 I got not a single internet mail...

I was very frustrated, you can imagine, eh?

GC> 1) Down

That's the only possible reason... :)

GC> 2) Dead

NO! ;-)

I did send a mail to amos-list, but I think it got lost:
--------------------------------------------------------------
BET:Problems :-(
MID:[e-mail]
EDA:19950510120705S+2
EMP:[e-mail]

Hello!

It's now one week ago I got my last amos-list or internet mail... I'm
now changing to a new address... to achieve maximum mail reliability,
please SUBSCRIBE me to *both* [e-mail] and
[e-mail] if possible.

If somebody has written me PMs and hasn't got a reply yet, please
resend it to [e-mail]...

Thank you for listening. ;-)

Greetings
     Chris Hodges <[e-mail]>
---------------------------------------------------------------

Bye, Chris Hodges <[e-mail]>       __    __
A500/290/CD/'30er; German Extension-Coder   __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Ardentinny (n.)
  One who rubs his hands eagerly together when he sits down in a
  restaurant.
(from: "The Deeper Meaning of Liff")

F
X-Status:

Can anyone send me an example of how the Turbo Plus Scene Editor is used? I do
not know the best way to detect walls etc.

F
X-Status:

Wanted beta testers (only two or three)

I've just finished writting a utility for amos that takes a anim (Dpaint
mode 5 animation) and converts it to a sprite/or bob bank.

The idea is that you can create your bobs in Dpaint (or any similar paint
program) and animate them there until you get nice smooth movement. Then
using my utility you can grab this animation as bobs for use in amos.

The first colour (background) must be black, but otherwise it's all
automatic, including automagically checking for bob sizes.

I now need a few beta testers to test it and give me feedback and
suggestions.
I will only give this out to a few people (I don't want a lot of buggy
versions going out and getting passed around before the final version is
finished), and I expect feedback.
Please do not ask for it if you are not willing to test it and give
feedback.
Darryl

-- Via DLG Pro v1.0


               #####\             _             /#####
               #( )# |          _( )__         | #( )#
               ##### |         /_    /         | #####
               #" "# |     ___m/I_ //_____     | #" "#
               # O # |____#-x.\ /++m\ /.x-#____| # O #
               #m.m# |   /" \ ///###\\\ / "\   | #m.m#
               #####/    ######/     \######    \#####

                 EX nightclub DJ (currently looking for work)

F
X-Status:

On Wed, 17 May 1995, Larry Mcgahey wrote:

> Can anyone send me an example of how the Turbo Plus Scene Editor is used? I do
> not know the best way to detect walls etc.
>

    I don't know anything about the Turbo Plus Scene Editor but if
        you mean that once you have a "map", how do you perform collision
        detection and object interaction then I can help you.

        By map, I mean that you some data that descibes how your scene or
        screen is to be created.

        In general you have to transform from screen- or object-space into
        map-space.  Once you are working in map-space you can easily
        determine what to do.

        To begin, lets use an example of a simple break-out (or araknoid)
        type game.  The map you create would represent the layout of
        level: this is map-space.  To construct the screen-space you
        have transform the representation from the map to the screen.
        If you're using an extension, such as Turbo, then this is most
        likely a built-in command.  This step would actually paste
        the tiles (or icons) to the screen.  Now the "ball" that
        travels around the screen is your object-space.  You have
        to transform the object-space into the map-space to determine
        if the ball has collided with a brick, a boudary, or nothing.

        To transform from object-space to map-space, you have to find
        how the current position of the object (the ball) on the screen
        corresonds to its location in the map.  For this case of
        a simple break-out game you would do the calculations as follows:

            MAP X Co-ordinate = (BALL X Co-ordinate) / (TILE X Size)
            MAP Y Co-ordinate = (BALL Y Co-ordinate) / (TILE Y Size)
            (NOTE: / here means integer division)

        so if the ball was at (220,100) on the screen and the tiles
        were 16x16 pixels then the ball would be at (13,6) in the
        map.  It should be noted here that in this case the screen
        and object-space are equivalent, but they won't always be.

        Now that you know the ball is at location (13,6) you have to
        look up into the map to find what "tile" is at (13,6).  This
        look up depends on how you have implemented your map-datatype.
        Once you know what tile the ball is on you have to decide what
        to do.  I usally implement this by having a set of values associated
        with each tile, these values could represent what type of tile
        this was.  Once you know what tile the ball is on, say 47, you
        have to look up the type of this tile.  This value could
        represent that their was no collision (i.e. a background tile)
        or that it has struck a brick.  If there was no collison then
        you do nothing but if their was a collison then you have to
        remove the brick from the screen and invert the sign on the
        balls y-vector.

        Of course the first value of the tile could give the type and
        once knowing it is a brick you could go and look up in another
        set of values what type of brick it was you struck.  This is
        all up to you.

        This can easily be extended to scrolling platform games, such
        as mario bros.  If you want this explained or you need more
        detail just ask.  I hope this helps.

        - Mike Sikorsky
          (Dept. of Computer Eng. University of Alberta)



F
X-Status:


    It seems noone know it... if I put a RUN instruction AMOS looks for
an AMOS source file and doesn't accept a compiled one.

    I have the compiler for Amos1.21 and I wish to chain two compiled
programs, but I can't do that. Please help me!

        Andy - [e-mail]
            http://www.cli.di.unipi.it/~naimoli/hi.html

F
X-Status:

Hi all,

I tried (a long time ago) to open the req. library using amos1.3
This worked fine. But I could not open the requester, everytime I
got a guru (8000 0004). So I started investigating but could not find
anything wrong.

The openlib-function from exec can be called as follows:

A1=pointer to "req.library"+chr(0) string
D0=version (eg. 0)
result=Execall(-552)

Works like a breeze and returns the library base in 'result'
Ok, you'd say. Just to test this, I tried for example a deek to
the library-base minus offset for some function. And indeed this
returns the op-code for the JMP command. A leek to the following four
bytes indeed shows a perfectly allright address within the 68k's space.

To simplify things even more I tried:

_EXEBASE = Leek(4)

This returns the current base of the 'exec.library' in _EXEBASE.
To open a library you'd think that a simple call to
call _exebase-552
and having set the a1 and d0 registers would work. Not so.
It will crash. Only calls to a library function that does NOT
need parameters will work.
What does this suggest ?, well clearly the 'Call' command does
not seem to be interested in the areg and dreg-settings.

Does anyone know if this can be overcome easily or do I have
to use a 'bank' with some MOVE.L's to registers and THEN the
call ? (rather bothersome I guess and close to the AmosPro
Lib Call command!)

'AMOS 1.3, load a library:
B$="intuition.library"+Chr$(0)
Areg(1)=Varptr(B$)
Dreg(0)=0

  'This WON't WORK!!
   _EXEBASE=Leek(4)
   Call Leek(_EXEBASE-552+2) : Rem openlin-function


 _INTBASE=Dreg(0)

  'And this WILL work *help*
   _INTBASE=Execall(-552)

Also, does anyone know what place to put the 'bitmap'-struct when
using amos ?
I tried opening a screen with intuition and intcall. Works ok.
Then I supply the screenbase with 'CUSTOM-BITMAP' (bit 6 of some offset)
and then I open an Amos-screen with the same dimension and put use:
loke (scr_base+184+8),leek (screen base)
loke (scr_base+184+12),leek (screen base+4)
to replace the pointers of the bitmap to my own amos-bitmaps
Works fine, when a 'rethinkdisplay' is called it shows the contents of
the amos-screen in the intuition-screen.
But this way, I trash the two bitplanes from the original intuition
screen!
(This way of putting an amos-screen in an intuitionscreen is VERY use-
full because the REQTOOL-requester will appear on your 'amos-screen'
in intuition, however, you'll have to rebuild you bitmap).

Sorry for the long message. I hope anyone will comment on this ?

---------------------------------------------------------------
Martijn Wehrens, Theoretical Chemistry Group Utrecht University
email [e-mail]
---------------------------------------------------------------

F
X-Status:

Hi Stephen,

Here's the commentated doscalls for reading a dir file
for file:

---------------------------------------------------

Input "drive:";A$ : Rem      input a$ :-)
A$=A$+Chr$(0) : Rem          add a null to a$
' strings are terminated by a null-sting, not in AMOS however
' so therefore, we add our ouwn zero.
Dreg(1)=Varptr(A$) : Rem     put a$-start into reg D1
Dreg(2)=-2 : Rem             put -2 into reg D2
' this is the mode we want to use for our 'lock':
' mode dreg(2) = -1 writing, only calling task allowed
' mode dreg(2) = -2 reading, all task may acces file
D1=Doscall(-84) : Rem Lock   call 'lock = Lock ( name, mode)
'                                   D0  =  -84    D1    D2
' D0 (called variabele D1 here :-) contains the address of the
' file-handle for the system
' or 0 if the file could not be found or something else went wrong
If D1<>0 : Rem                    not zero, we made it
   Dreg(1)=D1 : Rem               put 'Lock' in D1 for use with 'Examine'
   Reserve As Work 6,260 : Rem    workspace for 'fileinfo block'
' in C one would 'allocmem' memory for this purpose
   Dreg(2)=Start(6) : Rem         addres of space in D2
   D2=Doscall(-102) : Rem         Examine, i.e. read the fileinfoblock
   If D2<>0 : Rem                 if this was ok (not zero)
      Do
         Dreg(1)=D1 : Rem         put 'Lock' in D1 again
         Dreg(2)=Start(6) : Rem   address of 'free space' again
         D2=Doscall(-108) : Rem   Exnext
' (i.e. read the next file in the dir)
         Exit If D2=0 : Rem       not succesfull (last file or worse)
         R=Start(6)+8 : A$="" : Rem  get filename (from pos 8 onward)
' in Apro one can use the Peek$ - statement of course:
' A$=peek$(Start(6),30,chr$(0))
' I do *NOT* know where the filename actually is in this datastruct,
' if anyone knows: say so!
         Repeat : A$=A$+Chr$(Peek(R)) : Inc R : Until Peek(R)=0
' so a$ now contains the filename!
         If Right$(A$,5)<>".info" : Rem leave out the .info's
            If Leek(Start(6)+4)<0 : Rem is this a file ?
               Print Paper$(1);A$ : Rem yep, file
            Else
               Print Paper$(0);A$ : Rem directory...
            End If
         End If
      Loop
   End If
End If

---------------------------------------------------------------
Martijn Wehrens, Theoretical Chemistry Group Utrecht University
email [e-mail]
---------------------------------------------------------------

F
X-Status:


    Just incase some of you were wondering about my alias.....

My real name is Tim Lewis.
My alias is Eddie.

Why?  Well, I'm a big fan of Iron Maiden and for those of you who don't know
they have a mascot called "Eddie"!!!!  One things leads to another, advocacy
upon my part and then, what do you know!  I end up with the nickname of ...

    See if you can guess .....

    Thats right Eddie!!!! :-)

This might sound extremely sad & on reflection it is.  Then again I don't
really care!!!!  Up the Irons!

AND NOW A MORE RELEVANT MESSAGE ...........

    The AMOS scroll routine sucks!  I need to scroll an area of screen which
is about 154x154 pixels.  This needs to be done STRICTLY within these limits.
I can't use dual screens either because of the necessary display of this
screen.  Are there any clever bods out there who can (or want!) to tell me how
to this by screen memory copying etc.  I know it can be done, but I don't
really have the spare time to do it myself!!

    Any other solutions would also be appreciated.  It might send my mind
racing of in a new direction.

    I'm writing a LARGE!!! RPG at the moment.  I should hopefully have a Beta
test version sometime over summer!  If anyone is interested .....  If anyone
has any ideas about what they would like to see in an RPG then let me know.
I will try and incorporate them if I can.  I do need some fast scrolling
routines though.  HELP!!!!!!!

A POINT TO REFLECT UPON...........

    They may have been a good side to the C= liquidation.  Did you notice the
frequency of all the new stuff uploaded to Aminet increased since that
infamous start of procedings! :-)  Every cloud has a silver lining.

Long Live The AMIGA!

Tim.

F
X-Status:

I need a GET statement for AMOS Pro.  The a$=Input$(2,1) instruction will read
a single character from a file, but it WAITS until something is entered.

I need something that will check a file for a single character and GO ON if
there is nothing there.

In any normal basic, Get #2, A$ would work, but AMOS has no comparable
instruction that I can find.  I tried using:
Do
Z=Eof(2)
if Z=True Then Loop
A$=Input$(2,1)

But Z is ALWAYS true, so this method doesn't work either.

The program is monitoring the STDIN channel and waiting for a key.  I am trying
to make it so that the routine doesn't hang if there is no input from STDIN.  I
need a `Get a single byte, go on if nothing there' instruction.

Well met and godspeed,
                      Giark

F
X-Status:

Hi everyone, the other day somebody sent to the list a routine that reads
in a directory from a disk, what I would like to know, is can this routine
be modified to get the size of the files in the directory as well? If somebody
could help me with this I would appreciate it.

Cori Skagen
([e-mail])


F
X-Status:

[e-mail] (ryan nicholas d) wrote on 16.05.1995 some
text under the subject WB + CLI Compiled code.. I can't leave this
uncommentated ;-)

rd> The problem is this, I've been trying to get tool-types working for an Amos
rd> prog, but for some utterly weird reason, it doesn't seem to work. I would

Mine works ;-) The only stupid thing was to find out the name of the
own program.

rd> usually throw MonAn (Monitor/Debugger that came with DevPac3) at the task, but
rd> the code I wrote to read the tool types is in an included asm procedure, and I
rd> would have to run through the entire Amos -startup code (ie crap) before
rd> getting anywhere near where I am.

Just set an 'Illegal' asm command just before your tool types code,
then assemble it again and feed it into a procedure... save and start
monam... run through, until you come to this illegal exception and
then press Ctrl-S to skip it... then you are on the very right point
and you can debug your code as you like to ;-)

rd> Help! Has anybody got this working - and no, I've looked - Easylife doesn't
rd> *seem* to have it.

Yes. :-)

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Banteer (n.)
  (Archaic) A lusty and raucous old ballad sung after a particularly
  spectacular araglin (q.v) has been pulled off.
(from: "The Deeper Meaning of Liff")

F
X-Status:


Hi everyone!

Does anyone know how to do the equivalent of an "avail flush" in AmosPro?

I have a game that plays music using Paul Reece's Enhanced Music
Extension v4.64.  I find that if I play the game after I've been using
my Amiga for a while, the music causes random bits of garbled graphics
to show up on my screen - obviously these are unused bits of data lurking
in memory.  If I do an "avail flush" in CLI and then run the game, there
is no flickering.

So does anyone know how to flush memory? I'd welcome any help :))

*wave*
Stuart.

 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    Stuart Matthews - University of Queensland - [e-mail]

 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


F
X-Status:



> The program is monitoring the STDIN channel and waiting for a key.  I am trying
> to make it so that the routine doesn't hang if there is no input from STDIN.  I
> need a `Get a single byte, go on if nothing there' instruction.
>

Easylife has an instruction to do this, but I can't for the life of
me remember what I called it - It probably begins with "El", but
after that its anyones guess although to syllables "in" and "get"
probably feature somewhere in it :-). Look in the manual under
AmigaDos commands.

--------------------------------------------------------
 WORK: Eurotherm Controls, Worthing
 Email: [e-mail]

 UNIVERSITY: Imperial College, London
 Email: [e-mail]
--------------------------------------------------------
The opinions expressed above to not represent the views of
Eurotherm Controls plc, or Imperial College, or in fact Paul
Hickman, but are in fact the views of a small furry alien from
the planet zorg (called fred), so if you don't like them, sue
him/her/whatever. Death to all tennis players.

F
X-Status:


> Subject:       load_lib
> To:            [e-mail] (Amos-Mailing-List)
> Date:          Thu, 18 May 95 12:11:46 METDST
> From:          [e-mail] (Martijn Wehrens)
> Reply-to:      [e-mail]

[cut]

> To open a library you'd think that a simple call to
> call _exebase-552
> and having set the a1 and d0 registers would work. Not so.
> It will crash. Only calls to a library function that does NOT
> need parameters will work.
> What does this suggest ?, well clearly the 'Call' command does
> not seem to be interested in the areg and dreg-settings.
>
Call does read the Areg & Dreg settings, but it also expects the
routine called to preserve the values in registers A3-A6. Most
library routines will not do this so AMOS may crash after calling
them. LibCall knows this and stores the registers first. Also, some
libraries except to be called with their own base in register A6, and
call would not do this.

> Does anyone know if this can be overcome easily or do I have
> to use a 'bank' with some MOVE.L's to registers and THEN the
> call ? (rather bothersome I guess and close to the AmosPro
> Lib Call command!)

Thats the best way. You can of course make use of the fact that the
call command can take additional parameters which are placed on a
stack off register A3 to pass the library base & offset of the
routine to call. E.g. If the AMOS Code was "Call BNKNO,LIBBASE,OFFSET"

The bank code would be

    move.l    (a3)+,d7    ;d7=offset
    add.l        (a3)+,d7    ;d7=address to call
    movem.l    a3-6,-(sp)    ;Preserve registers
    move.l    d7,a6
    jsr        (a6)
    movem.l    (sp)+,a3-6
    rts

NOTE: This assumes the library call does not use register D7, but I
think your safe for most calls.
--------------------------------------------------------
 WORK: Eurotherm Controls, Worthing
 Email: [e-mail]

 UNIVERSITY: Imperial College, London
 Email: [e-mail]
--------------------------------------------------------
The opinions expressed above to not represent the views of
Eurotherm Controls plc, or Imperial College, or in fact Paul
Hickman, but are in fact the views of a small furry alien from
the planet zorg (called fred), so if you don't like them, sue
him/her/whatever. Death to all tennis players.

F
X-Status:



>   --Andy Church ([e-mail])
>     WWW: http://www.mbhs.edu/~achurch/

A friend of mine sent me the Demo version 1.3 of your extension.
I have tried it. But I didn't work. I had an message saying that tha Amos
could not loaded the extension and to use the default configuration.
This before I started SnoopDos to see where the Amos failed. The strange
thing is that the extension was loaded!
I repeated the event. Without SnoopDOs... nothing. Whith snoopdos running
(and so delaying everything) Amos loaded the extension and started
normally, but seeing the about on the loaded extension didn't give any
name in the slot n14. The extension was there but his name not.
What's wrong? It is a regular thing (I think it is not !)?
Has the registered version the same problems?

Has anyone had my problems?

THanks

MeF

F
X-Status:

Hi,

Ciro Skagen wrote:

... a routine that reads in a directory from a disk, what I would
 like to know, is can this routine ... size of the files

Yes, it can.

When the 'Examine'-call is made, the 'FileInfoBlock' is read
This block is the header of the file and contains all information
about the file such as the protection flags, the size, the time and
date and the filenote.
So all the information is already in the bank when the filename is read
from it.
I do not know where what information is stord, however :-(
But I try to print some more of the bank used and find the place
where the size of the file is kept in it.

---------------------------------------------------------------
Martijn Wehrens, Theoretical Chemistry Group Utrecht University
email [e-mail]
---------------------------------------------------------------

F
X-Status:

Hi all,

The other day, Tumu / BFB team <[e-mail]
posted a message containing the code needed to make use of the
reqtools requesters rtezrequest and rtfilrequest. They used a function

_RTFILENAMEPTR=Ma Malloc(108,Mem Public+Mem Clear)

in their code. I suppose this is to do a 'allocmem' from exec and this
can be done without the extension they propose. The easiest way is
to make use of :

Reserve as work 6,108
Poke$ Start(6),String$(chr$(0),108)
_RTFILENAMEPRT=Start(6)

Further, they use the Elmem$ functions to read out the text, this
can also be done with:

_RTFILE$=Peek$(_RTFILENAMEPRT,108,chr$(0))

which leads to the exact same result. However, there may be a reason
to do this as they did, if there is: Please post a message explaining
why!

Also:
I use the following code in my program to create an intuition-screen
and replace the bitmap-pointers with the ones from the amos-screen.
That way, you'll see your amos-screen in the intuiscreen, and thus it
should be possible to have the reqtools on your amos-screen. After
the requesters are done, you can put the intuiscreen away.
There are two serious flaws in this design (other than the ones you
might detect :-)
a) the openscreen-call will create two bitmaps i don't use.
   (this can be overcome because the bitmaps can buffer the
    amos-screen but it costs some chipmem...)
b) Nico Francois clearly states in his documents on RT that one
   should *not* use the 'screen'-tag to decide on which screen the
   requester is to come. One should use 'pubscreen' instead.
The problem is: how do I set the 'screen'-tag in the taglist or what
ever it is called in amos ? I studied the c-source 'demo.c'-file in the
distribution but I have no idea where I must 'loke' my screen-base in
order to have RT recognize it.

Anyone ?

code:
-------------------------------------------------------------snap---
Screen Open 0,640,256,4,Hires : Paper 0 : Pen 1 : Cls 0
_INTSCR[0,640,200,2,"My Screen"]
SCR=Param
Amos To Back
Dir
'
Lib Open 1,"reqtools.library",0
_RTEZREQUEST["Pull the screen!","return to amos"," "]
Lib Close
'
Amos To Front
'opruimen met close-screen
Areg(0)=SCR : If SCR=0 Then End
OK=Intcall(-66)
'-------------------------------------------
Procedure _INTSCR[NR,_WIDTH,_HEIGHT,_PLANES,NAM$]
'struct newscreen
   Reserve As Work 6,32 : S=Start(6)
   Doke S,0 : Add S,2
   Doke S,0 : Add S,2
   Doke S,_WIDTH : Add S,2
   Doke S,_HEIGHT : Add S,2
   Doke S,_PLANES : Add S,2
   Poke S,0 : Inc S
   Poke S,1 : Inc S
   Doke S,32768 : Add S,2 : Rem screen mode(hires)
   Doke S,15 : Add S,2
   Loke S,$0 : Add S,4
   A$=NAM$+Chr$(0) : FRE=Free
   Loke S,Varptr(A$) : Add S,4
   Loke S,$0 : Add S,4
   Loke S,$0 : Rem end of struct
'
Areg(0)=Start(6) : Rem where to find 'struct newscreen'
SCR=Intcall(-198) : Rem openscreen

   If SCR=0 Then Pop Proc[SCR] : Rem something is wrong!
   Erase(6) : Rem trash 'struct newscreen'
   Amos To Back
'set bit for 'Custom Bit Map'
   Doke SCR+20,Deek(SCR+20) or %1000000
'now to open an amos-screen of requested size:
   Screen Open NR,_WIDTH,_HEIGHT,2^_PLANES,Hires
   Cls 0 : Pen 1 : Paper 0
   SB=Screen Base : Rem get address for planes
   For R=0 To _PLANES-1
      Loke SCR+184+8+R*4,Leek(SB+R*4) : Rem   replace bitmap-pointers
   Next
   Areg(0)=SCR
   OK=Intcall(-252) : Rem  screen to front
   OK=Intcall(-384) : Rem  remake display (to see change)
End Proc[SCR]
Procedure _RTEZREQUEST[_REQ$,_GADS$,_TITLE$]
   Dreg(0)=1
   Areg(0)=0
   RES=Lib Call(1,-$1E) : Rem make buffer
   If RES=0 Then Pop Proc[-1]
   ' 0 for filereq
   ' 1 for easy req.
   ' 2 font
   ' 3 screen
   Areg(1)=Varptr(_REQ$)
   Areg(2)=Varptr(_GADS$)
   Areg(3)=RES
   Areg(4)=0
   Areg(0)=0
   X=Free : Rem make sure the only _title$ is this one
   Loke _REQBUFFER+12,Varptr(_TITLE$)

   RES2=Lib Call(1,-$42) : Rem ezrequest

   Areg(1)=_REQBUFFER
   RES3=Lib Call(1,-$24) : Rem clear reqbuffer
End Proc[RES2]
--------------------------------------------------------------

Anyone, please ?

---------------------------------------------------------------
Martijn Wehrens, Theoretical Chemistry Group Utrecht University
email [e-mail]
---------------------------------------------------------------

F
X-Status:

    Can anyone tell me where to obtain Paul Reeces music extension? I am
currently running Amos Creator 1.34 or something like that with the supplied
music extension. Can Paul Reece's work in Amos Creator? What differences are
there between the two? Any bugs I should know about? Are there any large
memory consumption differences?

    I know, many many questions, but I'm a questionable kind of guy so if
anyone has answers to the above, please let me know!

            Robert Currie

p.s. Is anyone interested in playtesting an Eye of Beholder style modular
RPG game that was created in Amos?

F
X-Status:

[e-mail] (Martijn Wehrens) wrote on 18.05.1995 some
text under the subject readdir. I can't leave this uncommentated ;-)

MW>          R=Start(6)+8 : A$="" : Rem  get filename (from pos 8 onward)
MW> ' in Apro one can use the Peek$ - statement of course:
MW> ' A$=peek$(Start(6),30,chr$(0))

Shouldn't it be:

A$=Peek$(R,30,Chr$(0))

or

A$=Peek$(Start(6)+8,30,Chr$(0))

MW> ' I do *NOT* know where the filename actually is in this datastruct,
MW> ' if anyone knows: say so!

 STRUCTURE FileInfoBlock,0
   LONG    fib_DiskKey      ;
   LONG    fib_DirEntryType ; Type of Directory. If < 0, then a plain
file.
                ; If > 0 a directory
   STRUCT fib_FileName,108  ; Null terminated. Max 30 chars used for
now
   LONG   fib_Protection    ; bit mask of protection, rwxd are 3-0.
   LONG   fib_EntryType
   LONG   fib_Size          ; Number of bytes in file
   LONG   fib_NumBlocks     ; Number of blocks in file
   STRUCT fib_DateStamp,ds_SIZEOF ; Date file last changed.
   STRUCT fib_Comment,80    ; Null terminated. Comment associated
with file

   ; Note: the following fields are not supported by all filesystems.
   ; They should be initialized to 0 sending an ACTION_EXAMINE
packet.
   ; When Examine() is called, these are set to 0 for you.
   ; AllocDosObject() also initializes them to 0.
   UWORD  fib_OwnerUID      ; owner's UID
   UWORD  fib_OwnerGID      ; owner's GID

   STRUCT fib_Reserved,32
   LABEL  fib_SIZEOF        ; FileInfoBlock

MW>          Repeat : A$=A$+Chr$(Peek(R)) : Inc R : Until Peek(R)=0
MW> ' so a$ now contains the filename!

I don't understand the meaning of this line... the other things
should work... ;-)

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Abwong (vb.)
   To bounce cheerfully on a bed.
(from: "The Deeper Meaning of Liff")

F
X-Status:

[e-mail] (Martijn Wehrens) wrote on 18.05.1995 some
text under the subject load_lib. I can't leave this uncommentated ;-)

MW> Does anyone know if this can be overcome easily or do I have

No. AMOS requires the registers A3-A6 and D6/D7 to be unmodified and
many library calls *require* the library base address in A6...

I had the same problems, and so I wrote some workaround...

Long time ago, I tried this and it worked (although you could do some
optimations)...

main
.dr0    ds.l    14
.ar6    dc.l    0
.rout   dc.l    0
    movem.l d0-d7/a0-a6,-(sp)
    lea     .ende(pc),a0
    move.l  a0,-(sp)
    lea     .dr0(pc),a6
    move.l  (a6)+,d0
    move.l  (a6)+,d1
    move.l  (a6)+,d2
    move.l  (a6)+,d3
    move.l  (a6)+,d4
    move.l  (a6)+,d5
    move.l  (a6)+,d6
    move.l  (a6)+,d7
    move.l  (a6)+,a0
    move.l  (a6)+,a1
    move.l  (a6)+,a2
    move.l  (a6)+,a3
    move.l  (a6)+,a4
    move.l  (a6)+,a5
    move.l  (a6)+,a6
    move.l  .ar6(pc),a6
    move.l  .rout(pc),-(sp)
    rts
.ende   lea     .ar6(pc),a6
    move.l  a5,-(a6)
    move.l  a4,-(a6)
    move.l  a3,-(a6)
    move.l  a2,-(a6)
    move.l  a1,-(a6)
    move.l  a0,-(a6)
    move.l  d7,-(a6)
    move.l  d6,-(a6)
    move.l  d5,-(a6)
    move.l  d4,-(a6)
    move.l  d3,-(a6)
    move.l  d2,-(a6)
    move.l  d1,-(a6)
    move.l  d0,-(a6)
    movem.l (sp)+,d0-d7/a0-a6
    rts

I assembled it and saved it into a bank... the first 15 longwords
hold the data registers d0 to d7 and a0-a6...

The address of the routine to call is placed in the 16th longword
(i.e LibBase+Offset)

The results can be read out from the same addresses the registers
were written into...

Then a wrote some library-procedures that allowed librarycalls in a
way like:

OPENLIB["dos"]
LIBCALL2["dos","Lock",Varptr(NAME$),-2]
LOCK=RES : If LOCK=0 Then Print "ERROR!" : Return
LIBCALL2["dos","Examine",LOCK,Start(9)]
If RES=0 Then Print "Empty!" : LIBCALL1["dos","Unlock",LOCK] : Return
...

It was rather slow (due of the offset search), but it worked...


MW> Also, does anyone know what place to put the 'bitmap'-struct when
MW> using amos ?

Hmm... I don't really know what you mean... give me some more
information please :)

MW> I tried opening a screen with intuition and intcall. Works ok.
MW> Then I supply the screenbase with 'CUSTOM-BITMAP' (bit 6 of some offset)

Are you opening the screen using CUSTOM_BITMAP or hacking it into
afterwards?

MW> and then I open an Amos-screen with the same dimension and put use:
MW> loke (scr_base+184+8),leek (screen base)
MW> loke (scr_base+184+12),leek (screen base+4)

You should use logbase(x) though...

MW> to replace the pointers of the bitmap to my own amos-bitmaps
MW> Works fine, when a 'rethinkdisplay' is called it shows the contents of
MW> the amos-screen in the intuition-screen.
MW> But this way, I trash the two bitplanes from the original intuition
MW> screen!

There a other way round... there *are* intuition elements and
structures for a AMOS screen already (rastport, bitmap, layer and
layerregion). Why not implement them directly into a intuition
screen?

MW> (This way of putting an amos-screen in an intuitionscreen is VERY use-
MW> full because the REQTOOL-requester will appear on your 'amos-screen'
MW> in intuition, however, you'll have to rebuild you bitmap).

You're right... But it is very useless to write intuition programs
using amos because
a) you create executables of huge size.
b) you need a lot of system experience
c) you could write it more easyly in C++.
d) AMOS causes a overhead which requires more processor time.
e) AMOS programs do not multitask (i.e if you start two amos
programs, one will be frozen)

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Acklings (pl.n.)
  The odd twinges you get in parts of your body when you scratch
  other parts.
(from: "The Deeper Meaning of Liff")

F
X-Status:

[e-mail] (Mike Pelletier) wrote on 18.05.1995 some
text under the subject Get. I can't leave this uncommentated ;-)

MP> I need something that will check a file for a single character and GO ON if
MP> there is nothing there.

Note that this will work only on RAW: Windows, not on CON: as they
only send their characters after a linefeed...

MP> The program is monitoring the STDIN channel and waiting for a key.  I am trying
MP> to make it so that the routine doesn't hang if there is no input from STDIN.  I
MP> need a `Get a single byte, go on if nothing there' instruction.

I added a procedure to my STDINOUT assigning program... note that it
will take the STDOUT for text input and a raw console for keyboard
input...

Amos To Back
_ASSIGNSTDOUT[1]
'_ASSIGNSTDIN[2]
Open In 2,"RAW:/200//100/RAW-Console/AUTO/WAIT/CLOSE"
Print #1,"Please enter a string: ";
T$=""
Do
  ' The syntax for _WAITFORCHAR is channel,timeout
  _WAITFORCHAR[2,50000]
  BOOL=Param
  If Not BOOL
    Print #1,".";
  Else
    A$=Input$(2,1)
    Exit If A$=Chr$(27)
    Print #1,Chr$(10);"You pressed a key: ";A$;Chr$(10);
    T$=T$+A$
  End If
  If Inkey$=Chr$(27) Then Exit
Loop
_REMSTD[1]
'_REMSTD[2]
Close 2
Amos To Front
Print "You typed: ";T$
End
Procedure _ASSIGNSTDIN[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  STIN=Doscall(-54)
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)<>0 Then Error 96
  Loke AD,STIN : Doke AD+4,$200 : Loke AD+6,0
End Proc
Procedure _ASSIGNSTDOUT[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  STIN=Doscall(-60)
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)<>0 Then Error 96
  Loke AD,STIN : Doke AD+4,$100 : Loke AD+6,0
End Proc
Procedure _REMSTD[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)=0 Then Error 97
  Loke AD,0 : Doke AD+4,0 : Loke AD+6,0
End Proc

'This is the new procedure:

Procedure _WAITFORCHAR[CHAN,TIMOUT]
  If CHAN<1 or CHAN>8 Then Error 23
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)=0 Then Error 97
  Dreg(1)=Leek(AD)
  Dreg(2)=TIMOUT
  RES=Doscall(-$CC)
End Proc[RES]

Note that again you should not break the program... use the ESC key
instead...
(otherwise type Loke Areg(5)+$8BC,0 in the direct mode immediately
after the interruption)

I hope this will help you out a bit...

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Albacete (n.)
  A single surprisingly long hair growing in the middle of nowhere.
(from: "The Deeper Meaning of Liff")

F
X-Status:

[e-mail] (Martijn Wehrens) wrote on 19.05.1995 some
text under the subject rtez_on_myscreen?. I can't leave this
uncommentated ;-)

MW> The problem is: how do I set the 'screen'-tag in the taglist or what
MW> ever it is called in amos ? I studied the c-source 'demo.c'-file in the
MW> distribution but I have no idea where I must 'loke' my screen-base in
MW> order to have RT recognize it.

OpenScreen is obsolete... OpenScreenTagList (V36+) is the one you're
longing for...

This a working solution and requires 2.0+:

' OpenIntScreen by Chris Hodges (21-May-95)
Screen Open 0,320,256,32,0
Curs Off : Flash Off : Paper 0 : Pen 1 : Cls
' Assign TAG list
Gosub INITSCREENTAGS
' Reserve memory for a colormap structure
Reserve As Work 14,33*8
ST=Start(14)
For A=0 To 31
  C=Colour(A)
  Doke ST,A
  Doke ST+2,(C and $F00)/256
  Doke ST+4,(C and $F0)/16
  Doke ST+6,C and $F
  Add ST,8
Next
Doke ST,-1
' Create Taglist... add tags at your choice...
TAGSTART
TAG[SA_QUIET,-1]
TAG[SA_WIDTH,320]
TAG[SA_HEIGHT,256]
TAG[SA_DEPTH,5]
TAG[SA_TYPE,$F] : Rem CUSTOMSCREEN, $2=PUBSCREEN
TAG[SA_BITMAP,Leek(Screen Base+$150)] : Rem AMOS Bitmap structure
TAG[SA_COLORS,Start(14)]
TAG[0,0]
' Now open the screen...
Amos To Back
Areg(0)=0 : Areg(1)=Start(15)
SCR=Intcall(-$264) : Rem OpenScreenTagList
If SCR<>0
  ' Draw some thingies on the screen...
  Gr Writing 0
  For A=1 To 31
    Ink A : Text 31-A,100-A,"This is AMOS calling intuition!"
    Circle 200,170,A
  Next
  ' Provisory "wait left mouse button" routine... no not use
  ' in your programs, write system conform replacements please...
  Repeat : Multi Wait : Until(Peek($BFE001) and 64)=0
  ' Close screen (or wait until all windows have been closed)
  Repeat
    Areg(0)=SCR : RES=Intcall(-$42) : Multi Wait
  Until RES=1
End If
Amos To Front
Erase 15
Erase 14
End
' Screen Tags... refer to intuition/screens.i includes for detailed
' information on the tags...
INITSCREENTAGS:
  TS=$80000000+33
  SA_LEFT=TS
  SA_TOP=TS+1
  SA_WIDTH=TS+2
  SA_HEIGHT=TS+3
  SA_DEPTH=TS+4
  SA_DETAILPEN=TS+5
  SA_BLOCKPEN=TS+6
  SA_TITLE=TS+7
  SA_COLORS=TS+8
  SA_ERRORCODE=TS+9
  SA_FONT=TS+10
  SA_SYSFONT=TS+11
  SA_TYPE=TS+12
  SA_BITMAP=TS+13
  SA_PUBNAME=TS+14
  SA_PUBSIG=TS+15
  SA_PUBTASK=TS+16
  SA_DISPLAYID=TS+17
  SA_DCLIP=TS+18
  SA_OVERSCAN=TS+19
  SA_OBSOLETE1=TS+20
' *BOOLEANS*
  SA_SHOWTITLE=TS+21
  SA_BEHIND=TS+22
  SA_QUIET=TS+23
  SA_AUTOSCROLL=TS+24
  SA_PENS=TS+25
  SA_FULLPALETTE=TS+26
  SA_COLORMAPENTRIES=TS+27
  SA_PARENT=TS+28
  SA_DRAGGABLE=TS+29
  SA_EXCLUSIVE=TS+30
  SA_SHAREPENS=TS+31
  SA_BACKFILL=TS+32
  SA_INTERLEAVED=TS+33
  SA_COLORS32=TS+34
  SA_VIDEOCONTROL=TS+35
  SA_FRONTCHILD=TS+36
  SA_BACKCHILD=TS+37
  SA_LIKEWORKBENCH=TS+38
  SA_RESERVED=TS+39
  SA_MINIMIZEISG=TS+40
Return
Procedure TAGSTART
  Shared TAGST
  Reserve As Work 15,128
  TAGST=Start(15)
End Proc
Procedure TAG[TAG,V]
  Shared TAGST
  Loke TAGST,TAG : Loke TAGST+4,V : Add TAGST,8
End Proc

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Alltami (n.)
  The ancient art of being able to balance the hot and cold shower
  taps.
(from: "The Deeper Meaning of Liff")

F
X-Status:

I was trying to come up with an algrorthm the other day to determine the
most frequently occuring number in a group of numbers.

ie out of X numbers how many times does such a number occur and what is the
most frequent number.

Note: This is different from taking the average

I came up with a solution, but it was ugly and very badly coded. I was also
long!

Does anyone have any neat answers?

Darryl

-- Via DLG Pro v1.0


               #####\             _             /#####
               #( )# |          _( )__         | #( )#
               ##### |         /_    /         | #####
               #" "# |     ___m/I_ //_____     | #" "#
               # O # |____#-x.\ /++m\ /.x-#____| # O #
               #m.m# |   /" \ ///###\\\ / "\   | #m.m#
               #####/    ######/     \######    \#####

                 EX nightclub DJ (currently looking for work)

F
X-Status:

On Fri, 19 May 1995, Martijn Wehrens wrote:

> Hi all,
>
> The other day, Tumu / BFB team <[e-mail]
> posted a message containing the code needed to make use of the
> reqtools requesters rtezrequest and rtfilrequest. They used a function
>
> _RTFILENAMEPTR=Ma Malloc(108,Mem Public+Mem Clear)
>
> in their code. I suppose this is to do a 'allocmem' from exec and this
> can be done without the extension they propose. The easiest way is
> to make use of :
>
> Reserve as work 6,108
> Poke$ Start(6),String$(chr$(0),108)
> _RTFILENAMEPRT=Start(6)
>
> Further, they use the Elmem$ functions to read out the text, this
> can also be done with:
>
> _RTFILE$=Peek$(_RTFILENAMEPRT,108,chr$(0))
>
> which leads to the exact same result. However, there may be a reason
> to do this as they did, if there is: Please post a message explaining
> why!

You can use both ways to reserve the reqbuffer and rtfilenamebuffer but I
like using Easylife and Makes lib commands because they are easier to
use and Makes lib memory allocation routines are System friendly.

And I forgotted to mention that you'll need Easylife extension as well if
you are going to use original Reqtools routines without modifications.

    Tumu / BFB Team


F
X-Status:


> Date:          Fri, 19 May 1995 17:08:38 +1000 (GMT+1000)
> From:          Stuart Matthews <[e-mail]>
> To:            Discussion List <[e-mail]>
> Subject:       Flushing memory in AmosPro

>
> Hi everyone!
>
> Does anyone know how to do the equivalent of an "avail flush" in AmosPro?
>
Try and allocate an impossibly large amount of memory to a bank - the
Amiga OS memory manager will find it cannot, and perform an avail
flush to try to free the memory. Something like Reserve bank
14,$70000000 should to the trick.

Or you can execute the "Avail" CLI command from within AMOS!

--------------------------------------------------------
 WORK: Eurotherm Controls, Worthing
 Email: [e-mail]

 UNIVERSITY: Imperial College, London
 Email: [e-mail]
--------------------------------------------------------
The opinions expressed above to not represent the views of
Eurotherm Controls plc, or Imperial College, or in fact Paul
Hickman, but are in fact the views of a small furry alien from
the planet zorg (called fred), so if you don't like them, sue
him/her/whatever. Death to all tennis players.

F
X-Status:

On Thu, 18 May 1995 [e-mail] wrote:

> Hi everyone, the other day somebody sent to the list a routine that reads
> in a directory from a disk, what I would like to know, is can this routine
> be modified to get the size of the files in the directory as well? If somebody
> could help me with this I would appreciate it.
>
> Cori Skagen
> ([e-mail])

If you wanna find the size of a file just try:

open in 1,"filename"
L=Lof(1)
close 1

Greg


F
X-Status:

Does anyone have a script that shows how to use the scenes created with the
Turbo Plus Scene Editor?
How can I get the amos list regularly? I don't remember where I down loaded it
from.

F
X-Status:

>> Hi everyone, the other day somebody sent to the list a routine that reads
>> in a directory from a disk, what I would like to know, is can this routine
>> be modified to get the size of the files in the directory as well? If somebody
>> could help me with this I would appreciate it.
>>
>It can, just extract the size from file info block returned by
>examine/examine next. I don't have the offset to hand, but look in
>the dos.h / dos.i file of commodore includes, or any decent C/Asm
>programming book for a description of the FIB structure.
>
>But why are you doing this - I know AMOS is full of bugs but I've
>never had trouble with the DirFirst$() / DirNext$ functions!

I never had problems with them either, BUT: when you do a 'DirFirst$',
amos read the entire directory. So there's not much sense in having to
do a 'DirNext$' other than the patternmatching. (Which is also in
some extensions..;-) The advantage of reading it using calls is the
possibillity to realy read one entry at a time and thus allow the
(diskette-)user to select an entry before the entire directory is read.
More like the filerequesters.

---------------------------------------------------------------
Martijn Wehrens, Theoretical Chemistry Group Utrecht University
email [e-mail]
---------------------------------------------------------------

From svcs1.digex.net!amos-request Mon May 22 09:02:04 1995 remote from cristal
Return-Path: <[e-mail]>
Received: from svcs1.digex.net by cristal.mpx.com.au with smtp
    (Smail3.1.28.1 #3) id m0sDK0G-00006dC; Mon, 22 May 95 09:01 AEST
Received: by svcs1.digex.net id AA05309
  (5.67b8/IDA-1.5 for amos-out); Sun, 21 May 1995 16:09:03 -0400
Received: from www1.access.digex.net by svcs1.digex.net with SMTP id AA05305
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 16:09:01 -0400
Received: from phoenix.inka.de by www1.access.digex.net with SMTP id AB15894
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 16:07:38 -0400
Received: from mips.pfalz.de by phoenix.inka.de with bsmtp
    (Smail3.1.29.1 #13) id m0sDHBr-0004B1C; Sun, 21 May 95 22:01 MET DST
Received: from kiste.pfalz.de by mips.pfalz.de with cbsmtp
    (Smail3.1.28.1 #4) id m0sDFIQ-000CnNC; Sun, 21 May 95 20:00 MET DST
X-Zc-Via: [e-mail]
X-Zc-Via: [e-mail]
From: [e-mail] (Chris Hodges)
Subject: Re: Get
Message-Id: <[e-mail]>
Date: Sun, 21 May 95 09:53:38 CET
X-Mailer: MicroDot 1.10 [REGISTERED 000261]
X-Zc-Telefon: V+49-89/8005856
X-Zc-Post: Kennedystrasse 8; D-82178 Puchheim; Germany
Path: kiste.pfalz.de!sixpack.pfalz.de
To: [e-mail]
X-Gateway: ZConnect CL kiste.pfalz.de [Connectline/AmigaOS]
References: <[e-mail]>
Status: RO
X-Status:

[e-mail] (Mike Pelletier) wrote on 18.05.1995 some
text under the subject Get. I can't leave this uncommentated ;-)

MP> I need something that will check a file for a single character and GO ON if
MP> there is nothing there.

Note that this will work only on RAW: Windows, not on CON: as they
only send their characters after a linefeed...

MP> The program is monitoring the STDIN channel and waiting for a key.  I am trying
MP> to make it so that the routine doesn't hang if there is no input from STDIN.  I
MP> need a `Get a single byte, go on if nothing there' instruction.

I added a procedure to my STDINOUT assigning program... note that it
will take the STDOUT for text input and a raw console for keyboard
input...

Amos To Back
_ASSIGNSTDOUT[1]
'_ASSIGNSTDIN[2]
Open In 2,"RAW:/200//100/RAW-Console/AUTO/WAIT/CLOSE"
Print #1,"Please enter a string: ";
T$=""
Do
  ' The syntax for _WAITFORCHAR is channel,timeout
  _WAITFORCHAR[2,50000]
  BOOL=Param
  If Not BOOL
    Print #1,".";
  Else
    A$=Input$(2,1)
    Exit If A$=Chr$(27)
    Print #1,Chr$(10);"You pressed a key: ";A$;Chr$(10);
    T$=T$+A$
  End If
  If Inkey$=Chr$(27) Then Exit
Loop
_REMSTD[1]
'_REMSTD[2]
Close 2
Amos To Front
Print "You typed: ";T$
End
Procedure _ASSIGNSTDIN[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  STIN=Doscall(-54)
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)<>0 Then Error 96
  Loke AD,STIN : Doke AD+4,$200 : Loke AD+6,0
End Proc
Procedure _ASSIGNSTDOUT[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  STIN=Doscall(-60)
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)<>0 Then Error 96
  Loke AD,STIN : Doke AD+4,$100 : Loke AD+6,0
End Proc
Procedure _REMSTD[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)=0 Then Error 97
  Loke AD,0 : Doke AD+4,0 : Loke AD+6,0
End Proc

'This is the new procedure:

Procedure _WAITFORCHAR[CHAN,TIMOUT]
  If CHAN<1 or CHAN>8 Then Error 23
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)=0 Then Error 97
  Dreg(1)=Leek(AD)
  Dreg(2)=TIMOUT
  RES=Doscall(-$CC)
End Proc[RES]

Note that again you should not break the program... use the ESC key
instead...
(otherwise type Loke Areg(5)+$8BC,0 in the direct mode immediately
after the interruption)

I hope this will help you out a bit...

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Albacete (n.)
  A single surprisingly long hair growing in the middle of nowhere.
(from: "The Deeper Meaning of Liff")

From svcs1.digex.net!amos-request Mon May 22 09:39:12 1995 remote from cristal
Return-Path: <[e-mail]>
Received: from svcs1.digex.net by cristal.mpx.com.au with smtp
    (Smail3.1.28.1 #3) id m0sDKaL-0000MPC; Mon, 22 May 95 09:39 AEST
Received: by svcs1.digex.net id AA05339
  (5.67b8/IDA-1.5 for amos-out); Sun, 21 May 1995 16:09:17 -0400
Received: from www1.access.digex.net by svcs1.digex.net with SMTP id AA05335
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 16:09:16 -0400
Received: from phoenix.inka.de by www1.access.digex.net with SMTP id AA15926
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 16:07:54 -0400
Received: from mips.pfalz.de by phoenix.inka.de with bsmtp
    (Smail3.1.29.1 #13) id m0sDHBr-0004B5C; Sun, 21 May 95 22:01 MET DST
Received: from kiste.pfalz.de by mips.pfalz.de with cbsmtp
    (Smail3.1.28.1 #4) id m0sDFIR-000CnTC; Sun, 21 May 95 20:00 MET DST
X-Zc-Via: [e-mail]
X-Zc-Via: [e-mail]
From: [e-mail] (Chris Hodges)
Subject: Re: rtez_on_myscreen?
Message-Id: <[e-mail]>
Date: Sun, 21 May 95 10:29:40 CET
X-Mailer: MicroDot 1.10 [REGISTERED 000261]
X-Zc-Telefon: V+49-89/8005856
X-Zc-Post: Kennedystrasse 8; D-82178 Puchheim; Germany
Path: kiste.pfalz.de!sixpack.pfalz.de
To: [e-mail]
X-Gateway: ZConnect CL kiste.pfalz.de [Connectline/AmigaOS]
References: <[e-mail]>
Status: RO
X-Status:

[e-mail] (Martijn Wehrens) wrote on 19.05.1995 some
text under the subject rtez_on_myscreen?. I can't leave this
uncommentated ;-)

MW> The problem is: how do I set the 'screen'-tag in the taglist or what
MW> ever it is called in amos ? I studied the c-source 'demo.c'-file in the
MW> distribution but I have no idea where I must 'loke' my screen-base in
MW> order to have RT recognize it.

OpenScreen is obsolete... OpenScreenTagList (V36+) is the one you're
longing for...

This a working solution and requires 2.0+:

' OpenIntScreen by Chris Hodges (21-May-95)
Screen Open 0,320,256,32,0
Curs Off : Flash Off : Paper 0 : Pen 1 : Cls
' Assign TAG list
Gosub INITSCREENTAGS
' Reserve memory for a colormap structure
Reserve As Work 14,33*8
ST=Start(14)
For A=0 To 31
  C=Colour(A)
  Doke ST,A
  Doke ST+2,(C and $F00)/256
  Doke ST+4,(C and $F0)/16
  Doke ST+6,C and $F
  Add ST,8
Next
Doke ST,-1
' Create Taglist... add tags at your choice...
TAGSTART
TAG[SA_QUIET,-1]
TAG[SA_WIDTH,320]
TAG[SA_HEIGHT,256]
TAG[SA_DEPTH,5]
TAG[SA_TYPE,$F] : Rem CUSTOMSCREEN, $2=PUBSCREEN
TAG[SA_BITMAP,Leek(Screen Base+$150)] : Rem AMOS Bitmap structure
TAG[SA_COLORS,Start(14)]
TAG[0,0]
' Now open the screen...
Amos To Back
Areg(0)=0 : Areg(1)=Start(15)
SCR=Intcall(-$264) : Rem OpenScreenTagList
If SCR<>0
  ' Draw some thingies on the screen...
  Gr Writing 0
  For A=1 To 31
    Ink A : Text 31-A,100-A,"This is AMOS calling intuition!"
    Circle 200,170,A
  Next
  ' Provisory "wait left mouse button" routine... no not use
  ' in your programs, write system conform replacements please...
  Repeat : Multi Wait : Until(Peek($BFE001) and 64)=0
  ' Close screen (or wait until all windows have been closed)
  Repeat
    Areg(0)=SCR : RES=Intcall(-$42) : Multi Wait
  Until RES=1
End If
Amos To Front
Erase 15
Erase 14
End
' Screen Tags... refer to intuition/screens.i includes for detailed
' information on the tags...
INITSCREENTAGS:
  TS=$80000000+33
  SA_LEFT=TS
  SA_TOP=TS+1
  SA_WIDTH=TS+2
  SA_HEIGHT=TS+3
  SA_DEPTH=TS+4
  SA_DETAILPEN=TS+5
  SA_BLOCKPEN=TS+6
  SA_TITLE=TS+7
  SA_COLORS=TS+8
  SA_ERRORCODE=TS+9
  SA_FONT=TS+10
  SA_SYSFONT=TS+11
  SA_TYPE=TS+12
  SA_BITMAP=TS+13
  SA_PUBNAME=TS+14
  SA_PUBSIG=TS+15
  SA_PUBTASK=TS+16
  SA_DISPLAYID=TS+17
  SA_DCLIP=TS+18
  SA_OVERSCAN=TS+19
  SA_OBSOLETE1=TS+20
' *BOOLEANS*
  SA_SHOWTITLE=TS+21
  SA_BEHIND=TS+22
  SA_QUIET=TS+23
  SA_AUTOSCROLL=TS+24
  SA_PENS=TS+25
  SA_FULLPALETTE=TS+26
  SA_COLORMAPENTRIES=TS+27
  SA_PARENT=TS+28
  SA_DRAGGABLE=TS+29
  SA_EXCLUSIVE=TS+30
  SA_SHAREPENS=TS+31
  SA_BACKFILL=TS+32
  SA_INTERLEAVED=TS+33
  SA_COLORS32=TS+34
  SA_VIDEOCONTROL=TS+35
  SA_FRONTCHILD=TS+36
  SA_BACKCHILD=TS+37
  SA_LIKEWORKBENCH=TS+38
  SA_RESERVED=TS+39
  SA_MINIMIZEISG=TS+40
Return
Procedure TAGSTART
  Shared TAGST
  Reserve As Work 15,128
  TAGST=Start(15)
End Proc
Procedure TAG[TAG,V]
  Shared TAGST
  Loke TAGST,TAG : Loke TAGST+4,V : Add TAGST,8
End Proc

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Alltami (n.)
  The ancient art of being able to balance the hot and cold shower
  taps.
(from: "The Deeper Meaning of Liff")

From svcs1.digex.net!amos-request Mon May 22 12:42:45 1995 remote from cristal
Return-Path: <[e-mail]>
Received: from svcs1.digex.net by cristal.mpx.com.au with smtp
    (Smail3.1.28.1 #3) id m0sDNRL-0000lAC; Mon, 22 May 95 12:41 AEST
Received: by svcs1.digex.net id AA10882
  (5.67b8/IDA-1.5 for amos-out); Sun, 21 May 1995 18:22:23 -0400
Received: from mail1.access.digex.net by svcs1.digex.net with SMTP id AA10876
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 18:22:21 -0400
Received: from rhea.otol.fi by mail1.access.digex.net with SMTP id AA00213
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 18:20:50 -0400
Received: from janus.otol.fi ([e-mail] [193.64.226.249]) by rhea.otol.fi (8.6.12/8.6.9) with ESMTP id BAA07258; Mon, 22 May 1995 01:22:13 +0300
Received: (from tumu@localhost) by janus.otol.fi (8.6.9/8.6.9) id BAA02571; Mon, 22 May 1995 01:22:11 +0300
Date: Mon, 22 May 1995 01:22:10 +0300 (EET DST)
From: Tuomas Lukinmaa <[e-mail]>
To: Martijn Wehrens <[e-mail]>
Cc: Amos-Mailing-List <[e-mail]>
Subject: Re: rtez_on_myscreen?
In-Reply-To: <[e-mail]>
Message-Id: <[e-mail]>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: RO
X-Status:

On Fri, 19 May 1995, Martijn Wehrens wrote:

> Hi all,
>
> The other day, Tumu / BFB team <[e-mail]
> posted a message containing the code needed to make use of the
> reqtools requesters rtezrequest and rtfilrequest. They used a function
>
> _RTFILENAMEPTR=Ma Malloc(108,Mem Public+Mem Clear)
>
> in their code. I suppose this is to do a 'allocmem' from exec and this
> can be done without the extension they propose. The easiest way is
> to make use of :
>
> Reserve as work 6,108
> Poke$ Start(6),String$(chr$(0),108)
> _RTFILENAMEPRT=Start(6)
>
> Further, they use the Elmem$ functions to read out the text, this
> can also be done with:
>
> _RTFILE$=Peek$(_RTFILENAMEPRT,108,chr$(0))
>
> which leads to the exact same result. However, there may be a reason
> to do this as they did, if there is: Please post a message explaining
> why!

You can use both ways to reserve the reqbuffer and rtfilenamebuffer but I
like using Easylife and Makes lib commands because they are easier to
use and Makes lib memory allocation routines are System friendly.

And I forgotted to mention that you'll need Easylife extension as well if
you are going to use original Reqtools routines without modifications.

    Tumu / BFB Team


F
X-Status:

[e-mail] (Mike Pelletier) wrote on 18.05.1995 some
text under the subject Get. I can't leave this uncommentated ;-)

MP> I need something that will check a file for a single character and GO ON if
MP> there is nothing there.

Note that this will work only on RAW: Windows, not on CON: as they
only send their characters after a linefeed...

MP> The program is monitoring the STDIN channel and waiting for a key.  I am trying
MP> to make it so that the routine doesn't hang if there is no input from STDIN.  I
MP> need a `Get a single byte, go on if nothing there' instruction.

I added a procedure to my STDINOUT assigning program... note that it
will take the STDOUT for text input and a raw console for keyboard
input...

Amos To Back
_ASSIGNSTDOUT[1]
'_ASSIGNSTDIN[2]
Open In 2,"RAW:/200//100/RAW-Console/AUTO/WAIT/CLOSE"
Print #1,"Please enter a string: ";
T$=""
Do
  ' The syntax for _WAITFORCHAR is channel,timeout
  _WAITFORCHAR[2,50000]
  BOOL=Param
  If Not BOOL
    Print #1,".";
  Else
    A$=Input$(2,1)
    Exit If A$=Chr$(27)
    Print #1,Chr$(10);"You pressed a key: ";A$;Chr$(10);
    T$=T$+A$
  End If
  If Inkey$=Chr$(27) Then Exit
Loop
_REMSTD[1]
'_REMSTD[2]
Close 2
Amos To Front
Print "You typed: ";T$
End
Procedure _ASSIGNSTDIN[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  STIN=Doscall(-54)
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)<>0 Then Error 96
  Loke AD,STIN : Doke AD+4,$200 : Loke AD+6,0
End Proc
Procedure _ASSIGNSTDOUT[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  STIN=Doscall(-60)
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)<>0 Then Error 96
  Loke AD,STIN : Doke AD+4,$100 : Loke AD+6,0
End Proc
Procedure _REMSTD[CHAN]
  If CHAN<1 or CHAN>8 Then Error 23
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)=0 Then Error 97
  Loke AD,0 : Doke AD+4,0 : Loke AD+6,0
End Proc

'This is the new procedure:

Procedure _WAITFORCHAR[CHAN,TIMOUT]
  If CHAN<1 or CHAN>8 Then Error 23
  AD=Areg(5)+$8BC+(CHAN-1)*10
  If Leek(AD)=0 Then Error 97
  Dreg(1)=Leek(AD)
  Dreg(2)=TIMOUT
  RES=Doscall(-$CC)
End Proc[RES]

Note that again you should not break the program... use the ESC key
instead...
(otherwise type Loke Areg(5)+$8BC,0 in the direct mode immediately
after the interruption)

I hope this will help you out a bit...

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Albacete (n.)
  A single surprisingly long hair growing in the middle of nowhere.
(from: "The Deeper Meaning of Liff")

From svcs1.digex.net!amos-request Mon May 22 09:39:12 1995 remote from cristal
Return-Path: <[e-mail]>
Received: from svcs1.digex.net by cristal.mpx.com.au with smtp
    (Smail3.1.28.1 #3) id m0sDKaL-0000MPC; Mon, 22 May 95 09:39 AEST
Received: by svcs1.digex.net id AA05339
  (5.67b8/IDA-1.5 for amos-out); Sun, 21 May 1995 16:09:17 -0400
Received: from www1.access.digex.net by svcs1.digex.net with SMTP id AA05335
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 16:09:16 -0400
Received: from phoenix.inka.de by www1.access.digex.net with SMTP id AA15926
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 16:07:54 -0400
Received: from mips.pfalz.de by phoenix.inka.de with bsmtp
    (Smail3.1.29.1 #13) id m0sDHBr-0004B5C; Sun, 21 May 95 22:01 MET DST
Received: from kiste.pfalz.de by mips.pfalz.de with cbsmtp
    (Smail3.1.28.1 #4) id m0sDFIR-000CnTC; Sun, 21 May 95 20:00 MET DST
X-Zc-Via: [e-mail]
X-Zc-Via: [e-mail]
From: [e-mail] (Chris Hodges)
Subject: Re: rtez_on_myscreen?
Message-Id: <[e-mail]>
Date: Sun, 21 May 95 10:29:40 CET
X-Mailer: MicroDot 1.10 [REGISTERED 000261]
X-Zc-Telefon: V+49-89/8005856
X-Zc-Post: Kennedystrasse 8; D-82178 Puchheim; Germany
Path: kiste.pfalz.de!sixpack.pfalz.de
To: [e-mail]
X-Gateway: ZConnect CL kiste.pfalz.de [Connectline/AmigaOS]
References: <[e-mail]>
Status: RO
X-Status:

[e-mail] (Martijn Wehrens) wrote on 19.05.1995 some
text under the subject rtez_on_myscreen?. I can't leave this
uncommentated ;-)

MW> The problem is: how do I set the 'screen'-tag in the taglist or what
MW> ever it is called in amos ? I studied the c-source 'demo.c'-file in the
MW> distribution but I have no idea where I must 'loke' my screen-base in
MW> order to have RT recognize it.

OpenScreen is obsolete... OpenScreenTagList (V36+) is the one you're
longing for...

This a working solution and requires 2.0+:

' OpenIntScreen by Chris Hodges (21-May-95)
Screen Open 0,320,256,32,0
Curs Off : Flash Off : Paper 0 : Pen 1 : Cls
' Assign TAG list
Gosub INITSCREENTAGS
' Reserve memory for a colormap structure
Reserve As Work 14,33*8
ST=Start(14)
For A=0 To 31
  C=Colour(A)
  Doke ST,A
  Doke ST+2,(C and $F00)/256
  Doke ST+4,(C and $F0)/16
  Doke ST+6,C and $F
  Add ST,8
Next
Doke ST,-1
' Create Taglist... add tags at your choice...
TAGSTART
TAG[SA_QUIET,-1]
TAG[SA_WIDTH,320]
TAG[SA_HEIGHT,256]
TAG[SA_DEPTH,5]
TAG[SA_TYPE,$F] : Rem CUSTOMSCREEN, $2=PUBSCREEN
TAG[SA_BITMAP,Leek(Screen Base+$150)] : Rem AMOS Bitmap structure
TAG[SA_COLORS,Start(14)]
TAG[0,0]
' Now open the screen...
Amos To Back
Areg(0)=0 : Areg(1)=Start(15)
SCR=Intcall(-$264) : Rem OpenScreenTagList
If SCR<>0
  ' Draw some thingies on the screen...
  Gr Writing 0
  For A=1 To 31
    Ink A : Text 31-A,100-A,"This is AMOS calling intuition!"
    Circle 200,170,A
  Next
  ' Provisory "wait left mouse button" routine... no not use
  ' in your programs, write system conform replacements please...
  Repeat : Multi Wait : Until(Peek($BFE001) and 64)=0
  ' Close screen (or wait until all windows have been closed)
  Repeat
    Areg(0)=SCR : RES=Intcall(-$42) : Multi Wait
  Until RES=1
End If
Amos To Front
Erase 15
Erase 14
End
' Screen Tags... refer to intuition/screens.i includes for detailed
' information on the tags...
INITSCREENTAGS:
  TS=$80000000+33
  SA_LEFT=TS
  SA_TOP=TS+1
  SA_WIDTH=TS+2
  SA_HEIGHT=TS+3
  SA_DEPTH=TS+4
  SA_DETAILPEN=TS+5
  SA_BLOCKPEN=TS+6
  SA_TITLE=TS+7
  SA_COLORS=TS+8
  SA_ERRORCODE=TS+9
  SA_FONT=TS+10
  SA_SYSFONT=TS+11
  SA_TYPE=TS+12
  SA_BITMAP=TS+13
  SA_PUBNAME=TS+14
  SA_PUBSIG=TS+15
  SA_PUBTASK=TS+16
  SA_DISPLAYID=TS+17
  SA_DCLIP=TS+18
  SA_OVERSCAN=TS+19
  SA_OBSOLETE1=TS+20
' *BOOLEANS*
  SA_SHOWTITLE=TS+21
  SA_BEHIND=TS+22
  SA_QUIET=TS+23
  SA_AUTOSCROLL=TS+24
  SA_PENS=TS+25
  SA_FULLPALETTE=TS+26
  SA_COLORMAPENTRIES=TS+27
  SA_PARENT=TS+28
  SA_DRAGGABLE=TS+29
  SA_EXCLUSIVE=TS+30
  SA_SHAREPENS=TS+31
  SA_BACKFILL=TS+32
  SA_INTERLEAVED=TS+33
  SA_COLORS32=TS+34
  SA_VIDEOCONTROL=TS+35
  SA_FRONTCHILD=TS+36
  SA_BACKCHILD=TS+37
  SA_LIKEWORKBENCH=TS+38
  SA_RESERVED=TS+39
  SA_MINIMIZEISG=TS+40
Return
Procedure TAGSTART
  Shared TAGST
  Reserve As Work 15,128
  TAGST=Start(15)
End Proc
Procedure TAG[TAG,V]
  Shared TAGST
  Loke TAGST,TAG : Loke TAGST+4,V : Add TAGST,8
End Proc

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Alltami (n.)
  The ancient art of being able to balance the hot and cold shower
  taps.
(from: "The Deeper Meaning of Liff")

From svcs1.digex.net!amos-request Mon May 22 12:42:45 1995 remote from cristal
Return-Path: <[e-mail]>
Received: from svcs1.digex.net by cristal.mpx.com.au with smtp
    (Smail3.1.28.1 #3) id m0sDNRL-0000lAC; Mon, 22 May 95 12:41 AEST
Received: by svcs1.digex.net id AA10882
  (5.67b8/IDA-1.5 for amos-out); Sun, 21 May 1995 18:22:23 -0400
Received: from mail1.access.digex.net by svcs1.digex.net with SMTP id AA10876
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 18:22:21 -0400
Received: from rhea.otol.fi by mail1.access.digex.net with SMTP id AA00213
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 18:20:50 -0400
Received: from janus.otol.fi ([e-mail] [193.64.226.249]) by rhea.otol.fi (8.6.12/8.6.9) with ESMTP id BAA07258; Mon, 22 May 1995 01:22:13 +0300
Received: (from tumu@localhost) by janus.otol.fi (8.6.9/8.6.9) id BAA02571; Mon, 22 May 1995 01:22:11 +0300
Date: Mon, 22 May 1995 01:22:10 +0300 (EET DST)
From: Tuomas Lukinmaa <[e-mail]>
To: Martijn Wehrens <[e-mail]>
Cc: Amos-Mailing-List <[e-mail]>
Subject: Re: rtez_on_myscreen?
In-Reply-To: <[e-mail]>
Message-Id: <[e-mail]>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: RO
X-Status:

On Fri, 19 May 1995, Martijn Wehrens wrote:

> Hi all,
>
> The other day, Tumu / BFB team <[e-mail]
> posted a message containing the code needed to make use of the
> reqtools requesters rtezrequest and rtfilrequest. They used a function
>
> _RTFILENAMEPTR=Ma Malloc(108,Mem Public+Mem Clear)
>
> in their code. I suppose this is to do a 'allocmem' from exec and this
> can be done without the extension they propose. The easiest way is
> to make use of :
>
> Reserve as work 6,108
> Poke$ Start(6),String$(chr$(0),108)
> _RTFILENAMEPRT=Start(6)
>
> Further, they use the Elmem$ functions to read out the text, this
> can also be done with:
>
> _RTFILE$=Peek$(_RTFILENAMEPRT,108,chr$(0))
>
> which leads to the exact same result. However, there may be a reason
> to do this as they did, if there is: Please post a message explaining
> why!

You can use both ways to reserve the reqbuffer and rtfilenamebuffer but I
like using Easylife and Makes lib commands because they are easier to
use and Makes lib memory allocation routines are System friendly.

And I forgotted to mention that you'll need Easylife extension as well if
you are going to use original Reqtools routines without modifications.

    Tumu / BFB Team


F
X-Status:

[e-mail] (Martijn Wehrens) wrote on 19.05.1995 some
text under the subject rtez_on_myscreen?. I can't leave this
uncommentated ;-)

MW> The problem is: how do I set the 'screen'-tag in the taglist or what
MW> ever it is called in amos ? I studied the c-source 'demo.c'-file in the
MW> distribution but I have no idea where I must 'loke' my screen-base in
MW> order to have RT recognize it.

OpenScreen is obsolete... OpenScreenTagList (V36+) is the one you're
longing for...

This a working solution and requires 2.0+:

' OpenIntScreen by Chris Hodges (21-May-95)
Screen Open 0,320,256,32,0
Curs Off : Flash Off : Paper 0 : Pen 1 : Cls
' Assign TAG list
Gosub INITSCREENTAGS
' Reserve memory for a colormap structure
Reserve As Work 14,33*8
ST=Start(14)
For A=0 To 31
  C=Colour(A)
  Doke ST,A
  Doke ST+2,(C and $F00)/256
  Doke ST+4,(C and $F0)/16
  Doke ST+6,C and $F
  Add ST,8
Next
Doke ST,-1
' Create Taglist... add tags at your choice...
TAGSTART
TAG[SA_QUIET,-1]
TAG[SA_WIDTH,320]
TAG[SA_HEIGHT,256]
TAG[SA_DEPTH,5]
TAG[SA_TYPE,$F] : Rem CUSTOMSCREEN, $2=PUBSCREEN
TAG[SA_BITMAP,Leek(Screen Base+$150)] : Rem AMOS Bitmap structure
TAG[SA_COLORS,Start(14)]
TAG[0,0]
' Now open the screen...
Amos To Back
Areg(0)=0 : Areg(1)=Start(15)
SCR=Intcall(-$264) : Rem OpenScreenTagList
If SCR<>0
  ' Draw some thingies on the screen...
  Gr Writing 0
  For A=1 To 31
    Ink A : Text 31-A,100-A,"This is AMOS calling intuition!"
    Circle 200,170,A
  Next
  ' Provisory "wait left mouse button" routine... no not use
  ' in your programs, write system conform replacements please...
  Repeat : Multi Wait : Until(Peek($BFE001) and 64)=0
  ' Close screen (or wait until all windows have been closed)
  Repeat
    Areg(0)=SCR : RES=Intcall(-$42) : Multi Wait
  Until RES=1
End If
Amos To Front
Erase 15
Erase 14
End
' Screen Tags... refer to intuition/screens.i includes for detailed
' information on the tags...
INITSCREENTAGS:
  TS=$80000000+33
  SA_LEFT=TS
  SA_TOP=TS+1
  SA_WIDTH=TS+2
  SA_HEIGHT=TS+3
  SA_DEPTH=TS+4
  SA_DETAILPEN=TS+5
  SA_BLOCKPEN=TS+6
  SA_TITLE=TS+7
  SA_COLORS=TS+8
  SA_ERRORCODE=TS+9
  SA_FONT=TS+10
  SA_SYSFONT=TS+11
  SA_TYPE=TS+12
  SA_BITMAP=TS+13
  SA_PUBNAME=TS+14
  SA_PUBSIG=TS+15
  SA_PUBTASK=TS+16
  SA_DISPLAYID=TS+17
  SA_DCLIP=TS+18
  SA_OVERSCAN=TS+19
  SA_OBSOLETE1=TS+20
' *BOOLEANS*
  SA_SHOWTITLE=TS+21
  SA_BEHIND=TS+22
  SA_QUIET=TS+23
  SA_AUTOSCROLL=TS+24
  SA_PENS=TS+25
  SA_FULLPALETTE=TS+26
  SA_COLORMAPENTRIES=TS+27
  SA_PARENT=TS+28
  SA_DRAGGABLE=TS+29
  SA_EXCLUSIVE=TS+30
  SA_SHAREPENS=TS+31
  SA_BACKFILL=TS+32
  SA_INTERLEAVED=TS+33
  SA_COLORS32=TS+34
  SA_VIDEOCONTROL=TS+35
  SA_FRONTCHILD=TS+36
  SA_BACKCHILD=TS+37
  SA_LIKEWORKBENCH=TS+38
  SA_RESERVED=TS+39
  SA_MINIMIZEISG=TS+40
Return
Procedure TAGSTART
  Shared TAGST
  Reserve As Work 15,128
  TAGST=Start(15)
End Proc
Procedure TAG[TAG,V]
  Shared TAGST
  Loke TAGST,TAG : Loke TAGST+4,V : Add TAGST,8
End Proc

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Alltami (n.)
  The ancient art of being able to balance the hot and cold shower
  taps.
(from: "The Deeper Meaning of Liff")

From svcs1.digex.net!amos-request Mon May 22 12:42:45 1995 remote from cristal
Return-Path: <[e-mail]>
Received: from svcs1.digex.net by cristal.mpx.com.au with smtp
    (Smail3.1.28.1 #3) id m0sDNRL-0000lAC; Mon, 22 May 95 12:41 AEST
Received: by svcs1.digex.net id AA10882
  (5.67b8/IDA-1.5 for amos-out); Sun, 21 May 1995 18:22:23 -0400
Received: from mail1.access.digex.net by svcs1.digex.net with SMTP id AA10876
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 18:22:21 -0400
Received: from rhea.otol.fi by mail1.access.digex.net with SMTP id AA00213
  (5.67b8/IDA-1.5 for <[e-mail]>); Sun, 21 May 1995 18:20:50 -0400
Received: from janus.otol.fi ([e-mail] [193.64.226.249]) by rhea.otol.fi (8.6.12/8.6.9) with ESMTP id BAA07258; Mon, 22 May 1995 01:22:13 +0300
Received: (from tumu@localhost) by janus.otol.fi (8.6.9/8.6.9) id BAA02571; Mon, 22 May 1995 01:22:11 +0300
Date: Mon, 22 May 1995 01:22:10 +0300 (EET DST)
From: Tuomas Lukinmaa <[e-mail]>
To: Martijn Wehrens <[e-mail]>
Cc: Amos-Mailing-List <[e-mail]>
Subject: Re: rtez_on_myscreen?
In-Reply-To: <[e-mail]>
Message-Id: <[e-mail]>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: RO
X-Status:

On Fri, 19 May 1995, Martijn Wehrens wrote:

> Hi all,
>
> The other day, Tumu / BFB team <[e-mail]
> posted a message containing the code needed to make use of the
> reqtools requesters rtezrequest and rtfilrequest. They used a function
>
> _RTFILENAMEPTR=Ma Malloc(108,Mem Public+Mem Clear)
>
> in their code. I suppose this is to do a 'allocmem' from exec and this
> can be done without the extension they propose. The easiest way is
> to make use of :
>
> Reserve as work 6,108
> Poke$ Start(6),String$(chr$(0),108)
> _RTFILENAMEPRT=Start(6)
>
> Further, they use the Elmem$ functions to read out the text, this
> can also be done with:
>
> _RTFILE$=Peek$(_RTFILENAMEPRT,108,chr$(0))
>
> which leads to the exact same result. However, there may be a reason
> to do this as they did, if there is: Please post a message explaining
> why!

You can use both ways to reserve the reqbuffer and rtfilenamebuffer but I
like using Easylife and Makes lib commands because they are easier to
use and Makes lib memory allocation routines are System friendly.

And I forgotted to mention that you'll need Easylife extension as well if
you are going to use original Reqtools routines without modifications.

    Tumu / BFB Team


F
X-Status:

On Fri, 19 May 1995, Martijn Wehrens wrote:

> Hi all,
>
> The other day, Tumu / BFB team <[e-mail]
> posted a message containing the code needed to make use of the
> reqtools requesters rtezrequest and rtfilrequest. They used a function
>
> _RTFILENAMEPTR=Ma Malloc(108,Mem Public+Mem Clear)
>
> in their code. I suppose this is to do a 'allocmem' from exec and this
> can be done without the extension they propose. The easiest way is
> to make use of :
>
> Reserve as work 6,108
> Poke$ Start(6),String$(chr$(0),108)
> _RTFILENAMEPRT=Start(6)
>
> Further, they use the Elmem$ functions to read out the text, this
> can also be done with:
>
> _RTFILE$=Peek$(_RTFILENAMEPRT,108,chr$(0))
>
> which leads to the exact same result. However, there may be a reason
> to do this as they did, if there is: Please post a message explaining
> why!

You can use both ways to reserve the reqbuffer and rtfilenamebuffer but I
like using Easylife and Makes lib commands because they are easier to
use and Makes lib memory allocation routines are System friendly.

And I forgotted to mention that you'll need Easylife extension as well if
you are going to use original Reqtools routines without modifications.

    Tumu / BFB Team


F
X-Status:

[e-mail] (Eddie) wrote on 18.05.1995 some text under the
subject Trivial. I can't leave this uncommentated ;-)

Ed>     Thats right Eddie!!!! :-)

Ahhhh... %-)))))))))) Normally, one would enter the nickname as
username and the realname in brakets, but this way it works, too ;-)

Ed>     The AMOS scroll routine sucks!  I need to scroll an area of screen which
Ed> is about 154x154 pixels.  This needs to be done STRICTLY within these limits.

Where's the problem?

Either use Screen Copy or the Scroll routines in conjunction with the
Clip-Command, which will perfectly stay inside your specified
limits...

However, I cannot imagine how to get such weird dimensions as
154x154... :-)

Look at this example program... it contains *one* Screen Copy command
for *all* directions...

Screen Open 0,320,256,4,0
Curs Off : Flash Off : Paper 0 : Pen 1
Palette 0,$FFF,$F00,$F0
Set Pattern 22
Ink 1,0 : Bar 0,0 To 320,256
X=100 : Y=20
Ink 6 : Box X-1,Y-1 To X+154,Y+154
Do
  Wait Vbl
  DX=Jleft(1)-Jright(1)
  DY=Jup(1)-Jdown(1)
  Screen Copy 0,X+Max(0,DX),Y+Max(0,DY),X+154+Min(DX,0),Y+154+Min(DY,0) To 0,X-Min(DX,0),Y-Min(DY,0)
Loop

Ed> I will try and incorporate them if I can.  I do need some fast scrolling
Ed> routines though.  HELP!!!!!!!

What kind of RPG is it? I hope you use maps&tiles and so on...

Ed>     They may have been a good side to the C= liquidation.  Did you notice the
Ed> frequency of all the new stuff uploaded to Aminet increased since that
Ed> infamous start of procedings! :-)  Every cloud has a silver lining.

I've been really shocked by this huge amount of software coming on
aminet recently ;-))))

Ed> Long Live The AMIGA!

Dito.

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Albuquerque (n.)
  The shapeless squiggle which is utterly unlike your normal signa-
  ture, but which is, nevertheless, all you are able to produce when
  asked formally to identify yourself.
  Muslims, whose religion forbids the making of graven images, use
  albuquerques to decorate there towels, menu cards and pyjamas.
(from: "The Deeper Meaning of Liff")

F
X-Status:

[e-mail] wrote on 18.05.1995 some text under the
subject Reading Directories. I can't leave this uncommentated ;-)

S0> Hi everyone, the other day somebody sent to the list a routine that reads
S0> in a directory from a disk, what I would like to know, is can this routine
S0> be modified to get the size of the files in the directory as well? If somebody
S0> could help me with this I would appreciate it.

I don't know who has posted this routine, but if it reads the
FileInfoBlock, you can read out any information you want as well...

These are the offsets of the FIB:

Fib_DiskKey     equ     0
Fib_DirEntryTyp equ     4
Fib_FileName    equ     8
Fib_Protection  equ     116
Fib_EntryType   equ     120
Fib_Size        equ     124
Fib_NumBlocks   equ     128
Fib_DateStamp   equ     132
Fib_Comment     equ     144
Fib_End         equ     260

e.g to get the comment of the file use:
=Peek$(FIBSTADR+144,80,Chr$(0))...

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Alcoy (adj.)
  Wanting to be bullied into having another drink.
(from: "The Deeper Meaning of Liff")

F
X-Status:

[e-mail] (Paul Hickman) wrote on 19.05.1995 some
text under the subject Re: Reading Directories. I can't leave this
uncommentated ;-)

PH> But why are you doing this - I know AMOS is full of bugs but I've
PH> never had trouble with the DirFirst$() / DirNext$ functions!

Every tried to get the a filename with spaces in the middle or at the
end?

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Aldclune (n.)
  One who collects ten-year-old telephone directories.
(from: "The Deeper Meaning of Liff")

F
X-Status:

[e-mail] (Stuart Matthews) wrote on 19.05.1995 some
text under the subject Flushing memory in AmosPro. I can't leave this
uncommentated ;-)

SM> Does anyone know how to do the equivalent of an "avail flush" in AmosPro?

Trap Reserve as Work 999,16000000

should do it...

SM> I have a game that plays music using Paul Reece's Enhanced Music
SM> Extension v4.64.  I find that if I play the game after I've been using
SM> my Amiga for a while, the music causes random bits of garbled graphics
SM> to show up on my screen

Euh? Should be something more serious, I think...

SM> obviously these are unused bits of data lurking in memory.

Such unused bits should never be accessed by any program...

IMHO it's a wrong screen display/screen offset setting which causes
this crap graphics... where do there errors appear? Only in the
bottom of the screen or right in the current screen graphics?

Bye, Chris Hodges <[e-mail]>      __    __
A4000/400/CD/'30er; AMOS Extension-Coder    __  ///   / / _____
GCS d H s-:++ !g p? !au a18 w++ v? C+++     \\\///   / /_/ ___/ LOGOUT
E---- N++ K- W------ -po+ t++@ !5 j-- R+ G?  \XX/    \__/ __/  FASCISM!
tv- b+ D-- B? e+(++)* u++ h! f !r n+ !y+ AMIGA RULEZ!   \/

Ambatolampy (n.)
  The bizarre assortment of objects collected by a sleepwalker.
(from: "The Deeper Meaning of Liff")

F
X-Status:

>> The other day, Tumu / BFB team <[e-mail]
>> posted a message containing the code needed to make use of the
>> reqtools requesters rtezrequest and rtfilrequest. They used a function
>>
>> _RTFILENAMEPTR=Ma Malloc(108,Mem Public+Mem Clear)
>>
>> in their code. I suppose this is to do a 'allocmem' from exec and this
>> can be done without the extension they propose. The easiest way is
>> to make use of :
>>
>> Reserve as work 6,108
>> Poke$ Start(6),String$(chr$(0),108)
>> _RTFILENAMEPRT=Start(6)
>>
>> Further, they use the Elmem$ functions to read out the text, this
>> can also be done with:
>>
>> _RTFILE$=Peek$(_RTFILENAMEPRT,108,chr$(0))
>>
>> which leads to the exact same result. However, there may be a reason
>> to do this as they did, if there is: Please post a message explaining
>> why!
>
>You can use both ways to reserve the reqbuffer and rtfilenamebuffer but I
>like using Easylife and Makes lib commands because they are easier to
>use and Makes lib memory allocation routines are System friendly.
>
>And I forgotted to mention that you'll need Easylife extension as well if
>you are going to use original Reqtools routines without modifications.

  Alternatively, you could just use the Intuition Extension's built-in
ReqTools support. :)

  --Andy Church ([e-mail])
    WWW: http://www.mbhs.edu/~achurch/
    AMOS Web Site: http://www.mbhs.edu/~achurch/amos/


F
X-Status:

I have now send off the beta version to all who wanted to be beta testers
of the anim2bob grabber. If you asked for a copy and did not receive it,
let me know so I can resend it.

I will release version 1.0 hopefully in a week or 2 when I get some feed
back from the beta testers.
Changes to come in V1.0 already include:

2 seperate versions.
     One for Hard disk users. This will load as an accessory and auto
install. Very easy to use, simply press install.
     The other for floppy users with small memory (ie a standard 500 or
2000). This will be a stand alone program that will save the bobs/sprites
into a file.

2 diferent ways of grabbing the anims.
     The version that was included in the beta version. It finds the
largest bob and grabs them all to that size.
     The other grabs the smallest size possible for each bob.

Option to save to disk as well as letting the object editor grab them.


I will post a uuencoded version of 1.0 in two weeks to the list (35K) if no
one objects ( and I mean no-one!). If some one does object to this let me
know (privately if you like) and I will come up with another way.
The reason for this is that some people pay for their messsages, and if
they don't want it, thay shouldn't pay for it. Fair enough!

Darryl

-- Via DLG Pro v1.0


               #####\             _             /#####
               #( )# |          _( )__         | #( )#
               ##### |         /_    /         | #####
               #" "# |     ___m/I_ //_____     | #" "#
               # O # |____#-x.\ /++m\ /.x-#____| # O #
               #m.m# |   /" \ ///###\\\ / "\   | #m.m#
               #####/    ######/     \######    \#####

                 EX nightclub DJ (currently looking for work)

F
X-Status:

On Mon, 22 May 1995, Larry Mcgahey wrote:

> How can I get the amos list regularly? I don't remember where I down loaded it
> from.
>

aminet:dev/amos/

I am not sure what the files are called, but if you dl/view the INDEX
file first you'll be able to find out yourself.

assign Aminet: ftp://ftp.doc.ic.ac.uk/packages/aminet/
assign Aminet: ftp://wuarchive.wustl.edu/pub/aminet/

:-)

.......................................................................
.       Branko Collin                .           'WHOP                .
.                                    .        whisssssCRAC!           .
.   //  [e-mail]        .        THUMPA THUMPA'          .
. \X/   [e-mail]               .  Meccano      -      Gilette   .
.......................................................................

F
X-Status:

    The way HAM works is that the first pixel of every line defines a
color. The pixel beside it changes only one aspect of that color, either the
red, green or blue aspect. This pattern is repeated across the entire line
of  the graphics screen, each pixel only containing the change in one aspect pf
the color. By getting the pixel color, you are really only getting the value
of the change in one aspect of the previouse color. To calculate the color you
must actually apply the changes made to the first pixel of the line up to and
including the pixel in question. I am not sure of the exact form of the changes
as in which change goes in which order(I think it is Red then Green then Blue)
but I am sure someone could supply you with that information.

                Robert Currie

F
X-Status:

On Tue, 23 May 1995, Darryl Lewis wrote:

> Does any one know how to add a user funtion to the USER pull down menu at
> the top of the editor screen.

You dont need to do this.  The menu items have keyboard shortcuts
anyway.  You can call them by using the Call Editor or Ask Editor
commands.  Look up the docs on accessories.

> I know that you can click on ADD from that menu, but I'd like to do it in
> software.
> I'm trying to make a piece of software self installing, similar to the
> compiler.

What do you mean self-installing?

>
> Can anyone help?
> Darryl

             ----====>>>> From the desk of ... <<<<====----
     _______   _______   _____   __  __
    / ___  /  / ___  /  / __ |  / / / /
   / /__/ /  / /  / /  / / / / / /_/ /   >Live long and prosper ;-)<
  /  __  /  / /  / /  / / / / /__ __/
 / /  / /  / /  / /  / /_/ /   / /   Email : [e-mail]
/_/  /_/  /_/  /_/  /_____/   /_/    Web Page : http://www.staffs.ac.uk


F
X-Status:

On Tue, 23 May 1995, Darryl Lewis wrote:

> Does any one know why the number returned using the point command on a ham
> picture is different from the colour seen?
> For example:
> Load in a Ham picture to a screen.
> Use the POINT command on an area and a colour that you know.
> Cls using the number returned from the point function.
>
> The resulting screen is a different colour thaa what you'd think.
> Can anyone help?

    Well as it says in the Amos Manual, a Ham screen doesn't map the
colors on the screen to the color index.  Only the first 16 colors can be
set from the palette, the other 48 are all preset.  Not sure what order its
in (I don't have the manual with me), but I think its something like
Red,Blue,green - thats is 16-31 are shades of red etc....

    This is cos of the way ham works, it uses the base 16 colors as
normal, but the reds, greens and blues just alter the current pixels color
relative to the previous pixel.

    Say you have black as color 0.  And you plot a point with it.  And
then the next point is 31 (max red) and then the next is whatever max blue
is.  You'd end up with a black pixel, then a red pixel, and then a purple
pixel.  That is where the color fringing you get comes from on really badly
scanned images.

    So to find out the color of a point on a ham screen - you'll have to
look at all the pixels to the left of it, until you find the left hand side
of the screen, or one of the first 16 colors.

Chris
PS Of course that lot is all for the OCS ham mode (4096 colors) *8).

--
[e-mail]   | "What are you a man or a Windoze user?"
[e-mail]  |
--------------+----------+----------------------------------------------------
The Gathering | zippy.spods.dcs.kcl.ac.uk 5000
URL           | http://www.dcs.rhbnc.ac.uk/~christ


F
X-Status:



>     The way HAM works is that the first pixel of every line defines a
> color. The pixel beside it changes only one aspect of that color, either the
> red, green or blue aspect. This pattern is repeated across the entire line
> of  the graphics screen, each pixel only containing the change in one aspect pf
> the color. By getting the pixel color, you are really only getting the value
> of the change in one aspect of the previouse color. To calculate the color you
> must actually apply the changes made to the first pixel of the line up to and
> including the pixel in question. I am not sure of the exact form of the changes
> as in which change goes in which order(I think it is Red then Green then Blue)
> but I am sure someone could supply you with that information.

That's not precise.
It is not the firs pixel on the screen which determine the colour...
otherwise you could not have black and withe on the same line.

The HAM works this way (for HAM6):

The HAM is based on the fact that real good picture have lots of shade
which can be "compressed" in less bits.
You have a base palette of 16 colors. Now, the screen is done by 6 planes
(thta's why HAM6!).
The color of a pixel is described as follows:
the first 4 bit of the pixel (the first 4 bit-planes) tell you which color
of the palette had to be used as the reference color.
The last 2 bits (last two bitplanes) carry these informations:
If both bits are 00 the color is used pure that's you see the same color
which is stored in the palette base.
If I remember right the following value refer to RGB shades as follows...
If bit 1 is set to 1 you have a shade in the RED value of the color.
If bit 2 is set to 1 you have a shade in the GREEN value of the color
IF bit 1 is set to 1 and bit 2 is sey to 1 you have a shade in the BLUE
value of the color.

So a pixel can different from the next on the left only of a single
component of the color or it have to belong to the base palette set.

That's why you can't have smooth horizontal grey shades (but you can have
smooth horizontal red, blue, green shades). A grey shade implies you
firstly pass through the RED, the GREEN and the BLUE shade to finally
reach the desired grey shade.
That is there are two pixels on the screen which are not of the desired
color... that's called fringing.

Hope this will help.

MeF

F