[Coco] Color Forth Assistance Request

Stephen Pereira spereira1952 at comcast.net
Sat Jan 17 19:38:13 EST 2015


Thanks very much for this information, Art.

This will give me something to anchor my effort and experiments.

smp
--
Stephen M. Pereira
Bedford, NH  03110
KB1SXE


> On Jan 17, 2015, at 7:30 PM, Arthur Flexser <flexser at fiu.edu> wrote:
> 
> I found a snatch of old code of mine that contains enough information that
> I can give you something with more detail.  It doesn't exactly use the
> method of the previous message, but instead jumps directly into the
> SAVEM command (to $CF7E in Disk Basic 1.1) after storing the appropriate
> information to the file name buffer (8 bytes at $94C) and extension buffer
> (3 bytes at $954).  It appears that if you then jump to $CF7E with the
> registers set as below and pushed onto the stack with a single PSHS
> instruction, everything should work:
> 
> Y = start address - 1
> X = end address
> D = exec address
> U = address in your program where you want to resume execution after
> the       savem happens
> 
> Art
> 
> 
> 
> On Sat, Jan 17, 2015 at 2:02 PM, Arthur Flexser <flexser at fiu.edu> wrote:
> 
>> One approach to using Basic operations from a machine-language program is
>> to set up a buffer containing the command as it would be typed in to Basic,
>> possibly already tokenized (i.e., use a "save" token byte instead of Ascii
>> SAVE), and then point Basic to this buffer and have it execute the
>> command.  There was an article in Rainbow with examples;  maybe someone can
>> provide the reference.  Pointing Basic to the buffer essentially involves
>> putting its address into $00A6, the location of Basic's parse pointer.
>> 
>> Art
>> 
>> On Sat, Jan 17, 2015 at 12:57 PM, Stephen Pereira <
>> spereira1952 at comcast.net> wrote:
>> 
>>> Hello again, all,
>>> 
>>> As you know from my postings about Color Forth, I have been successful
>>> with typing in the source code for the Color Forth nucleus (thanks again,
>>> Brett, for your assistance finding the last few typos!).  With this, the
>>> ORG of the assembly code can be changed to allow Color Forth to be loaded
>>> from DECB (HDBDOS in my case) without overwriting any of the disk
>>> capabilities.  I also have devised a method to pre-load the code for 1 to 8
>>> Forth screens of Forth code that can be subsequently loaded into Color
>>> Forth to add in more of the original capability, or the editor, etc.
>>> 
>>> Now, I would like to add in a simple method for loading and saving a set
>>> of Forth screens from disk.  What I would like to be able to do is perform
>>> LOADM"SCRN1TO8" and SAVEM"SCRN1TO8" &H2380 &H43A0 &H5979 from inside Color
>>> Forth.
>>> 
>>> I have taken a look at "Disk Basic Unravelled," and I found memory
>>> locations like Disk File Name Buffer (DNAMBF) and Disk File Extension Name
>>> Buffer (DEXTBF) which look like places that I could load the file name and
>>> extension.  I also found the routine at LC48D which is commented as "Open
>>> Disk File For Read or Write."
>>> 
>>> However, I have never written my own code to patch into the DECB system
>>> before, so I am skittish, to say the least.
>>> 
>>> What I would like to do is write my own code to load the file name and
>>> extension, the starting and ending addresses for the memory block that I
>>> would like to save, and call a routine to perform a SAVEM command as if it
>>> had come in from the Basic command line.  Similarly, I would like to write
>>> the code to load the file name and extension, and call a routine to perform
>>> a LOADM command as if it had come in from the Basic command line.
>>> 
>>> Can anyone offer me some advice on where to look to pick up on the
>>> specific things that I would have to do in my code to accomplish these
>>> operations?
>>> 
>>> Thanks very much, in advance, for your attention and advice.
>>> 
>>> smp
>>> --
>>> Stephen M. Pereira
>>> Bedford, NH  03110
>>> KB1SXE
>>> 
>>> 
>>> 
>>> --
>>> Coco mailing list
>>> Coco at maltedmedia.com
>>> https://pairlist5.pair.net/mailman/listinfo/coco
>>> 
>> 
>> 
> 
> -- 
> Coco mailing list
> Coco at maltedmedia.com
> https://pairlist5.pair.net/mailman/listinfo/coco



More information about the Coco mailing list