I'm trying to implement a kind of "conditional :" in ATLAST, the reasoning being I have a file that gets FLOADed multiple times to handle multiple steps of my program flow (I'm essentially abusing Forth as an assembler, step 1 does a first parsing for references, etc. and in step 2 the instruction words actually emit bytes).
So when declaring words for "macros" in that file, it errors out in step 2, because they were already declared in step 1, but I also can't just FORGET them, because that would forget everything that came afterwards, such as the references I just collected in step 1.
So essentially I need a ": that only runs in step 1", my idea being something like this:
VARIABLE STAGE
: ::
STAGE # 0 = IF
[COMPILE] : ( be a word declaration )
EXIT
THEN
BEGIN ( eat the disabled declaration )
' ( get the address of the next word )
['] ; ( get the address of semicolon )
= ( loop until they are equal )
UNTIL
; IMMEDIATE
:: FIVE 5 ; ( declares as expected )
FIVE . ( prints 5 )
1 STAGE ! ( up to here everything's fine )
:: FIVE 6 ; ( is supposed to do nothing, but errors out )
FIVE . ( is supposed to print 5 again )
The traced error message (starting from 1 STAGE !):
Trace: !
Trace: ::
Trace: STAGE
Trace: #
Trace: (LIT) 0
Trace: =
Trace: ?BRANCH
Trace: '
Trace: (LIT) 94721509587192
Trace: =
Trace: ?BRANCH
Trace: '
Word not specified when expected.
Trace: ;
Compiler word outside definition.
Walkback:
;
KEY ( -- ch ) as common in some other Forths for reading a single character from the input stream ( outside the :: declaration, since it's IMMEDIATE ) doesn't exist in ATLAST, the only related words I could find are:
': is supposed to read a word from the input stream, then pushes its compile address
[']: like ' but reads a word from the current line (the inside of the :: declaration)
(LIT)/(STRLIT): are supposed to read literals from the input stream according to the documentation, I could only ever make them segmentation fault, I think they're for compiler-internal use only (e.g., if the compiler encounters a number literal it will compile the (LIT) word to make it push that number onto the stack)
There aren't any WORD or PARSE either, as in some other Forths.
As you can see, ' is struggling actually getting something from the input stream for some weird reason, and it looks like ['] is failing to capture the ; which then errors out because it's suddenly encountering a ; where it doesn't belong.
I suspect it actually ran ' ['], even though it's supposed to work on the input stream, not the immediate line, and I'm clearly in compile mode there.
I did a similar thing with conditionally declaring variables, there it was rather easy to just [COMPILE] ' DROP to skip a single word (turning RES x into ' x DROP), but here I'm pretty sure I can't actually compile those instructions, because I can't emit a loop outside of a declaration. Unless there is a way to somehow compile similar code that recursively gets rid of everything until the ;.
A problem is that ' cannot find a number. A possible solution is to use a special dummy name for the definition, instead of skip it over:
: ::
STAGE # 0 = IF : EXIT THEN
' DROP \ this xt isn't needed
" : _dummy" EVALUATE ( -- n ) DROP
;
Or maybe use a new name every time:
: ::
STAGE # 0 = IF : EXIT THEN
' >NAME # \ ( s1 ) \ should be checked
": _dummy_" DUP >R S+
R> EVALUATE ( -- n ) DROP
;
But due to non standard words it might not work. Another problem is that non colon-definitions are out of the scope.
Perhaps, a better solution is a preprocessing by external means.
It appears that ATLAST is a primitive Forth, that doesn't allow you to go to a more sophisticated handling of sources. But all is not lost!
For example, a Forth implementation according to the ISO standard will handle the matter with ease with one or more of: REQUIRE [IF] [THEN] [DEFINED] SRC >IN NAME WORD FIND.
As you have a Forth, you can steal these words from another Forth and compile the code.
Another solution that may help directly is executing EXIT in interpret mode while loading a file.
You have to find out whether you can create a flag whether to abandon the input source. Then this definition might help:
: ?abandon IF S" EXIT" EVALUATE THEN ;
S" FIVE" FOUND ?abandon
Note that ?abandon must be executed in interpret mode.
Related
I'm interesting in testing some of the limits of Gforth and would like to have it execute arbitrary code that I "hand compile" into allocated memory. Here is my attempt.
100 cells allocate throw constant &mem
\ store at &mem: docol: . EXIT
docol: &mem !
comp' . &mem 1 cells + ! drop \ drop "execution token"
comp' EXIT &mem 2 cells + ! drop
42 \ something to print
&mem execute
Unfortunately this fails with:
in file included from *OS command line*:-1
notes/execute.fs:8: Invalid memory address
&mem >>>execute<<<
Backtrace:
$7EFC61175B28 execute
I have to use comp' instead of ', because it doesn't work for getting the xt of EXIT.
I would have thought this should work, unless Gforth doesn't operate in any way like JonesForth did where docol: starts executing the xt's next to it.
Is this possible in either Gforth or ANS forth in general?
You can execute an arbitrary list of xt, but you have to use your own word to execute this list, by applying execute to each xt from the list.
By the current standard, a standard program cannot compile arbitrary code into allocated memory. The program may only compile into the code space of the dictionary, and in the frame of the current definition (i.e., that is not yet completed). Compilation can be performed via compile, ( xt -- ) or postpone ( i*x "name" -- j*x ) words. Also the words literal, 2literal, sliteral, fliteral (or their counterparts lit,, 2lit,, slit,, flit,) can be used to compile literals.
In Gforth you can also compile into another dictionary ("section"), that can be allocated using word extra-section ( size "name" -- ).
10000 extra-section execute-in-my-section
\ execute-in-my-section ( i*x xt -- j*x )
unused cr . \ free space in the default dictionary
[:
unused cr . \ free space in the current section
:noname
postpone .
postpone ;
( xt-new )
unused cr . \ free space after compile the new definition
;] execute-in-my-section ( xt-new )
\ test
123 swap execute
See also section.fs source, and Sections paper by Anton Ertl, 2016.
When reading about the tick (') operator I wondered if it can be useful inside a word definition. I know that there is ['] to be used inside a definition, but I thought about using it to read the word name following invocation.
An example:
4 variable cnt
: cycle: ( arg fn -- )
'
4 cnt !
begin
cr
dup execute
-1 cnt +!
cnt # 0 = until
drop
;
I can use cycle: to repeat some word invocation, as follows.
: hello ." hello" ;
cycle: hello
Which prints hello four times, as expected.
But the following code won't define a word that prints hello four times:
: 4hello cycle: hello ;
The tick operator still expects a word from the input stream following invocation of 4hello.
Is it possible to inject it somehow when using cycle: in a word definition, so it won't "leak" outside?
Yes, it's possible. You would have to make cycle: immediate. And then also change it to postpone its actions, rather than perform them at runtime.
Postponing means to delay the actions of words. Immediate words are compiled into the current definition, and normal words are arranged to be compiled when the current definition is executing.
In this case it might look something like this.
: (cycle) 4 0 do dup execute loop drop ;
: cycle: ' postpone literal postpone (cycle) ; immediate
Note that this version no longer works outside definitions.
I am using the interpreter directives (non ANS standard) control structures of Gforth as described in the manual section 5.13.4 Interpreter Directives. I basically want to use the loop words to create a dynamically sized word containing literals. I came up with this definition for example:
: foo
[ 10 ] [FOR]
1
[NEXT]
;
Yet this produces an Address alignment exception after the [FOR] (yes, I know you should not use a for loop in Forth at all. This is just for an easy example).
In the end it turned out that you have to write loops as one-liners in order to ensure their correct execution. So doing
: foo [ 10 [FOR] ] 1 [ [NEXT] ] ;
instead works as intended. Running see foo yields:
: foo
1 1 1 1 1 1 1 1 1 1 1 ; ok
which is exactly what I want.
Is there a way to get new lines in the word definition? The words I would like to write are way more complex, and for a presentation I would need them better formatted.
It would really be best to use an immediate word instead. For example,
: ones ( n -- ) 0 ?do 1 postpone literal loop ; immediate
: foo ( -- ten ones ) [ 10 ] ones ;
With SEE FOO resulting in the same as your example. With POSTPONE, especially with Gforth's ]] .. [[ syntax, the repeated code can be as elaborate as you like.
A multiline [FOR] would need to do four things:
Use REFILL to read in subsequent lines.
Save the read-in lines, because you'll need to evaluate them one by one to preserve line-expecting parsing behavior (such as from comments: \ ).
Stop reading in lines, and loop, when you match the terminating [NEXT].
Take care to leave >IN right after the [NEXT] so that interpretation can continue normally.
You might still run into issues with some code, like code checking SOURCE-ID.
For an example of using REFILL to parse across multiple lines, here's code from a recent posting from CLF, by Gerry:
: line, ( u1 caddr2 u2 -- u3 )
tuck here swap chars dup allot move +
;
: <text> ( "text" -- caddr u )
here 0
begin
refill
while
bl word count s" </text>" compare
while
0 >in ! source line, bl c, 1+
repeat then
;
This collects everything between <text> and a </text> that's on its own line, as with a HERE document, while also adding spaces. To save the individual lines for [FOR] in an easy way, I'd recommend leaving 0 as a sentinel on the data stack and then drop SAVE-MEM 'd lines on top of it.
(I'm back with yet another question :-) )
Given the following PostScript code:
/riverside { 5 pop } def
/star { 6 pop 2 {riverside} repeat } def
star
I'm wondering how nested procedures should be handled. (I'm creating my own interpreter).
When I execute the star procedure, halfway it finds a nameObjec(riverside) and replaces it with an executable array containing the values from the riverside procedure and executes them.
If I execute the repeat operator the interpreter crashes because there is only one item left on the stack.
Should I actually execute an executable array (=procedure) directly when I'm already in an executable array (=prodecure), or should the executable arrays (=procedures) always be pushed on the (operand?/execution?)stack? or only be executed by another operator?
How many times should this riverside be executed? (2 or 3 times?) I guess 2?
For your information: this is the situation that I have when I execute star on the 3rd line (see the ERROR):
% begin execute 3rd line (star)
% OP = operand stack
% EX = execution stack
% handle 6
OP: 6
EX: star
% handle pop (removes 6 from OP)
OP: -
EX: star
% handle 2
OP: 2
EX: star
% set the riverside executable array on the EX, execute the values
OP: 2
EX: star riverside
% repeat operator:
CRASH, only one item on the OP left, but repeat operator requires 2 operands.
OP: 5
EX:
% end
Please shine a light on this matter, because it is somewhat complex/confusing :-)
Update:
another code sample might be this one:
/starside
{ 72 0 lineto
currentpoint translate
-144 rotate } def
/star
{ moveto
currentpoint translate
4 {starside} repeat
closepath
gsave
.5 setgray fill
grestore
stroke } def
200 200 star
showpage
when the interpreter tokenizes /star { moveto ... if it encounters the nested {starside} how will that be treated? (+ what if there was {starside 5 2 mul pop} instead of only {starside} ?)
I believe you need to look at section 3.5.3 of the PLRM. Although this deals with a simple executable array the concept is the same. When the token scanner encounters a '{' it starts to build an executable array. Until it reaches a matching '}' token the scanner simply stores what it encounters on the operand stack. When it encounters the matching '{' then the objects are converted into an executable array (and stored on the operand stack)
In the case of the scanner encountering an executable name, it stores the name on the operand stack. It does not execute the name, nor does it even perform lookup on it to retrieve the associated object.
So immediately before the execution of '}' in your example, the operand stack would contain twp objects, the '{' opening array, and the executable name riverside. When you encounter the '}' then the scanner creates the actual executable array and stores it on the operand stack. (Note, implementation details vary here)
So immediately before the execution of 'repeat' you would have two objects on the stack, the counter and an executable array containing a single executable name.
You don't look up the name until the executable array containing the name is executed.
This might make it clearer:
%!
/test {(This is my initial string\n) print} def
2 {test} repeat
2 {test} /test {(This is my second string\n) print} def repeat
Notice that I've redefined 'test' after creating the executable array containing the executable name 'test', yet the execution uses the later definition of test. As you can see, its vitally important not to do name lookup too early!
I have many text files of this format
....
<snip>
'FOP' 0.19 1 24 1 25 7 8 /
'FOP' 0.18 1 24 1 25 9 11 /
/
TURX
560231
300244
70029
200250
645257
800191
900333
600334
770291
300335
220287
110262 /
SUBTRACT
'TURX' 'TURY'/
</snip>
......
where the portions I snipped off contain other various data in various formats. The file format is inconsistent (machine generated), the only thing one is assured of is the keyword TURX which may appear more than once. If it appears alone on one line, then the next few lines will contain numbers that I need to fetch into an array. The last number will have a space then a forward slash (/). I can then use this array in other operations afterwards.
How do I "search" or parse a file of unknown format in fortran, and how do I get a loop to fetch the rest of the data, please? I am really new to this and I HAVE to use fortran. Thanks.
Fortran 95 / 2003 have a lot of string and file handling features that make this easier.
For example, this code fragment to process a file of unknown length:
use iso_fortran_env
character (len=100) :: line
integer :: ReadCode
ReadLoop: do
read (75, '(A)', iostat=ReadCode ) line
if ( ReadCode /= 0 ) then
if ( ReadCode == iostat_end ) then
exit ReadLoop
else
write ( *, '( / "Error reading file: ", I0 )' ) ReadCode
stop
end if
end if
! code to process the line ....
end do ReadLoop
Then the "process the line" code can contain several sections depending on a logical variable "Have_TURX". If Have_TRUX is false you are "seeking" ... test whether the line contains "TURX". You could use a plain "==" if TURX is always at the start of the string, or for more generality you could use the intrinsic function "index" to test whether the string "line" contains TURX.
Once the program is in the mode Have_TRUX is true, then you use "internal I/O" to read the numeric value from the string. Since the integers have varying lengths and are left-justified, the easiest way is to use "list-directed I/O": combining these:
read (line, *) integer_variable
Then you could use the intrinsic function "index" again to test whether the string also contains a slash, in which case you change Have_TRUX to false and end reading mode.
If you need to put the numbers into an array, it might be necessary to read the file twice, or to backspace the file, because you will have to allocate the array, and you can't do that until you know the size of the array. Or you could pop the numbers into a linked list, then when you hit the slash allocate the array and fill it from the linked list. Or if there is a known maximum number of values you could use a temporary array, then transfer the numbers to an allocatable output array. This is assuming that you want the output argument of the subroutine be an allocatable array of the correct length, and the it returns one group of numbers per call:
integer, dimension (:), allocatable, intent (out) :: numbers
allocate (numbers (1: HowMany) )
P.S. There is a brief summary of the language features at http://en.wikipedia.org/wiki/Fortran_95_language_features and the gfortran manual has a summary of the intrinsic procedures, from which you can see what built in functions are available for string handling.
I'll give you a nudge in the right direction so that you can finish your project.
Some basics:
Do/While as you'll need some sort of loop
structure to loop through the file
and then over the numbers. There's
no for loop in Fortran, so use this
type.
Read
to read the strings.
To start you need something like this:
program readlines
implicit none
character (len=30) :: rdline
integer,dimension(1000) :: array
! This sets up a character array with 30 positions and an integer array with 1000
!
open(18,file='fileread.txt')
do
read(18,*) rdline
if (trim(rdline).eq.'TURX') exit !loop until the trimmed off portion matches TURX
end do
See this thread for way to turn your strings into integers.
Final edit: Looks like MSB has got most of what I just found out. The iostat argument of the read is the key to it. See this site for a sample program.
Here was my final way around it.
PROGRAM fetchnumbers
implicit none
character (len=50) ::line, numdata
logical ::is_numeric
integer ::I,iost,iost2,counter=0,number
integer, parameter :: long = selected_int_kind(10)
integer, dimension(1000)::numbers !Can the number of numbers be up to 1000?
open(20,file='inputfile.txt') !assuming file is in the same location as program
ReadLoop: do
read(20,*,iostat=iost) line !read data line by line
if (iost .LT. 0) exit !end of file reached before TURX was found
if (len_trim(line)==0) cycle ReadLoop !ignore empty lines
if (index(line, 'TURX').EQ.1) then !prepare to begin capturing
GetNumbers: do
read(20, *,iostat=iost2)numdata !read in the numbers one by one
if (.NOT.is_numeric(numdata)) exit !no more numbers to read
if (iost2 .LT. 0) exit !end of file reached while fetching numbers
read (numdata,*) number !read string value into a number
counter = counter + 1
Storeloop: do I =1,counter
if (I<counter) cycle StoreLoop
numbers(counter)=number !storing data into array
end do StoreLoop
end do GetNumbers
end if
end do ReadLoop
write(*,*) "Numbers are:"
do I=1,counter
write(*,'(I14)') numbers(I)
end do
END PROGRAM fetchnumbers
FUNCTION is_numeric(string)
IMPLICIT NONE
CHARACTER(len=*), INTENT(IN) :: string
LOGICAL :: is_numeric
REAL :: x
INTEGER :: e
is_numeric = .FALSE.
READ(string,*,IOSTAT=e) x
IF (e == 0) is_numeric = .TRUE.
END FUNCTION is_numeric