ref: e5f0c0798b44ffaa03772477138ff0ae4df4763b
dir: /doc/ficl.html/
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="Author" content="john sadler">
<meta name="GENERATOR" content="Mozilla/4.5 [en] (Win95; U) [Netscape]">
<meta name="Description" content="all about the coolest scripting language ever">
<title>ficl 2.03 release notes</title>
</head>
<body>
<center>
<h1>
<b>ficl 2.03 release notes</b></h1></center>
<table BORDER=0 CELLSPACING=3 WIDTH="600" >
<tr>
<td><b>Forth Inspired Command Language </b></td>
<td ROWSPAN="4"><img SRC="ficl_logo.jpg" height=64 width=64></td>
</tr>
<tr>
<td><b>Author: John Sadler (<a href="mailto:john_sadler@alum.mit.edu">john_sadler@alum.mit.edu</a>)</b></td>
</tr>
<tr>
<td><b>Created: 19 July 1997 </b></td>
</tr>
<tr>
<td><b>Revision 2.03: 20 May 1999 </b></td>
</tr>
</table>
<h2>
Contents</h2>
<ul>
<li>
<a href="#whatis">What is ficl?</a></li>
<li>
<a href="#features">Ficl features</a></li>
<li>
<a href="#porting">Porting</a></li>
<li>
<a href="#api">Application Programming Interface</a></li>
<li>
<a href="#manifest">Distribution source files</a></li>
<li>
<a href="#whatsnew">What's new in this release</a></li>
<li>
<a href="#objects">Objects in ficl</a></li>
<ul>
<li>
<a href="ficl.html#oopgloss">OOP glossary</a></li>
<li>
<a href="#glossinstance">Instance variable glossary</a></li>
<li>
<a href="#glossclass">Class methods glossary</a></li>
<li>
<a href="#objectgloss">Object base class methods glossary</a></li>
<li>
<a href="#stockclasses">Supplied Classes</a></li>
</ul>
<li>
<a href="#extras">Ficl extras</a></li>
<li>
<a href="#ansinfo">ANS required information</a></li>
<li>
<a href="#links">Forth and Ficl references, <b><font color="#000000">download</font></b></a></li>
<li>
<font color="#000000"><a href="#includesficl">Some software that includes
ficl</a></font></li>
<li>
<a href="#lawyerbait">Disclaimer & License</a></li>
</ul>
<hr WIDTH="100%">
<table CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
<a NAME="whatis"></a>What is ficl?</h2>
Ficl (Forth inspired command language) is an ANS Forth interpreter written
in C. Unlike traditional Forths, this interpreter is designed to be embedded
into other systems as a command/macro/development prototype language. Ficl
provides object extensions that can be used to wrap methods and structures
of the host system without altering them. See below for examples of <a href="#includesficl">software
that includes ficl</a>.</td>
</tr>
<tr>
<td>Where Forths usually view themselves as the center of the system and
expect the rest of the system to be coded in Forth, Ficl acts as a component
of the system. It is easy to export code written in C or ASM to Ficl in
the style of TCL, or to invoke Ficl code from a compiled module. This allows
you to do incremental development in a way that combines the best features
of threaded languages (rapid development, quick code/test/debug cycle,
reasonably fast) with the best features of C (everyone knows it, easier
to support large blocks of code, efficient, type checking). In addition,
Ficl provides a simple object model that can act as an object oriented
adapter for code written in C (or asm, Forth, C++...). </td>
</tr>
<tr>
<td><b>Ficl Design goals</b>
<ul>
<li>
Target 32 bit processors (<i>version 2.03 targets 64 bit processors too</i>)</li>
<li>
Scripting, prototyping, and extension language for systems written also
in C</li>
<li>
Supportable - code is as transparent as I can make it</li>
<li>
Interface to functions written in C</li>
<li>
Conform to the Forth DPANS 94</li>
<li>
Minimize porting effort - require an ANSI C runtime environment and minimal
glue code</li>
<li>
Provide object oriented extensions</li>
</ul>
</td>
</tr>
</table>
<br>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
<a NAME="features"></a>Ficl features</h2>
<ul>
<li>
Code is written in ANSI C for portability. </li>
<li>
Standard: Implements the ANS Forth CORE word set, part of the CORE EXT
word set, SEARCH and SEARCH EXT, TOOLS and part of TOOLS EXT, LOCAL and
LOCAL EXT, EXCEPTION, MEMORY, and various extras.</li>
<li>
Extensible: you can export code written in Forth, C, or asm in a straightforward
way. Ficl provides open facilities for extending the language in an application
specific way. You can even add new control structures (not surprising if
you're familiar with Forth)</li>
<li>
Ficl and C can interact in two ways: Ficl can wrap C code, and C functions
can invoke ficl code.</li>
<li>
Ficl code is thread safe and re-entrant: All Ficl VMs share one system
dictionary; each Ficl virtual machine has an otherwise complete state,
and each can be bound to a separate I/O channel (or none at all). An optional
function called ficlLockDictionary() can control exclusive dictionary access.
This function is stubbed out by default (See FICL_MULTITHREAD in sysdep.h).
As long as there is only one "session" that can compile words into the
dictionary, you do not need exclusive dictionary access for multithreading.
<font color="#000099"><b>Note</b>:
while the code is re-entrant, there are still restrictions on how you can
use it safely in a multithreaded system. Specifically, the VM itself maintains
state, so you generally need a VM per thread in a multithreaded system.
If interrupt service routines make calls into Ficl code that alters VM
state, then these generally need their own VM as well. Alternatively, you
could provide a mutual exclusion mechanism to serialize access to a VM
from multiple threads.</font></li>
<li>
Simple incorporation into existing systems: the sample implementation requires
three Ficl function calls (see the example program in testmain.c).</li>
<li>
ROM able: Ficl is designed to work in RAM based and ROM code / RAM data
environments. It does require somewhat more memory than a pure ROM implementation
because it builds its system dictionary in RAM at startup time.</li>
<li>
Written an ANSI C to be as simple as I can make it to understand, support,
debug, and port. Compiles without complaint at /Az /W4 (require ANSI C,
max. warnings) under Microsoft VC++</li>
<li>
Does full 32 bit math (but you need to implement two mixed precision math
primitives (see sysdep.c))</li>
<li>
Type 1 indirect threaded interpreter</li>
</ul>
</td>
</tr>
</table>
<p>
<hr WIDTH="100%">
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
<a NAME="porting"></a>Porting ficl</h2>
To install ficl on your target system, you need an ANSI C compiler and
its runtime library. Inspect the system dependent macros and functions
in <tt>sysdep.h</tt> and <tt>sysdep.c</tt> and edit them to suit your system.
For example, <tt>INT16</tt> is a <tt>short</tt> on some compilers and an
<tt>int</tt>
on others. Check the default <tt>CELL</tt> alignment controlled by <tt>FICL_ALIGN</tt>.
If necessary, add new definitions of <tt>ficlMalloc, ficlFree, ficlLockDictionary</tt>,
and <tt>ficlTextOut</tt> to work with your operating system. Finally, use
<tt>testmain.c</tt>
as a guide to installing the ficl system and one or more virtual machines
into your code. You do not need to include <tt>testmain.c</tt> in your
build.
<p>Feel free to stub out the double precision math functions (which are
presently implemented as inline assembly because it's so easy on many 32
bit processors) with kludge code that only goes to 32 bit precision. In
most applications, you won't notice the difference. If you're doing a lot
of number crunching, consider implementing them correctly.
<h3>
Build controls</h3>
The file sysdep.h contains default values for build controls. Most of these
are written such that if you define them on the compiler command line,
the defaults are overridden. I suggest you take the defaults on everything
below the "build controls" section until you're confident of your port.
Beware of declaring too small a dictionary, for example. You need about
3200 cells for a full system, about 2000 if you strip out most of the "soft"
words.
<h3>
To-Do List (target system dependent words)</h3>
<ul>
<li>
Unimplemented system dependent <tt>CORE</tt> word: <tt>KEY</tt> </li>
<li>
Kludged <tt>CORE</tt> word: <tt>ACCEPT</tt></li>
</ul>
</td>
</tr>
</table>
<br>
<table COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
<a NAME="api"></a>Application Programming Interface</h2>
<i>See the comments in ficl.c and ficl.h for additional information, and
the example in file testmain.c.</i>
<br>
<dl>
<dt>
<b>void ficlInitSystem(int nDictCells)</b></dt>
<dd>
Initializes Ficl's shared system data structures, and creates the dictionary
allocating the specified number of CELLs from the heap (by a call to ficlMalloc)</dd>
<dt>
<b>void ficlTermSystem(void)</b></dt>
<dd>
Reclaims memory allocated for the ficl system including all dictionaries
and all virtual machines created by vmCreate. Any uses of the memory allocation
words (allocate and resize) are your problem.</dd>
<dt>
<b>int ficlBuild(char *name, FICL_CODE code, char flags)</b></dt>
<dd>
Create a primitive word in ficl's main dictionary with the given name,
code pointer, and properties (immediate, compile only, etc) as described
by the flags (see ficl.h for flag descriptions of the form FW_XXXX)</dd>
<dt>
<b>int ficlExec(FICL_VM *pVM, char *text)</b></dt>
<dd>
Feed the specified C string ('\0' terminated) to the given virtual machine
for evaluation. Returns various exception codes (VM_XXXX in ficl.h) to
indicate the reason for returning. Normal exit condition is VM_OUTOFTEXT,
indicating that the VM consumed the string successfully and is back for
more. ficlExec calls can be nested, and the function itself is re-entrant,
but note that a VM is static, so you have to take reasonable precautions
(for example, use one VM per thread in a multithreaded system if you want
multiple threads to be able to execute commands).</dd>
<dt>
<b>int ficlExecC(FICL_VM *pVM, char *text, int nChars)</b></dt>
<dd>
Same as ficlExec, but takes a count indicating the length of the supplied
string. Setting nChars to -1 is equivalent to ficlExec (expects '\0' termination).</dd>
<dt>
<b>int ficlExecXT(FICL_VM *pVM, FICL_WORD *pFW)</b></dt>
<dd>
Same as ficlExec, but takes a pointer to a FICL_WORD instead of a string.
Executes the word and returns after it has finished. If executing the word
results in an exception, this function will re-throw the same code if it
is nested under another ficlExec family function, or return the exception
code directly if not. This function is useful if you need to execute the
same word repeatedly - you save the dictionary search and outer interpreter
overhead.</dd>
<dt>
<b>FICL_VM *ficlNewVM(void)</b></dt>
<dd>
Create, initialize, and return a VM from the heap using ficlMalloc. Links
the VM into the system VM list for later reclamation by ficlTermSystem.</dd>
<dt>
<b>FICL_WORD *ficlLookup(char *name)</b></dt>
<dd>
Returns the address (also known as an XT in this case) of the specified
word in the main dictionary. If not found, returns NULL. The address can
be used in a call to ficlExecXT.</dd>
<dt>
<b>FICL_DICT *ficlGetDict(void)</b></dt>
<dd>
Returns a pointer to the main system dictionary, or NULL if the system
is uninitialized.</dd>
<dt>
<b>FICL_DICT *ficlGetEnv(void)</b></dt>
<dd>
Returns a pointer to the environment dictionary. This dictionary stores
information that describes this implementation as required by the Standard.</dd>
<dt>
<b>void ficlSetEnv(char *name, UNS32 value)</b></dt>
<dd>
Enters a new constant into the environment dictionary, with the specified
name and value.</dd>
<dt>
<b>void ficlSetEnvD(char *name, UNS32 hi, UNS32 lo)</b></dt>
<dd>
Enters a new double-cell constant into the environment dictionary with
the specified name and value.</dd>
<dt>
<b>FICL_DICT *ficlGetLoc(void)</b></dt>
<dd>
Returns a pointer to the locals dictionary. This function is defined only
if FICL_WANT_LOCALS is #defined as non-zero (see sysdep.h). The locals
dictionary is the symbol table for <a href="#locals">local variables</a>.</dd>
<dt>
<b>void ficlCompileCore(FICL_DICT *dp)</b></dt>
<dd>
Defined in words.c, this function builds ficl's primitives. </dd>
<dt>
<b>void ficlCompileSoftCore(FICL_VM *pVM)</b></dt>
<dd>
Defined in softcore.c, this function builds ANS required words and ficl
extras by evaluating a text string (think of it as a memory mapped file
;-) ). The string itself is built from files in the softwords directory
by PERL script softcore.pl. </dd>
</dl>
</td>
</tr>
</table>
<hr WIDTH="100%">
<table BORDER=0 CELLSPACING=5 WIDTH="600" >
<tr>
<td COLSPAN="2">
<h2>
<a NAME="manifest"></a>Ficl Source Files</h2>
</td>
</tr>
<tr>
<td><b>ficl.h</b></td>
<td>Declares most public functions and all data structures. Includes sysdep.h
and math.h</td>
</tr>
<tr>
<td><b>sysdep.h</b></td>
<td>Declares system dependent functions and contains build control macros.
Edit this file to port to another system.</td>
</tr>
<tr>
<td><b>math.h</b></td>
<td>Declares functions for 64 bit math</td>
</tr>
<tr>
<td><b>words.c</b></td>
<td>Exports ficlCompileCore(), the run-time dictionary builder, and contains
all primitive words as static functions.</td>
</tr>
<tr>
<td><b>vm.c</b></td>
<td>Virtual Machine methods</td>
</tr>
<tr>
<td><b>stack.c</b></td>
<td>Stack methods</td>
</tr>
<tr>
<td><b>ficl.c</b></td>
<td>System initialization, termination, and ficlExec</td>
</tr>
<tr>
<td><b>dict.c</b></td>
<td>Dictionary</td>
</tr>
<tr>
<td><b>math64.c</b></td>
<td>Implementation of 64 bit math words (except the two unsigned primitives
declared in sysdep.h and implemented in sysdep.c)</td>
</tr>
<tr>
<td><b>softcore.c</b></td>
<td>Contains all of the "soft" words - those written in Forth and compiled
by Ficl at startup time. Sources for these words are in the softwords directory.
The files softwords/softcore.bat and softwords/softcore.pl generate softcore.c
from the .fr sources.</td>
</tr>
<tr>
<td><b>sysdep.c</b></td>
<td>Implementation of system dependent functions declared in sysdep.h</td>
</tr>
<tr>
<td><b>softwords/</b></td>
<td>Directory contains sources and translation scripts for the words defined
in softcore.c. Softcore.c depends on most of the files in this directory.
See softcore.bat for the actual list of files that contribute to softcore.c.
This is where you'll find source code for the object oriented extensions.</td>
</tr>
</table>
<hr WIDTH="100%">
<table BORDER=0 CELLPADDING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
<a NAME="whatsnew"></a>What's new in version 2.03</h2>
This is the first version of Ficl that includes contributed code. Thanks
especially to Daniel Sobral, Michael Gauland for contributions and bug
finding.
<p>New words
<ul>
<li>
<tt><a href="#clock">clock</a>
(FICL)</tt></li>
<li>
<tt><a href="#clockspersec">clocks/sec</a>
(FICL)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans8.htm#8.6.1.1230">dnegate</a>
(DOUBLE)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans10.htm#10.6.2.1905">ms</a>
(FACILITY EXT - replaces MSEC)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans9.htm#9.6.1.2275">throw</a>
(EXCEPTION)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans9.htm#9.6.1.0875">catch</a>
(EXCEPTION)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans14.htm#14.6.1.0707">allocate</a>
(MEMORY)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans14.htm#14.6.1.1605">free</a>
(MEMORY)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans14.htm#14.6.1.2145">resize</a>
(MEMORY)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans6.htm#6.2.2440">within</a>
(CORE EXT)</tt></li>
<li>
<tt><a href="#alloc">alloc</a>
(class method)</tt></li>
<li>
<tt><a href="#allocarray">alloc-array</a>
(class method)</tt></li>
<li>
<tt><a href="#oofree">free</a>
(class method)</tt></li>
</ul>
Bugs Fixed
<ul>
<li>
Bug fix in isNumber(): used to treat chars between 'Z' and 'a' as valid
in base 10... (harmless, but weird)</li>
<li>
ficlExec pushes the <i>ip</i> and <tt>interpret</tt>s at the right times
so that nested calls to ficlExec behave the way you'd expect them to.</li>
<li>
<tt>evaluate</tt> respects count parameter, and also passes exceptional
return conditions back out to the calling instance of ficlExec.</li>
<li>
VM_QUIT now clears the locals dictionary in ficlExec.</li>
</ul>
Ficlwin Enhancements
<ul>
<li>
File Menu: recent file list and Open now load files.</li>
<li>
Text ouput function is now faster through use of string caching. Cache
flushes at the end of each line and each time ficlExec returns.</li>
<li>
Edit/paste now behaves more reasonably for text. File/open loads the specified
file.</li>
<li>
Registry entries specify dictionary and stack sizes, default window placement,
and whether or not to create a splitter for multiple VMs. See HKEY_CURRENT_USER/Software/CodeLab/ficlwin/Settings</li>
</ul>
Ficl Enhancements
<ul>
<li>
This version includes changes to make it <b>64 bit friendly</b>. This unfortunately
meant that I had to tweak some core data types and structures. I've tried
to make this transparent to 32 bit code, but a couple of things got renamed.
INT64 is now DPINT. UNS64 is now DPUNS. FICL_INT and FICL_UNS are synonyms
for INT32 and UNS32 in 32 bit versions, but a are obsolescent. Please use
the new data types instead. Typed stack operations on INT32 and UNS32 have
been renamed because they operate on CELL scalar types, which are 64 bits
wide on 64 bit systems. Added BITS_PER_CELL, which has legal values of
32 or 64. Default is 32.</li>
<li>
ficl.c: Added ficlExecXT() - executes an xt completely before returning,
passing back any exception codes generated in the process. Normal exit
code is VM_INNEREXIT.</li>
<li>
ficl.c: Added ficlExecC() to operate on counted strings as opposed to zero
terminated ones.</li>
<li>
ficlExec pushes ip and executes interpret at the right times so that nested
calls to ficlExec behave the way you'd expect them to.</li>
<li>
ficlSetStackSize() allows specification of stack size at run-time (affects
subsequent invocations of ficlNewVM()).</li>
<li>
vm.c: vmThrow() checks for (pVM->pState != NULL) before longjmping it.
vmCreate nulls this pointer initially. </li>
<li>
EXCEPTION wordset contributed by Daniel Sobral of FreeBSD</li>
<li>
MEMORY-ALLOC wordset contributed by Daniel Sobral, too. Added class methods
<tt>alloc</tt>
and <tt>alloc-array</tt> in softwords/oo.fr to allocate objects from the
heap.</li>
<li>
Control structure match check upgraded (thanks to Daniel Sobral for this
suggestion). Control structure mismatches are now errors, not warnings,
since the check accepts all syntactally legal constructs.</li>
<li>
Added vmInnerLoop() to vm.h. This function/macro factors the inner
interpreter out of ficlExec so it can be used in other places. Function/macro
behavior is conditioned on INLINE_INNER_LOOP in sysdep.h. Default: 1 unless
_DEBUG is set. In part, this is because VC++ 5 goes apoplectic when trying
to compile it as a function. See </li>
<br>comments in vm.c
<li>
EVALUATE respects the count parameter, and also passes exceptional return
conditions back out to the calling instance of ficlExec.</li>
<li>
VM_QUIT clears locals dictionary in ficlExec()</li>
<li>
Added Michael Gauland's ficlLongMul and ficlLongDiv and support routines
to math64.c and .h. These routines are coded in C, and are compiled only
if PORTABLE_LONGMULDIV == 1 (default is 0).</li>
<li>
Added definition of ficlRealloc to sysdep.c (needed for memory allocation
wordset). If your target OS supports realloc(), you'll probably want to
redefine ficlRealloc in those terms. The default version does ficlFree
followed by ficlMalloc.</li>
<li>
testmain.c: Changed gets() in testmain to fgets() to appease the security
gods.</li>
<li>
testmain: <tt>msec</tt> renamed to <tt><a href="#ficlms">ms</a></tt> in
line with the ANS</li>
<li>
softcore.pl now removes comments & spaces at the start and end of lines.
As a result: sizeof (softWords) == 7663 bytes (used to be 20000)
and consumes 11384 bytes of dictionary when compiled</li>
<li>
Deleted license paste-o in readme.txt (oops).</li>
</ul>
</td>
</tr>
</table>
<hr WIDTH="100%">
<table BORDER=0 CELLPADDING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
What's new in version 2.02</h2>
New words
<ul>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans6.htm#6.2.1850">marker</a>
(CORE EXT)</tt></li>
<li>
<tt><a href="http://www.taygeta.com/forth/dpans15.htm#15.6.2.1580">forget</a>
(TOOLS EXT)</tt></li>
<li>
<tt><a href="#ficlforgetwid">forget-wid</a>
(FICL)</tt></li>
<li>
<tt><a href="#ficlwordlist">ficl-wordlist</a> (FICL)</tt></li>
<li>
<tt><a href="#ficlvocabulary">ficl-vocabulary</a> (FICL)</tt></li>
<li>
<tt><a href="#ficlhide">hide</a>
(FICL)</tt></li>
<li>
<tt><a href="#ficlhidden">hidden</a>
(FICL)</tt></li>
<li>
<a href="#jhlocal">Johns Hopkins local variable syntax</a> (as best I can
determine)</li>
</ul>
Bugs Fixed
<ul>
<li>
<tt>forget</tt> now adjusts the dictionary pointer to remove the name of
the word being forgotten (name chars come before the word header in ficl's
dictionary)</li>
<li>
<tt>:noname</tt> used to push the colon control marker and its execution
token in the wrong order</li>
<li>
<tt>source-id</tt> now behaves correctly when loading a file.</li>
<li>
<tt>refill</tt> returns zero at EOF (Win32 load). Win32 <tt><a href="#ficlload">load</a></tt>
command continues to be misnamed. Really ought to be called <tt>included</tt>,
but does not exactly conform to that spec either (because <tt>included</tt>
expects a string signature on the stack, while Ficl's <tt><a href="#ficlload">load</a></tt>
expects a filename upon invocation). The "real" <tt>LOAD</tt> is a <tt>BLOCK</tt>
word.</li>
</ul>
Enhancements (IMHO)
<ul>
<li>
dictUnsmudge no longer links anonymous definitions into the dictionary</li>
<li>
<tt>oop</tt> is no longer the default compile wordlist at startup, nor
is it in the search order. Execute <b><tt>also oop definitions</tt></b>
to use Ficl OOP.</li>
<li>
Revised oo.fr extensively to make more use of early binding</li>
<li>
Added <tt>meta</tt> - a constant that pushes the address of metaclass.
See oo.fr for examples of use.</li>
<li>
Added classes: <tt>c-ptr c-bytePtr c-2bytePtr c-cellPtr
</tt>These
classes model pointers to non-object data, but each knows the size of its
referent.</li>
</ul>
</td>
</tr>
</table>
<hr WIDTH="100%">
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
What's new in version 2.01</h2>
<ul>
<li>
Bug fix: <tt>(local)</tt> used to leave a value on the stack between the
first and last locals declared. This value is now stored in a static.</li>
<li>
Added new local syntax with parameter re-ordering. <a href="#newlocal">See
description below</a>. (No longer compiled in version 2.02, in favor of
the Johns Hopkins syntax)</li>
</ul>
</td>
</tr>
</table>
<hr WIDTH="100%">
<table BORDER=0 CELLPADDING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
What's new in version 2.0</h2>
<ul>
<li>
New ANS Forth words: <tt>TOOLS</tt> and part of <tt>TOOLS EXT, SEARCH</tt>
and <tt>SEARCH EXT, LOCALS</tt> and <tt>LOCALS EXT</tt> word sets, additional
words from <tt>CORE EXT, DOUBLE</tt>, and <tt>STRING</tt>. (See the function
ficlCompileCore in words.c for an alphabetical list by word set).</li>
<li>
Simple <tt>USER</tt> variable support - a user variable is a virtual machine
instance variable. User variables behave as <tt>VARIABLE</tt>s in all other
respects.</li>
<li>
Object oriented syntax extensions (see below)</li>
<li>
Optional stack underflow and overflow checking in many CORE words (enabled
when FICL_ROBUST >= 2)</li>
<li>
Various bug fixes</li>
</ul>
</td>
</tr>
</table>
<hr WIDTH="100%">
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h3>
<a NAME="locals"></a>Local Variables</h3>
Ficl includes support for <tt>LOCALS</tt> and <tt>LOCALS EXT</tt> words
(all three of them!). I've implemented both of the local variable syntaxes
suggested in DPANS Appendix A.13. Examples: (By the way, Ficl implements
<tt>-ROT</tt>
as <tt>: -rot 2 -roll ;</tt> )
<ul><b><tt>\ Using LOCALS| from LOCALS EXT</tt></b>
<br><b><tt>: -rot ( a b c -- c a b )</tt></b>
<br><b><tt> locals| c b a |</tt></b>
<br><b><tt> c a b </tt></b>
<br><b><tt>;</tt></b>
<br><b><tt>\ Using LOCAL END-LOCAL</tt></b>
<br><b><tt>: -rot ( a b c -- c a b )</tt></b>
<br><b><tt> local c</tt></b>
<br><b><tt> local b</tt></b>
<br><b><tt> local a</tt></b>
<br><b><tt> end-locals</tt></b>
<br><b><tt> c a b</tt></b>
<br><b><tt>;</tt></b></ul>
Local variable support is optional because it adds a small amount of overhead
to the outer interpreter. You can disable it by setting FICL_WANT_LOCALS
to 0 in sysdep.h. Beware: much of the OOP code described below uses local
variables, so if you disable locals, you're going to lose other capabilities
too. Local variables can make Forth code quite a bit easier to read, so
I'd encourage you to experiment with them.
<br>The default maximum number of local variables is 16. It's controlled
by FICL_MAX_LOCALS in sysdep.h.
<br><a NAME="jhlocal"></a>Ficl 2.02 includes by default an implementation
of the Johns Hopkins local syntax (as best I can determine it from examples
on the web). This syntax lets you declare local variables that look very
much like a stack comment. Variables in the declaration appear in the "correct"
order for a stack comment. Everything after the -- is treated as a comment.
In addition, you can insert a | before the -- to declare one or more zero-initialized
locals. Example:
<blockquote><b><tt>:tuck0 { a b c | d -- 0 a b c }</tt></b>
<br><b><tt> d a b c ;</tt></b></blockquote>
The | and -- delimiters can appear at most once, and must appear in the
order shown in the example to work correctly. The local declaration ends
at the first occurrence of }. The declaration must all be on one line as
presently implemented.
<br><a NAME="newlocal"></a>Ficl 2.01 added yet another local syntax that
models a stack comment. This one is not compiled in the release, but you
can add it by editing softwords/softcore.bat to include the file ficllocal.fr.
In this case, parameters are re-ordered so that the rightmost initialized
param comes from the top of the stack. The syntax is:
<ul><b><tt>{{ <initialized params> -- <cleared params> }}</tt></b></ul>
You can omit either the initialized or the cleared parameters. Parameters
after the double dash are set to zero initially. Those to the left are
initialized from the stack at execution time. Examples (lame ones, admittedly):
<ul><b><tt>: -rot ( a b c -- c a b )</tt></b>
<br><b><tt> {{ a b c }}</tt></b>
<br><b><tt> c a b </tt></b>
<br><b><tt>;</tt></b>
<br><b><tt>: tuck0 ( a b c -- 0 a b c )</tt></b>
<br><b><tt> {{ a b c -- d }}</tt></b>
<br><b><tt> d a b c </tt></b>
<br><b><tt>; </tt></b></ul>
<h3>
Search Order</h3>
Ficl implements many of the search order words in terms of two primitives
called <tt><a href="#tosearch">>SEARCH</a></tt> and <tt><a href="#searchfrom">SEARCH></a></tt>.
As their names suggest (assuming you're familiar with Forth), they push
and pop the search order stack. See the list of <a href="#extras">Ficl
extras</a> for details.
<br>The standard does not appear to specify any conditions under which
the search order is reset to a sane state. Ficl resets the search order
to its default state whenever <tt>ABORT</tt> happens. This includes stack
underflows and overflows. <tt>QUIT</tt> does not affect the search order.
The minimum search order (set by <tt>ONLY</tt>) is equivalent to
<br><b><tt>FORTH-WORDLIST 1 SET-ORDER</tt></b>
<br>There is a default maximum of 16 wordlists in the search order. This
can be changed by redefining FICL_DEFAULT_VOCS (declared in sysdep.h).
<h3>
Soft Words</h3>
Many words from all the supported wordsets are written in Forth, and stored
as a big string that Ficl compiles when it starts. The sources for all
of these words are in directory ficl/softwords. There is a .bat file (softcore.bat)
and a PERL 5 script (softcore.pl) that convert Forth files into the file
softcore.c, so softcore.c is really dependent on the Forth sources. This
is not reflected in the Visual C++ project database. For the time being,
it's a manual step. You can edit softcore.bat to change the list of files
that contribute to softcore.c. </td>
</tr>
</table>
<h2>
<hr WIDTH="100%"></h2>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
<a NAME="objects"></a>Objects in ficl</h2>
Ficl is not the first Forth to include Object Oriented extensions. Ficl's
OO syntax owes a debt to the work of John Hayes and Dick Pountain, among
others. OO Ficl is different from other OO Forths in a few ways, though
(some things never change). First, unlike several implementations, the
syntax is documented (<a href="#ootutorial">below</a>) beyond the source
code. In Ficl's spirit of working with C code, the OO syntax provides means
to adapt existing data structures. I've tried to make Ficl's OO model simple
and safe by unifying classes and objects, providing late binding by default,
and separating namespaces so that methods and regular Forth words are not
easily confused.
<h3>
Design goals of Ficl OO syntax</h3>
Ficl's object extensions provide the traditional OO benefits of associating
data with the code that manipulates it, and reuse through single inheritance.
Ficl also has some unusual capabilities that support interoperation with
systems written in C.
<ul>
<li>
Ficl objects are normally late bound for safety (late binding guarantees
that the appropriate method will always be invoked for a particular object).
Early binding is also available, provided you know the object's class at
compile-time.</li>
<li>
Ficl OOP supports single inheritance, aggregation, and arrays of objects.</li>
<li>
Classes have independent name spaces for their methods: methods are only
visible in the context of a class or object. Methods can be overridden
or added in subclasses; there is no fixed limit on the number of methods
of a class or subclass.</li>
<li>
Ficl OOP syntax is regular and unified over classes and objects. In ficl,
a classes are objects. Class methods include the ability to subclass and
instantiate.</li>
<li>
Ficl can adapt legacy data structures with object wrappers. You can model
a structure in a Ficl class, and create an instance that refers to an address
in memory that holds an instance of the structure. The <i>ref object</i>
can then manipulate the structure directly. This lets you wrap data structures
written and instantiated in C.</li>
</ul>
</td>
</tr>
</table>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h3>
Ficl Object Model</h3>
All classes in Ficl are derived from the common base class <tt><a href="#objectgloss">OBJECT</a></tt>.
All classes are instances of <tt><a href="#glossclass">METACLASS</a></tt>.
This means that classes are objects, too. <tt>METACLASS</tt> implements
the methods for messages sent to classes. Class methods create instances
and subclasses, and give information about the class. Classes have exactly
three elements:
<ul>
<li>
The address ( <tt>.CLASS</tt> ) of a parent class, or zero if it's a base
class (only <tt>OBJECT</tt> and <tt>METACLASS</tt> have this property)</li>
<li>
The size ( <tt>.SIZE</tt> ) in address units of an instance of the class</li>
<li>
A wordlist ID ( <tt>.WID</tt> ) for the methods of the class</li>
</ul>
In the figure below, <tt>METACLASS</tt> and <tt>OBJECT</tt> are system-supplied
classes. The others are contrived to illustrate the relationships among
derived classes, instances, and the two system base classes. The dashed
line with an arrow at the end indicates that the object/class at the arrow
end is an instance of the class at the other end. The vertical line with
a triangle denotes inheritance.
<p>Note for the curious: <tt>METACLASS</tt> behaves like a class - it responds
to class messages and has the same properties as any other class. If you
want to twist your brain in knots, you can think of <tt>METACLASS</tt>
as an instance of itself.
<br> </td>
</tr>
</table>
<img SRC="ficl_oop.jpg" VSPACE=10 height=442 width=652>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
<a NAME="ootutorial"></a>Ficl OO Syntax Tutorial</h2>
<h3>
Introduction</h3>
Ficl objects associate a class with an instance (really the storage for
one set of instance variables). This is done explicitly, in that any Ficl
object is represented by the cell pair:
<ul><b><tt>( instance-addr class-addr )</tt></b></ul>
on the stack. Whenever a named Ficl object executes, it leaves this "signature".
All methods expect a class and instance on the stack when they execute,
too. In many other OO languages, including C++, instances contain information
about their classes (a vtable pointer, for example). By making this pairing
explicit rather than implicit, Ficl can be OO about chunks of data that
don't realize that they are objects, without sacrificing any robustness
for native objects. Whenever you create an object in Ficl, you specify
its class. After that, the object always pushes its class and the address
of its payload (instance variable space) when invoked by name.
<p>Classes are special kinds of objects that store the methods of their
instances, the size of an instance's payload, and a parent class pointer.
Classes themselves are instances of a special base class called <tt>METACLASS</tt>,
and all classes inherit from class <tt>OBJECT</tt>. This is confusing at
first, but it means that Ficl has a very simple syntax for constructing
and using objects. Class methods include subclassing (<tt>SUB</tt>), creating
initialized and uninitialized instances (<tt>NEW</tt> and <tt>INSTANCE</tt>),
and creating reference instances (<tt>REF</tt>). Classes also have methods
for disassembling their methods (<tt>SEE</tt>), identifying themselves
(<tt>ID</tt>), and listing their pedigree (<tt>PEDIGREE</tt>). All objects
inherit methods for initializing instances and arrays of instances, for
performing array operations, and for getting information about themselves.
<h3>
Methods and messages</h3>
Methods are the chunks of code that objects execute in response to messages.
A message is a request to an object for a behavior that the object supports.
When it receives a message, the target object looks up a method that performs
the behavior for its class, and executes it. Any specific message will
be bound to different methods in different objects, according to class.
This separation of messages and methods allows objects to behave polymorphically.
(In Ficl, methods are words defined in the context of a class, and messages
are the names of those words.) Ficl classes associate messages with methods
for their instances (a fancy way of saying that each class owns a wordlist).
Ficl provides a late-binding operator <b><tt>--></tt></b> that sends messages
to objects at run-time, and an early-binding operator
<b><tt>=></tt></b>
that compiles a specific class's method. These operators are the only supported
way to invoke methods. Regular Forth words are not visible to the method-binding
operators, so there's no chance of confusing a message with a regular
word of the same name. </td>
</tr>
</table>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h3>
Tutorial (finally!)</h3>
Since this is a tutorial, I'm assuming you're following along by pasting
the examples into ficlWin, the Win32 version of Ficl (or some other build
that includes the OO part of softcore.c). I also assume that you're familiar
with Forth. If not, please see one of the <a href="#links">references</a>,
below. Ficl's OOP words are in vocabulary OOP. To put OOP in the search
order and make it the compilation wordlist from the default search order
(as set by <tt>ONLY</tt>), type:
<ul><b><tt>ONLY ( reset to default search order )</tt></b>
<br><b><tt>ALSO OOP DEFINITIONS</tt></b></ul>
To start, we'll work with the two base classes <tt>OBJECT</tt> and <tt>METACLASS</tt>.
Try this:
<ul><b><tt>metaclass --> methods</tt></b></ul>
The line above contains three words. The first is the name of a class,
so it pushes its signature on the stack. Since all classes are instances
of <tt>METACLASS</tt>, <tt>METACLASS</tt> behaves as if it is an instance
of itself (this is the only class with this property). It pushes the same
address twice: once for the class and once for the payload, since they
are the same. The next word finds a method in the context of a class and
executes it. In this case, the name of the method is <tt>methods</tt>.
Its job is to list all the methods that a class knows. What you get when
you execute this line is a list of all the class methods Ficl provides.
<ul><b><tt>object --> sub c-foo</tt></b></ul>
Causes base-class <tt>OBJECT</tt> to derive from itself a new class called
c-foo. Now we'll add some instance variables and methods to the new class...
<ul><b><tt>cell: m_cell1</tt></b>
<br><b><tt>4 chars: m_chars</tt></b>
<br><b><tt>: init ( inst class -- )</tt></b>
<br><b><tt> locals| class inst |</tt></b>
<br><b><tt> 0 inst class --> m_cell1 !</tt></b>
<br><b><tt> inst class --> m_chars 4 0 fill</tt></b>
<br><b><tt> ." initializing an instance of c_foo at "
inst x. cr</tt></b>
<br><b><tt>;</tt></b>
<br><b><tt>end-class</tt></b></ul>
The first two lines add named instance variables to the class, and create
a method for each. <i>Untyped</i> instance variable methods (like those
created by <tt>cell: cells: char:</tt> and <tt>chars:</tt>) just push the
address of the corresponding instance variable when invoked on an instance
of the class. It's up to you to remember the size of the instance variable
and manipulate it with the usual Forth words for fetching and storing (we'll
see below how to aggregate objects, which do know their size). We've also
defined a method called <tt>init</tt> that clears the instance variables.
Notice that the method expects the addresses of the class and instance
when it's called. It stashes those in local variables to avoid stack tricks,
and puts them onto the stack whenever it calls a method. In this case,
we're storing zero to the two member variables.
<p>The <tt>init</tt> method is special for Ficl objects: whenever you create
an initialized instance using <b><tt>new</tt></b> or <b><tt>new-array</tt></b>,
Ficl calls the class's <tt>init</tt> method for you on that instance. The
default <tt>init</tt> method supplied by class <tt>object</tt> clears the
instance, so we didn't really need to override it in this case (see the
source code in ficl/softwords/oo.fr).
<br>Now make an instance of the new class:
<ul><b><tt>c-foo --> new foo-instance</tt></b></ul>
And try a few things...
<ul><b><tt>foo-instance --> methods</tt></b>
<br><b><tt>foo-instance --> pedigree</tt></b></ul>
Or you could type this with the same effect:
<ul><b><tt>foo-instance 2dup --> methods --> pedigree</tt></b></ul>
Notice that we've overridden the init method supplied by object, and added
two more methods for the member variables. If you type WORDS, you'll see
that these methods are not visible outside the context of the class that
contains them. The method finder --> uses the class to look up methods.
You can use this word in a definition, as we did in <tt>init</tt>, and
it performs late binding, meaning that the mapping from message (method
name) to method (the code) is deferred until run-time. To see this, you
can decompile the init method like this:
<ul><b><tt>c-foo --> see init</tt></b>
<br>or
<br><b><tt>foo-instance --> class --> see init</tt></b></ul>
Ficl also provides early binding, but you have to ask for it. Ficl's early
binding operator pops a class off the stack and compiles the method you've
named, so that that method executes regardless of the class of object it's
used on. This can be dangerous, since it defeats the data-to-code matching
mechanism object oriented languages were created to provide, but it does
increase run-time speed by binding the method at compile time. In many
cases, such as the init method, you can be reasonably certain of the class
of thing you're working on. This is also true when invoking class methods,
since all classes are instances of <tt>metaclass</tt>. Here's an example
from the definition of <tt>metaclass</tt> in oo.fr (don't paste this into
ficlWin - it's already there):
<ul><b><tt>: new \ ( class metaclass "name" -- )</tt></b>
<br><b><tt> metaclass => instance --> init ;</tt></b></ul>
Try this...
<ul><b><tt>metaclass --> see new</tt></b></ul>
Decompiling the method with <tt>SEE</tt> shows the difference between the
two strategies. The early bound method is compiled inline, while the late-binding
operator compiles the method name and code to find and execute it in the
context of whatever class is supplied on the stack at run-time.
<br>Notice that the early-binding operator requires a class at compile
time. For this reason, classes are <tt>IMMEDIATE</tt>, meaning that they
push their signature at compile time or run time. I'd recommend that you
avoid early binding until you're very comfortable with Forth, object-oriented
programming, and Ficl's OOP syntax.
<p>As advertised earlier, Ficl provides ways to objectify existing data
structures without changing them. Instead, you can create a Ficl class
that models the structure, and instantiate a <b>ref </b>from this class,
supplying the address of the structure. After that, the <i>ref instance</i>
behaves as a Ficl object, but its instance variables take on the values
in the existing structure. Example (from ficlclass.fr):
<br>
<ul><b><tt>object subclass c-wordlist \ OO model of FICL_HASH</tt></b>
<br><b><tt> cell: .parent</tt></b>
<br><b><tt> cell: .size</tt></b>
<br><b><tt> cell: .hash</tt></b>
<p><b><tt> : push drop >search ;</tt></b>
<br><b><tt> : pop 2drop previous ;</tt></b>
<br><b><tt> : set-current drop set-current ;</tt></b>
<br><b><tt> : words --> push words previous ;</tt></b>
<br><b><tt>end-class</tt></b>
<p><b><tt>: named-wid ( "name" -- ) </tt></b>
<br><b><tt> wordlist postpone c-wordlist
metaclass => ref ;</tt></b></ul>
In this case, <tt>c-wordlist</tt> describes Ficl's wordlist structure;
named-wid creates a wordlist and binds it to a ref instance of <tt>c-wordlist</tt>.
The fancy footwork with <tt>POSTPONE</tt> and early binding is required
because classes are immediate. An equivalent way to define named-wid with
late binding is:
<ul><b><tt>: named-wid ( "name" -- )</tt></b>
<br><b><tt> wordlist postpone c-wordlist
--> ref ;</tt></b></ul>
To do the same thing at run-time (and call it my-wordlist):
<ul><b><tt>wordlist c-wordlist --> ref my-wordlist</tt></b></ul>
Now you can deal with the wordlist through the ref instance:
<ul><b><tt>my-wordlist --> push</tt></b>
<br><b><tt>my-wordlist --> set-current</tt></b>
<br><b><tt>order</tt></b></ul>
Ficl can also model linked lists and other structures that contain pointers
to structures of the same or different types. The class constructor word
<b><tt><a href="#exampleref:">ref:</a></tt></b>
makes an aggregate reference to a particular class. See the <a href="#glossinstance">instance
variable glossary</a> for an <a href="#exampleref:">example</a>.
<p>Ficl can make arrays of instances, and aggregate arrays into class descripions.
The <a href="#glossclass">class methods</a> <b><tt>array</tt></b> and <b><tt>new-array</tt></b>
create uninitialized and initialized arrays, respectively, of a class.
In order to initialize an array, the class must define (or inherit) a reasonable
<b><tt>init</tt></b>
method. <b><tt>New-array</tt></b> invokes it on each member of the array
in sequence from lowest to highest. Array instances and array members use
the object methods <b><tt>index</tt></b>, <b><tt>next</tt></b>, and <b><tt>prev</tt></b>
to navigate. Aggregate a member array of objects using <b><tt><a href="#arraycolon">array:</a></tt></b>.
The objects are not automatically initialized in this case - your class
initializer has to call <b><tt>array-init</tt></b> explicitly if you want
this behavior.
<p>For further examples of OOP in Ficl, please see the source file ficl/softwords/ficlclass.fr.
This file wraps several Ficl internal data structures in objects and gives
use examples. </td>
</tr>
</table>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h3>
<a NAME="oopgloss"></a>OOP Glossary</h3>
Note: with the exception of the binding operators (the first two definitions
here), all of the words in this section are internal factors that you don't
need to worry about. These words provide method binding for all classes
and instances. Also described are supporting words and execution factors.
All are defined in softwords/oo.fr.
<dl>
<dt>
<b><tt>--> ( instance class "method-name" -- xn )</tt></b></dt>
<dd>
Late binding: looks up and executes the given method in the context of
the class on top of the stack. </dd>
<dt>
<b><tt>=> comp: ( class meta "method-name" -- ) exec:
( inst class -- xn )</tt></b></dt>
<dd>
Early binding: compiles code to execute the method of the class specified
at compile time.</dd>
<dt>
<b><tt>do-do-instance</tt></b></dt>
<dd>
When executed, causes the instance to push its ( instance class ) stack
signature. Implementation factor of <b><tt>metaclass --> sub</tt></b>.
Compiles <b><tt>.do-instance</tt></b> in the context of a class; <tt>.do-instance</tt>
implements the <tt>does></tt> part of a named instance. </dd>
<dt>
<b><tt>exec-method ( instance class c-addr u -- xn )</tt></b></dt>
<dd>
Given the address and length of a message (method name) on the stack, finds
the method in the context of the specified class and invokes it. Upon entry
to the method, the instance and class are on top of the stack, as usual.
If unable to find the method, prints an error message and aborts.</dd>
<dt>
<b><tt>find-method-xt ( class "method-name" -- class xt )</tt></b></dt>
<dd>
Attempts to map the message to a method in the specified class. If successful,
leaves the class and the execution token of the method on the stack. Otherwise
prints an error message and aborts.</dd>
<dt>
<b><tt>lookup-method ( class c-addr u -- class xt )</tt></b></dt>
<dd>
Given the address and length of a message (method name) on the stack, finds
the method in the context of the specified class. If unable to find the
method, prints an error message and aborts.</dd>
<dt>
<b><tt>parse-method comp: ( "method-name" -- ) exec:
( -- c-addr u )</tt></b></dt>
<dd>
Parse "name" from the input stream and compile code to push its length
and address when the enclosing definition runs.</dd>
</dl>
</td>
</tr>
</table>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h3>
<a NAME="glossinstance"></a>Instance Variable Glossary</h3>
<b>Note</b>: these words are only visible when creating a subclass! To
create a subclass, use the <tt>sub</tt> method on <tt>object</tt> or any
class derived from it (<i>not</i> <tt>metaclass</tt>). Source code for
Ficl OOP is in ficl/softwords/oo.fr.
<dt>
Instance variable words do two things: they create methods that do an action
appropriate for the type of instance variable they represent, and they
reserve space in the class template for the instance variable. We'll use
the term <i>instance variable</i> to refer both to the method that gives
access to a particular field of an object, and to the field itself. Rather
than give esentially the same example over and over, here's one example
that shows several of the instance variable construction words in use:</dt>
<ul>
<dt>
<tt>object subclass c-example</tt></dt>
<dt>
<tt> cell:
.cell0</tt></dt>
<br><tt> c-4byte obj: .nCells</tt>
<br><tt> 4 c-4byte array: .quad</tt>
<br><tt> char:
.length</tt>
<br><tt> 79 chars:
.name</tt>
<br><tt>end-class</tt></ul>
This class only defines instance variables, and it inherits some methods
from <tt>object</tt>. Each untyped instance variable (.cell0, .length,
.name) pushes its address when executed. Each object instance variable
pushes the address and class of the aggregate object. Similar to C, an
array instance variable leaves its base address (and its class) when executed.
The word <tt>subclass</tt> is shorthand for "<tt>--> sub</tt>"
<br>
<dt>
<b><tt>cell: ( offset "name"
-- offset' )</tt></b></dt>
<dt>
<b><tt>
Execution: ( -- cell-addr )</tt></b></dt>
<dl>
<dd>
Create an untyped instance variable one cell wide. The instance variable
leaves its payload's address when executed. </dd>
<dt>
<b><tt>cells: ( offset nCells "name"
-- offset' )</tt></b></dt>
<dt>
<b><tt>
Execution: ( -- cell-addr )</tt></b></dt>
<dd>
Create an untyped instance variable n cells wide.</dd>
<dt>
<b><tt>char: ( offset "name"
-- offset' )</tt></b></dt>
<dt>
<b><tt>
Execution: ( -- char-addr )</tt></b></dt>
<dd>
Create an untyped member variable one char wide</dd>
<dt>
<b><tt>chars: ( offset nChars "name"
-- offset' )</tt></b></dt>
<dt>
<b><tt>
Execution: ( -- char-addr )</tt></b></dt>
<dd>
Create an untyped member variable n chars wide.</dd>
<dt>
<b><tt>obj: ( offset class
meta "name" -- offset' )</tt></b></dt>
<dt>
<b><tt>
Execution: ( -- instance class )</tt></b></dt>
<dd>
Aggregate an uninitialized instance of <b>class</b> as a member variable
of the class under construction.</dd>
<dt>
<a NAME="arraycolon"></a><b><tt>array:
( offset n class meta "name" -- offset' )</tt></b></dt>
<dt>
<b><tt>
Execution: ( -- instance class )</tt></b></dt>
<dd>
Aggregate an uninitialized array of instances of the class specified as
a member variable of the class under construction.</dd>
<dt>
<a NAME="exampleref:"></a><b><tt>ref:
( offset class meta "name" -- offset' )</tt></b></dt>
<dt>
<b><tt>
Execution: ( -- ref-instance ref-class )</tt></b></dt>
<dd>
Aggregate a reference to a class instance. There is no way to set the value
of an aggregated ref - it's meant as a way to manipulate existing data
structures with a Ficl OO model. For example, if your system contains a
linked list of 4 byte quantities, you can make a class that represents
a list element like this: </dd>
<dl>
<dd>
<tt>object subclass c-4list</tt></dd>
<dd>
<tt>c-4list ref: .link</tt></dd>
<dd>
<tt>c-4byte obj: .payload</tt></dd>
<dd>
<tt>end-class;</tt></dd>
<dd>
<tt>address-of-existing-list c-4list --> ref mylist</tt></dd>
</dl>
<dd>
The last line binds the existing structure to an instance of the class
we just created. The link method pushes the link value and the class c_4list,
so that the link looks like an object to Ficl and like a struct to C (it
doesn't carry any extra baggage for the object model - the Ficl methods
alone take care of storing the class information). </dd>
<dd>
Note: Since a ref: aggregate can only support one class, it's good for
modeling static structures, but not appropriate for polymorphism. If you
want polymorphism, aggregate a c_ref (see classes.fr for source) into your
class - it has methods to set and get an object.</dd>
<dd>
By the way, it is also possible to construct a pair of classes that contain
aggregate pointers to each other. Here's a rough example:</dd>
<dl>
<dd>
<tt>object subclass c-fee</tt></dd>
<dd>
<tt>object subclass c-fie</tt></dd>
<dd>
<tt> c-fee ref: .fee</tt></dd>
<dd>
<tt>end-class \ done with
c-fie</tt></dd>
<dd>
<tt> c-fie ref: .fie</tt></dd>
<br><tt>end-class
\ done with c-fee</tt></dl>
</dl>
</td>
</tr>
</table>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h3>
<a NAME="glossclass"></a>Class Methods Glossary</h3>
These words are methods of <tt>metaclass</tt>. They define the manipulations
that can be performed on classes. Methods include various kinds of instantiation,
programming tools, and access to member variables of classes. Source is
in softwords/oo.fr.
<dl>
<dt>
<b><tt>instance ( class metaclass "name" -- instance
class )</tt></b> </dt>
<dd>
Create an uninitialized instance of the class, giving it the name specified.
The method leaves the instance 's signature on the stack (handy if you
want to initialize). Example:</dd>
<dd>
<tt>c_ref --> instance uninit-ref 2drop</tt></dd>
<dt>
<b><tt>new ( class
metaclass "name" -- )</tt></b> </dt>
<dd>
Create an initialized instance of class, giving it the name specified.
This method calls init to perform initialization. </dd>
<dt>
<b><tt>array ( nObj class metaclass
"name" -- nObjs instance class )</tt></b> </dt>
<dd>
Create an array of nObj instances of the specified class. Instances are
not initialized. Example:</dd>
<dd>
<tt>10 c_4byte --> array 40-raw-bytes 2drop drop</tt></dd>
<dt>
<b><tt>new-array ( nObj class metaclass "name" -- )</tt></b> </dt>
<dd>
Creates an initialized array of nObj instances of the class. Same syntax
as <tt>array</tt></dd>
<dt>
<a NAME="alloc"></a><b><tt>alloc ( class metaclass -- instance
class )</tt></b></dt>
<dd>
Creates an anonymous instance of <b>class</b> from the heap (using a call
to ficlMalloc() to get the memory). Leaves the payload and class addresses
on the stack. Usage example:</dd>
<dd>
<tt>c-ref --> alloc 2constant instance-of-ref</tt></dd>
<dd>
Creates a double-cell constant that pushes the payload and class address
of a heap instance of c-ref.</dd>
<dt>
<a NAME="allocarray"></a><b><tt>alloc-array ( nObj class metaclass
-- instance class )</tt></b></dt>
<dd>
Same as new-array, but creates anonymous instances from the heap using
a call to ficlMalloc(). Each instance is initialized using the class's
<tt>init</tt>
method</dd>
<dt>
<b><tt>ref ( instance-addr
class metaclass "name" -- )</tt></b> </dt>
<dd>
Make a ref instance of the class that points to the supplied instance address.
No new instance space is allotted. Instead, the instance refers to the
address supplied on the stack forever afterward. For wrapping existing
structures.</dd>
</dl>
<dl>
<dt>
<b><tt>sub ( class
metaclass -- old-wid addr[size] size )</tt></b></dt>
<dd>
Derive a subclass. You can add or override methods, and add instance variables.
Alias: <tt>subclass</tt>. Examples:</dd>
<dl>
<dd>
<tt>c_4byte --> sub c_special4byte</tt></dd>
<dd>
<tt>( your new methods and instance variables here )</tt></dd>
<dd>
<tt>end-class</tt></dd>
<dd>
or</dd>
<dd>
<tt>c_4byte subclass c_special4byte</tt></dd>
<dd>
<tt>( your new methods and instance variables here )</tt></dd>
<dd>
<tt>end-class</tt></dd>
</dl>
<dt>
<b><tt>.size ( class metaclass
-- instance-size )</tt></b> </dt>
<dd>
Returns address of the class's instance size field, in address units. This
is a metaclass member variable.</dd>
<dt>
<b><tt>.super ( class metaclass --
superclass )</tt></b> </dt>
<dd>
Returns address of the class's superclass field. This is a metaclass member
variable.</dd>
<dt>
<b><tt>.wid ( class metaclass
-- wid )</tt></b> </dt>
<dd>
Returns the address of the class's wordlist ID field. This is a metaclass
member variable.</dd>
<dt>
<b><tt>get-size</tt></b></dt>
<dd>
Returns the size of an instance of the class in address units. Imeplemented
as</dd>
<dd>
<tt>: get-size metaclass => .size @ ;</tt></dd>
<dt>
<b><tt>get-wid</tt></b></dt>
<dd>
Returns the wordlist ID of the class. Implemented as </dd>
<dd>
<tt>: get-wid metaclass => .wid @ ;</tt></dd>
<dt>
<b><tt>get-super</tt></b></dt>
<dd>
Returns the class's superclass. Implemented as</dd>
<dd>
<tt>: get-super metaclass => .super @ ;</tt></dd>
<dt>
<b><tt>id (
class metaclass -- c-addr u )</tt></b> </dt>
<dd>
Returns the address and length of a string that names the class.</dd>
<dt>
<b><tt>methods ( class metaclass -- )</tt></b> </dt>
<dd>
Lists methods of the class and all its superclasses</dd>
<dt>
<b><tt>offset-of ( class metaclass "name" -- offset )</tt></b></dt>
<dd>
Pushes the offset from the instance base address of the named member variable.
If the name is not that of an instance variable method, you get garbage.
There is presently no way to detect this error. Example:</dd>
<dl>
<dd>
<tt>metaclass --> offset-of .wid</tt></dd>
</dl>
<dt>
<b><tt>pedigree ( class metaclass -- )</tt></b> </dt>
<dd>
Lists the pedigree of the class (inheritance trail)</dd>
<dt>
<b><tt>see ( class
metaclass "name" -- )</tt></b> </dt>
<dd>
Decompiles the specified method - obect version of <tt>SEE</tt>, from the
<tt>TOOLS</tt>
wordset.</dd>
</dl>
</td>
</tr>
</table>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h3>
<a NAME="objectgloss"></a><tt>object</tt> base-class Methods Glossary</h3>
These are methods that are defined for all instances by the base class
<tt>object</tt>.
The methods include default initialization, array manipulations, aliases
of class methods, upcasting, and programming tools.
<dl>
<dt>
<b><tt>init ( instance
class -- )</tt> </b></dt>
<dd>
Default initializer called automatically for all instances created with
<tt>new</tt>
or <tt>new-array</tt>. Zero-fills the instance. You do not normally need
to invoke <tt>init</tt> explicitly.</dd>
<dt>
<b><tt>array-init ( nObj instance class -- )</tt></b> </dt>
<dd>
Applies <tt>init</tt> to an array of objects created by <tt>new-array</tt>.
Note that <tt>array:</tt> does not cause aggregate arrays to be initialized
automatically. You do not normally need to invoke <tt>array-init</tt> explicitly.</dd>
<dt>
<a NAME="oofree"></a><b><tt>free ( instance class -- )</tt></b></dt>
<dd>
Releases memory used by an instance previously created with <tt>alloc</tt>
or <tt>alloc-array</tt>. Note - this method is not presently protected
against accidentally deleting something from the dictionary. If you do
this, Bad Things are likely to happen. Be careful for the moment to apply
free only to instances created with <tt>alloc</tt> or <tt>alloc-array</tt>.</dd>
<dt>
<b><tt>class ( instance class
-- class metaclass )</tt></b> </dt>
<dd>
Convert an object signature into that of its class. Useful for calling
class methods that have no object aliases.</dd>
<dt>
<b><tt>super ( instance class
-- instance parent-class )</tt></b> </dt>
<dd>
Upcast an object to its parent class. The parent class of <tt>object</tt>
is zero. Useful for invoking an overridden parent class method.</dd>
<dt>
<b><tt>pedigree ( instance class -- )</tt></b> </dt>
<dd>
Display an object's pedigree - its chain of inheritance. This is an alias
for the corresponding class method.</dd>
<dt>
<b><tt>size ( instance
class -- sizeof(instance) )</tt></b> </dt>
<dd>
Returns the size, in address units, of one instance. Does not know about
arrays! This is an alias for the class method <tt>get-size</tt></dd>
<dt>
<b><tt>methods ( instance class -- )</tt></b> </dt>
<dd>
Class method alias. Displays the list of methods of the class and all superclasses
of the instance.</dd>
<dt>
<b><tt>index ( n instance class
-- instance[n] class )</tt></b> </dt>
<dd>
Convert array-of-objects base signature into signature for array element
n. No check for bounds overflow. Index is zero-based, like C, so </dd>
<dl>
<dd>
<tt>0 my-obj --> index</tt> </dd>
</dl>
<dd>
is equivalent to </dd>
<dl>
<dd>
<tt>my-obj</tt></dd>
</dl>
<dd>
Check out the <a href="#minusrot">description of <tt>-ROT</tt></a> for
help in dealing with indices on the stack.</dd>
<dt>
<b><tt>next ( instance[n]
class -- instance[n+1] class )</tt></b> </dt>
<dd>
Convert an array-object signature into the signature of the next
object in the array. No check for bounds overflow.</dd>
<dt>
<b><tt>prev ( instance[n]
class -- instance[n-1] class )</tt></b> </dt>
<br>Convert an object signature into the signature of the previous object
in the array. No check for bounds underflow.</dl>
</td>
</tr>
</table>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h3>
<a NAME="stockclasses"></a>Supplied Classes (See classes.fr)</h3>
<dl>
<dt>
<b><tt>metaclass </tt></b></dt>
<dd>
Describes all classes of Ficl. Contains class methods. Should never be
directly instantiated or subclassed. Defined in oo.fr. Methods described
above.</dd>
<dt>
<b><tt>object</tt> </b></dt>
<dd>
Mother of all Ficl objects. Defines default initialization and array indexing
methods. Defined in oo.fr. Methods described above.</dd>
<dt>
<b><tt>c-ref</tt> </b></dt>
<dd>
Holds the signature of another object. Aggregate one of these into a data
structure or container class to get polymorphic behavior. Methods &
members: </dd>
<dd>
<tt>get ( inst class -- ref-inst ref-class )</tt></dd>
<dd>
<tt>set ( ref-inst ref-class inst class -- )</tt></dd>
<dd>
<tt>.instance ( inst class -- a-addr ) </tt>cell member that
holds the instance</dd>
<dd>
<tt>.class ( inst class -- a-addr ) </tt>cell member that holds
the class</dd>
<dt>
<b><tt>c-byte </tt></b></dt>
<dd>
Primitive class derived from <tt>object</tt>, with a 1-byte payload. Set
and get methods perform correct width fetch and store. Methods & members:</dd>
<dd>
<tt>get ( inst class -- c )</tt></dd>
<dd>
<tt>set ( c inst class -- )</tt></dd>
<dd>
<tt>.payload ( inst class -- addr ) </tt>member holds instance's
value</dd>
<dt>
<b><tt>c-2byte</tt></b> </dt>
<dd>
Primitive class derived from <tt>object</tt>, with a 2-byte payload. Set
and get methods perform correct width fetch and store. Methods & members:</dd>
<dd>
<tt>get ( inst class -- 2byte )</tt></dd>
<dd>
<tt>set ( 2byte inst class -- )</tt></dd>
<dd>
<tt>.payload ( inst class -- addr ) </tt>member holds instance's
value</dd>
<dt>
<b><tt>c-4byte</tt></b> </dt>
<dd>
Primitive class derived from <tt>object</tt>, with a 4-byte (cell) payload.
Set and get methods perform correct width fetch and store. Methods &
members:</dd>
<dd>
<tt>get ( inst class -- x )</tt></dd>
<dd>
<tt>set ( x inst class -- )</tt></dd>
<dd>
<tt>.payload ( inst class -- addr ) </tt>member holds instance's
value</dd>
<dt>
<b><tt>c-ptr</tt></b></dt>
<dd>
Base class derived from <tt>object</tt> for pointers to non-object types.
This class is not complete by itself: several methods depend on a derived
class definition of <tt>@size</tt>. Methods & members:</dd>
<dd>
<tt>.addr ( inst class -- a-addr )</tt> member variable - holds
the pointer address</dd>
<dd>
<tt>get-ptr ( inst class -- ptr )</tt></dd>
<dd>
<tt>set-ptr ( ptr inst class -- )</tt></dd>
<dd>
<tt>inc-ptr ( inst class -- )</tt> Adds @size to pointer address</dd>
<dd>
<tt>dec-ptr ( inst class -- )</tt> Subtracts @size from pointer
address</dd>
<dd>
<tt>index-ptr ( i inst class -- )</tt> Adds i*@size to pointer
address</dd>
<dt>
<b><tt>c-bytePtr</tt></b></dt>
<dd>
Pointer to byte derived from c-ptr. Methods & members:</dd>
<dd>
<tt>@size ( inst class -- size )</tt> Push size of the pointed-to
thing</dd>
<dd>
<tt>get ( inst class -- c ) </tt>Fetch the pointer's
referent byte</dd>
<dd>
<tt>set ( c inst class -- ) </tt>Store c at the pointer address</dd>
<dt>
<b><tt>c-2bytePtr</tt></b></dt>
<dd>
Pointer to double byte derived from c-ptr. Methods & members:</dd>
<dd>
<tt>@size ( inst class -- size )</tt> Push size of the pointed-to
thing</dd>
<dd>
<tt>get ( inst class -- x ) </tt>Fetch the pointer's
referent 2byte</dd>
<dd>
<tt>set ( x inst class -- )</tt> Store 2byte x at the pointer
address</dd>
<dt>
<b><tt>c-cellPtr</tt></b></dt>
<dd>
Pointer to cell derived from c-ptr. Methods & members:</dd>
<dd>
<tt>@size ( inst class -- size )</tt> Push size of the pointed-to
thing</dd>
<dd>
<tt>get ( inst class -- x ) </tt>Fetch the pointer's
referent cell</dd>
<dd>
<tt>set ( x inst class -- )</tt> Storex at the pointer address</dd>
<dt>
<b><tt>c-string</tt></b> </dt>
<dd>
Counted string (thin)</dd>
</dl>
</td>
</tr>
</table>
<h2>
<hr WIDTH="100%"></h2>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<dl>
<h2>
<a NAME="extras"></a>Ficl extras</h2>
<h3>
Number syntax</h3>
You can precede a number with "0x", as in C, and it will be interpreted
as a hex value regardless of the value of <tt>BASE</tt>. Example:
<dl><tt>ok> decimal 123 . cr</tt>
<br><tt>123 </tt>
<br><tt>ok> 0x123 . cr</tt>
<br><tt>291 </tt></dl>
<h3>
Search order words</h3>
Note: Ficl resets the search order whenever it does <tt>ABORT</tt>. If
you don't like this behavior, just comment out the dictResetSearchOrder()
lines in ficlExec().
<br>
<dt>
<a NAME="tosearch"></a><tt>>search ( wid -- )</tt></dt>
<dd>
Push <tt>wid</tt> onto the search order. Many of the other search order
words are written in terms of the <tt>SEARCH></tt> and <tt>>SEARCH</tt>
primitives.</dd>
<dt>
<a NAME="searchfrom"></a><tt>search> ( -- wid )</tt></dt>
<dd>
Pop <tt>wid</tt> off the search order</dd>
<dt>
<a NAME="ficlsetcurrent"></a><tt>ficl-set-current ( wid --
old-wid )</tt></dt>
<dd>
Set wid as compile wordlist, leaving the previous compile wordlist on the
stack</dd>
<dt>
<a NAME="ficlvocabulary"></a><tt>ficl-vocabulary ( nBins "name"
-- )</tt></dt>
<dd>
Creates a <tt>ficl-wordlist</tt> with the specified number of hash table
bins, binds it to the name, and associates the semantics of <tt>vocabulary</tt>
with it (replaces the top wid in the search order list with its own wid
when executed)</dd>
<dt>
<a NAME="ficlwordlist"></a><tt>ficl-wordlist ( nBins -- wid
)</tt></dt>
<dd>
Creates a <font face="">wordlist</font> with the specified number of hash
table bins, and leaves the address of the wordlist on the stack. A <tt>ficl-wordlist</tt>
behaves exactly as a regular wordlist, but it may search faster depending
on the number of bins chosen and the number of words it contains at search
time. As implemented in ficl, a <tt>wordlist</tt> is single threaded by
default. </dd>
<dt>
<a NAME="ficlforgetwid"></a><tt>forget-wid ( wid -- )</tt></dt>
<dd>
Iterates through the specified wordlist and unlinks all definitions whose
xt addresses are greater than or equal to the value of <tt>HERE</tt>, the
dictionary fill pointer. </dd>
<dt>
<a NAME="ficlhide"></a><tt>hide ( -- current-wid-was )</tt></dt>
<dd>
Push the <tt>hidden</tt> wordlist onto the search order, and set it as
the current compile wordlist (unsing <tt>ficl-set-current</tt>). Leaves
the previous compile wordlist ID. I use this word to hide implementation
factor words that have low reuse potential so that they don't clutter the
default wordlist. To undo the effect of hide, execute <b><tt>previous
set-current</tt></b></dd>
<dt>
<a NAME="ficlhidden"></a><tt>hidden ( -- wid )</tt></dt>
<dd>
Wordlist for storing implementation factors of ficl provided words. To
see what's in there, try: <b><tt>hide words previous set-current</tt></b></dd>
<dt>
<tt>wid-set-super ( wid -- )</tt></dt>
<dd>
Ficl wordlists have a parent wordlist pointer that is not specified in
standard Forth. Ficl initializes this pointer to NULL whenever it creates
a wordlist, so it ordinarily has no effect. This word sets the parent pointer
to the wordlist specified on the top of the stack. Ficl's implementation
of <tt>SEARCH-WORDLIST</tt> will chain backward through the parent link
of the wordlist when searching. This simplifies Ficl's object model in
that the search order does not need to reflect an object's class hierarchy
when searching for a method. It is possible to implement Ficl object syntax
in strict ANS Forth, but method finders need to manipulate the search order
explicitly.</dd>
</dl>
<h3>
User variables</h3>
<dl>
<dt>
<tt>user ( -- ) name</tt></dt>
<dd>
Create a user variable with the given name. User variables are virtual
machine local. Each VM allocates a fixed amount of storage for them. You
can change the maximum number of user variables allowed by defining FICL_USER_CELLS
on your compiiler's command line. Default is 16 user cells.</dd>
</dl>
<h3>
Miscellaneous</h3>
<dl>
<dt>
<tt>-roll ( xu xu-1 ... x0 u -- x0 xu-1 ... x1 ) </tt></dt>
<dd>
Rotate u+1 items on top of the stack after removing u. Rotation is in the
opposite sense to <tt>ROLL</tt></dd>
</dl>
<dl>
<dt>
<a NAME="minusrot"></a><tt>-rot ( a b c -- c a b )</tt></dt>
<dd>
Rotate the top three stack entries, moving the top of stack to third place.
I like to think of this as <tt>1<sup>1</sup>/<sub>2</sub>swap</tt> because
it's good for tucking a single cell value behind a cell-pair (like an object). </dd>
</dl>
<dl>
<dt>
<tt>.env ( -- )</tt></dt>
<dd>
List all environment variables of the system</dd>
<dt>
<tt>.hash ( -- )</tt></dt>
<dd>
List hash table performance statistics of the wordlist that's first in
the search order</dd>
<dt>
<tt>.ver ( -- )</tt></dt>
<dd>
Display ficl version ID</dd>
<dt>
<tt>>name ( xt -- c-addr u )</tt></dt>
<dd>
Convert a word's execution token into the address and length of its name</dd>
<dt>
<tt>body> ( a-addr -- xt )</tt></dt>
<dd>
Reverses the effect of <tt>CORE</tt> word <tt>>body </tt>(converts a parameter
field address to an execution token)</dd>
<dt>
<tt>compile-only</tt></dt>
<dd>
Mark the most recently defined word as being executable only while in compile
state. Many <tt>immediate</tt> words have this property.</dd>
<dt>
<tt>empty ( -- )</tt> </dt>
<dd>
Empty the parameter stack</dd>
<dt>
<tt>endif</tt></dt>
<dd>
Synonym for <tt>THEN</tt></dd>
<dt>
<tt>parse-word ( <spaces>name -- c-addr u )</tt></dt>
<dd>
Skip leading spaces and parse name delimited by a space. c-addr is the
address within the input buffer and u is the length of the selected string.
If the parse area is empty, the resulting string has a zero length. (From
the Standard)</dd>
<dt>
<tt>w@ ( addr -- x )</tt></dt>
<dd>
Fetch a 16 bit quantity from the specified address</dd>
<dt>
<tt>w! ( x addr -- )</tt></dt>
<dd>
Store a 16 bit quantity to the specified address (the low 16 bits of the
given value)</dd>
<dt>
<tt>x. ( x -- )</tt></dt>
<dd>
Pop and display the value in hex format, regardless of the current value
of <tt>BASE</tt></dd>
</dl>
<h3>
FiclWin Extras (defined in testmain.c)</h3>
<dl>
<dt>
<tt>break ( -- )</tt></dt>
<dd>
Does nothing - just a handy place to set a debugger breakpoint</dd>
<dt>
<tt>cd ( "directory-name<newline>" --
)</tt></dt>
<dd>
Executes the Win32 chdir() function, changing the program's logged directory.</dd>
<dt>
<a NAME="clock"></a><tt>clock ( -- now )</tt></dt>
<dd>
Wrapper for the ANSI C clock() function. Returns the number of clock ticks
elapsed since process start.</dd>
<dt>
<a NAME="clockspersec"></a><tt>clocks/sec ( -- clocks_per_sec
)</tt></dt>
<dd>
Pushes the number of ticks in a second as returned by <tt>clock</tt></dd>
<dt>
<a NAME="ficlload"></a><tt>load ( "filename<newline>"
-- )</tt></dt>
<dd>
Opens the Forth source file specified and loads it one line at a time,
like <tt>INCLUDED (FILE)</tt></dd>
<dt>
<tt>pwd ( -- )</tt></dt>
<dd>
Prints the current working directory as set by <tt>cd</tt></dd>
<dt>
<tt>system ( "command<newline>" -- )</tt></dt>
<dd>
Issues a command to a shell; implemented with the Win32 system() call.</dd>
<dt>
<tt>spewhash ( "filename<newline>" -- )</tt></dt>
<dd>
Dumps all threads of the current compilation wordlist to the specified
text file. This was useful when I thought there might be some point in
attempting to optimize the hash function. I no longer harbor those illusions.</dd>
<h3>
FiclWin Exclusives (no source provided)</h3>
<dt>
<tt>!oreg ( c -- )</tt></dt>
<dd>
Set the value of the simulated LED register as specified (0..255)</dd>
<dt>
<tt>@ireg ( -- c )</tt></dt>
<dd>
Gets the value of the simulated switch block (0..255)</dd>
<dt>
<tt>!dac ( c -- )</tt></dt>
<dd>
Sets the value of the bargraph control as specified. Valid values range
from 0..255</dd>
<dt>
<tt>@adc ( -- c )</tt></dt>
<dd>
Fetches the current position of the slider control. Range is 0..255</dd>
<dt>
<tt>status" ( "ccc<quote>" -- )</tt></dt>
<dd>
Set the mainframe window's status line to the text specified, up to the
first trailing quote character.</dd>
<dt>
<a NAME="ficlms"></a><tt><a href="http://www.taygeta.com/forth/dpans10.htm#10.6.2.1905">ms</a>
( u -- )</tt></dt>
<dd>
Causes the running virtual machine to sleep() for the number of milliseconds
specified by the top-of-stack value.</dd>
</dl>
</td>
</tr>
</table>
<h2>
<hr WIDTH="100%"><a NAME="ansinfo"></a>ANS Required Information</h2>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td><b>ANS Forth System</b>
<br><b>Providing names from the Core Extensions word set </b>
<br><b>Providing the Exception word set</b>
<br><b>Providing names from the Exception Extensions word set</b>
<br><b>Providing the Locals word set </b>
<br><b>Providing the Locals Extensions word set </b>
<br><b>Providing the Memory Allocation word set</b>
<br><b>Providing the Programming-Tools word set</b>
<br><b>Providing names from the Programming-Tools Extensions word set</b>
<br><b>Providing the Search-Order word set</b>
<br><b>Providing the Search-Order Extensions word set</b>
<h3>
Implementation-defined Options</h3>
The implementation-defined items in the following list represent characteristics
and choices left to the discretion of the implementor, provided that the
requirements of the Standard are met. A system shall document the values
for, or behaviors of, each item.
<ul>
<li>
<b>aligned address requirements (3.1.3.3 Addresses);</b> </li>
<br><font color="#000000">System dependent. You can change the default
address alignment by defining FICL_ALIGN on your compiler's command line.
The default value is set to 2 in sysdep.h. This causes dictionary entries
and <tt>ALIGN</tt> and <tt>ALIGNED</tt> to align on 4 byte boundaries.
To align on <b>2<sup><font face="">n</font></sup></b> byte boundaries,
set FICL_ALIGN to <b>n</b>. </font>
<li>
<b>behavior of 6.1.1320 EMIT for non-graphic characters</b>; </li>
<br><font color="#000000">Depends on target system, C runtime library,
and your implementation of ficlTextOut().</font>
<li>
<b>character editing of 6.1.0695 ACCEPT and 6.2.1390 EXPECT</b>; </li>
<br><font color="#000000">None implemented in the versions supplied in
words.c. Because ficlExec() is supplied a text buffer externally, it's
up to your system to define how that buffer will be obtained.</font>
<li>
<b>character set (3.1.2 Character types, 6.1.1320 EMIT, 6.1.1750 KEY)</b>; </li>
<br><font color="#000000">Depends on target system and implementation of
ficlTextOut()</font>
<li>
<b>character-aligned address requirements (3.1.3.3 Addresses)</b>; </li>
<br><font color="#000000">Ficl characters are one byte each. There are
no alignment requirements.</font>
<li>
<b>character-set-extensions matching characteristics (3.4.2 Finding definition
n<font color="#000000">ames)</font></b><font color="#000000">; </font></li>
<br><font color="#000000">No special processing is performed on characters
beyond case-folding. Therefore, extended characters will not match their
unaccented counterparts.</font>
<li>
<b>conditions under which control characters match a space delimiter (3.4.1.1
Delimiters)</b>;<font color="#FF6666"> </font></li>
<br><font color="#000000">Ficl uses the Standard C function isspace() to
distinguish space characters. The rest is up to your library vendor.</font>
<li>
<b>format of the control-flow stack (3.2.3.2 Control-flow stack)</b>; </li>
<br><font color="#000000">Uses the data stack</font>
<li>
<b>conversion of digits larger than thirty-five (3.2.1.2 Digit conversion)</b>; </li>
<br><font color="#000000">The maximum supported value of <tt>BASE</tt>
is 36. Ficl will assertion fail in function ltoa of vm.c if the base is
found to be larger than 36 or smaller than 2. There will be no effect if
NDEBUG is defined</font>, however, other than possibly unexpected behavior.
<li>
<b>display after input terminates in 6.1.0695 ACCEPT and 6.2.1390 EXPECT</b>; </li>
<br><font color="#000000">Target system dependent</font>
<li>
<b>exception abort sequence (as in 6.1.0680 ABORT")</b>; </li>
<br><font color="#000000">Does <tt>ABORT</tt></font>
<li>
<b>input line terminator (3.2.4.1 User input device)</b>;<font color="#FF0000"> </font></li>
<br><font color="#000000">Target system dependent (implementation of outer
loop that calls ficlExec)</font>
<li>
<b>maximum size of a counted string, in characters (3.1.3.4 Counted strings,
6.1.2450 WORD)</b>; </li>
<br><font color="#000000">255</font>
<li>
<b>maximum size of a parsed string (3.4.1 Parsing)</b>; </li>
<br>Limited by available memory and the maximum unsigned value that can
fit in a CELL (2<sup>32</sup>-1).
<li>
<b>maximum size of a definition name, in characters (3.3.1.2 Definition
names)</b>; </li>
<br><font color="#000000">Ficl stores the first 31 characters of a definition
name.</font>
<li>
<b>maximum string length for 6.1.1345 ENVIRONMENT?, in characters</b>; </li>
<br><font color="#000000">Same as maximum definition name length</font>
<li>
<b>method of selecting 3.2.4.1 User input device</b>; </li>
<br>None supported. This is up to the target system
<li>
<b>method of selecting 3.2.4.2 User output device</b>; </li>
<br>None supported. This is up to the target system
<li>
<b>methods of dictionary compilation (3.3 The Forth dictionary)</b>; </li>
<li>
<b>number of bits in one address unit (3.1.3.3 Addresses)</b>; </li>
<br><font color="#000000">Target system dependent. Ficl generally supports
processors that can address 8 bit quantities, but there is no dependency
that I'm aware of.</font>
<li>
<b>number representation and arithmetic (3.2.1.1 Internal number representation)</b>; </li>
<br>System dependent. Ficl represents a CELL internally as a union that
can hold INT32 (a signed 32 bit scalar value), UNS32 (32 bits unsigned),
and an untyped pointer. No specific byte ordering is assumed.
<li>
<b>ranges for n, +n, u, d, +d, and ud (3.1.3 Single-cell types, 3.1.4 Cell-pair
types)</b>; </li>
<br>Assuming a 32 bit implementation, range for signed single-cell values
is -2<sup>31</sup>..2<sup>31</sup>-1. Range for unsigned single cell values
is 0..2<sup>32</sup>-1. Range for signed double-cell values is -2<sup>63</sup>..2<sup>63</sup>-1.
Range for unsigned single cell values is 0..2<sup>64</sup>-1.
<li>
<b>read-only data-space regions (3.3.3 Data space)</b>;</li>
<br>None
<li>
<b>size of buffer at 6.1.2450 WORD (3.3.3.6 Other transient regions)</b>; </li>
<br>Default is 255. Depends on the setting of nPAD in ficl.h.
<li>
<b>size of one cell in address units (3.1.3 Single-cell types)</b>; </li>
<br><font color="#000000">System dependent, generally four.</font>
<li>
<b>size of one character in address units (3.1.2 Character types)</b>; </li>
<br><font color="#000000">System dependent, generally one.</font>
<li>
<b>size of the keyboard terminal input buffer (3.3.3.5 Input buffers)</b>; </li>
<br><font color="#000000">This buffer is supplied by the host program.
Ficl imposes no practical limit.</font>
<li>
<b>size of the pictured numeric output string buffer (3.3.3.6 Other transient
regions)</b>; </li>
<br>Default is 255 characters. Depends on the setting of nPAD in ficl.h.
<li>
<b>size of the scratch area whose address is returned by 6.2.2000 PAD (3.3.3.6
Other transient regions)</b>; </li>
<br>Not presently supported
<li>
<b>system case-sensitivity characteristics (3.4.2 Finding definition names)</b>; </li>
<br><font color="#000000">Ficl is not case sensitive</font>
<li>
<b>system prompt (3.4 The Forth text interpreter, 6.1.2050 QUIT)</b>; </li>
<br><font color="#000000">"ok>"</font>
<li>
<b>type of division rounding (3.2.2.1 Integer division, 6.1.0100 */, 6.1.0110
*/MOD, 6.1.0230 /, 6.1.0240 /MOD, 6.1.1890 MOD)</b>; </li>
<br><font color="#000000">Symmetric</font>
<li>
<b>values of 6.1.2250 STATE when true</b>; </li>
<br><font color="#000000">One (no others)</font>
<li>
<b>values returned after arithmetic overflow (3.2.2.2 Other integer operations)</b>; </li>
<br>System dependent. Ficl makes no special checks for overflow.
<li>
<b>whether the current definition can be found after 6.1.1250 DOES> (6.1.0450
:)</b>. </li>
<br><font color="#000000">No. Definitions are unsmudged after ; only, and
only then if no control structure matching problems have been detected.</font></ul>
<h3>
Ambiguous Conditions</h3>
A system shall document the system action taken upon each of the general
or specific ambiguous conditions identified in this Standard. See 3.4.4
Possible actions on an ambiguous condition.
<p>The following general ambiguous conditions could occur because of a
combination of factors:
<ul>
<dl>
<li>
<b>a name is neither a valid definition name nor a valid number during
text interpretation (3.4 The Forth text interpreter)</b>; </li>
<br><font color="#000000">Ficl does <tt>ABORT</tt> and prints the name
followed by " not found".</font>
<li>
<b>a definition name exceeded the maximum length allowed (3.3.1.2 Definition
names)</b>; </li>
<br><font color="#000000">Ficl stores the first 31 characters of the definition
name, and uses all characters of the name in computing its hash code. The
actual length of the name, up to 255 characters, is stored in the definition's
length field.</font>
<li>
<b>addressing a region not listed in 3.3.3 Data Space</b>; </li>
<br><font color="#000000">No problem: all addresses in ficl are absolute.
You can reach any 32 bit address in Ficl's address space.</font>
<li>
<b>argument type incompatible with specified input parameter, e.g., passing
a flag to a word expecting an n (3.1 Data types)</b>; </li>
<br><font color="#000000">Ficl makes no check for argument type compatibility.
Effects of a mismatch vary widely depending on the specific problem and
operands.</font></dl>
<li>
<b>attempting to obtain the execution token, (e.g., with 6.1.0070 ', 6.1.1550
FIND, etc.) of a definition with undefined interpretation semantics</b>; </li>
<br><font color="#000000">Ficl returns a valid token, but the result of
executing that token while interpreting may be undesirable.</font>
<li>
<b>dividing by zero (6.1.0100 */, 6.1.0110 */MOD, 6.1.0230 /, 6.1.0240
/MOD, 6.1.1561 FM/MOD, 6.1.1890 MOD, 6.1.2214 SM/REM, 6.1.2370 UM/MOD,
8.6.1.1820 M*/)</b>;</li>
<br><font color="#000000">Results are target procesor dependent. Generally,
Ficl makes no check for divide-by-zero. The target processor will probably
throw an exception.</font>
<li>
<b>insufficient data-stack space or return-stack space (stack overflow)</b>; </li>
<br><font color="#000000">With FICL_ROBUST (sysdep.h) set >= 2, most parameter
stack operations are checked for underflow and overflow. Ficl does not
check the return stack.</font>
<li>
<b>insufficient space for loop-control parameters</b>; </li>
<br><font color="#000000">No check - Evil results.</font>
<li>
<b>insufficient space in the dictionary</b>; </li>
<br><font color="#000000">Ficl generates an error message if the dictionary
is too full to create a definition header. It checks <tt>ALLOT</tt> as
well, but it is possible to make an unchecked allocation request that overflows
the dictionary.</font>
<li>
<b>interpreting a word with undefined interpretation semantics</b>; </li>
<br><font color="#000000">Ficl protects all ANS Forth words with undefined
interpretation semantics from being executed while in interpret state.
It is possible to defeat this protection using ' (tick) and <tt>EXECUTE</tt>,
though.</font>
<li>
<b>modifying the contents of the input buffer or a string literal (3.3.3.4
Text-literal regions, 3.3.3.5 Input buffers)</b>; </li>
<br><font color="#000000">Varies depending on the nature of the buffer.
The input buffer is supplied by ficl's host function, and may reside in
read-only memory. If so, writing the input buffer can ganerate an exception.
String literals are stored in the dictionary, and are writable.</font>
<li>
<b>overflow of a pictured numeric output string</b>;</li>
<br>In the unlikely event you are able to construct a pictured numeric
string of more than 255 characters, the system will be corrupted unpredictably.
The buffer area that holds pictured numeric output is at the end of the
virtual machine. Whatever is mapped after the offending VM in memory will
be trashed, along with the heap structures that contain it.
<li>
<b>parsed string overflow</b>;</li>
<br>Ficl does not copy parsed strings unless asked to. Ordinarily, a string
parsed from the input buffer during normal interpretation is left in-place,
so there is no possibility of overflow. If you ask to parse a string into
the dictionary, as in <tt>SLITERAL</tt>, you need to have enough room for
the string, otherwise bad things may happen. This is not usually a problem.
<li>
<b>producing a result out of range, e.g., multiplication (using *) results
in a value too big to be represented by a single-cell integer (6.1.0090
*, 6.1.0100 */, 6.1.0110 */MOD, 6.1.0570 >NUMBER, 6.1.1561 FM/MOD, 6.1.2214
SM/REM, 6.1.2370 UM/MOD, 6.2.0970 CONVERT, 8.6.1.1820 M*/)</b>; </li>
<br><font color="#000000">Value will be truncated</font>
<li>
<b>reading from an empty data stack or return stack (stack underflow)</b>; </li>
<br><font color="#000000">Most stack underflows are detected and prevented
if FICL_ROBUST (sysdep.h) is set to 2 or greater. Otherwise, the stack
pointer and size are likely to be trashed.</font>
<li>
<b>unexpected end of input buffer, resulting in an attempt to use a zero-length
string as a name</b>; </li>
<br><font color="#000000">Ficl returns for a new input buffer until a non-empty
one is supplied.</font></ul>
The following specific ambiguous conditions are noted in the glossary entries
of the relevant words:
<ul>
<li>
<b>>IN greater than size of input buffer (3.4.1 Parsing)</b></li>
<br>Bad Things occur - unpredictable bacause the input buffer is supplied
by the host program's outer loop.
<li>
<b>6.1.2120 RECURSE appears after 6.1.1250 DOES></b></li>
<br>It finds the address of the definition before <tt>DOES></tt>
<li>
<b>argument input source different than current input source for 6.2.2148
RESTORE-INPUT</b></li>
<br>Not implemented
<li>
<b>data space containing definitions is de-allocated (3.3.3.2 Contiguous
regions)</b></li>
<br>This is OK until the cells are overwritten with something else. The
dictionary maintains a hash table, and the table must be updated in order
to de-allocate words without corruption.
<li>
<b>data space read/write with incorrect alignment (3.3.3.1 Address alignment)</b></li>
<br>Target processor dependent. Consequences include: none (Intel), address
error exception (68K).
<li>
<b>data-space pointer not properly aligned (6.1.0150 ,, 6.1.0860 C,)</b></li>
<br>See above on data space read/write alignment
<li>
<b>less than u+2 stack items (6.2.2030 PICK, 6.2.2150 ROLL)</b></li>
<br>Ficl detects a stack underflow and reports it, executing <tt>ABORT,</tt>
as long as FICL_ROBUST is two or larger.
<li>
<b>loop-control parameters not available ( 6.1.0140 +LOOP, 6.1.1680 I,
6.1.1730 J, 6.1.1760 LEAVE, 6.1.1800 LOOP, 6.1.2380 UNLOOP)</b></li>
<br>Loop initiation words are responsible for checking the stack and guaranteeing
that the control parameters are pushed. Any underflows will be detected
early if FICL_ROBUST is set to two or greater. Note however that Ficl only
checks for return stack underflows at the end of each line of text.
<li>
<b>most recent definition does not have a name (6.1.1710 IMMEDIATE)</b></li>
<br>No problem.
<li>
<b>name not defined by 6.2.2405 VALUE used by 6.2.2295 TO</b></li>
<br>Ficl's version of <tt>TO</tt> works correctly with <tt>VALUE</tt>s,
<tt>CONSTANT</tt>s
and <tt>VARIABLE</tt>s.
<li>
<b>name not found (6.1.0070 ', 6.1.2033 POSTPONE, 6.1.2510 ['], 6.2.2530
[COMPILE])</b></li>
<br>Ficl prints an error message and does <tt>ABORT</tt>
<li>
<b>parameters are not of the same type (6.1.1240 DO, 6.2.0620 ?DO, 6.2.2440
WITHIN)</b></li>
<br>No check. Results vary depending on the specific problem.
<li>
<b>6.1.2033 POSTPONE or 6.2.2530 [COMPILE] applied to 6.2.2295 TO</b></li>
<br>The word is postponed correctly.
<li>
<b>string longer than a counted string returned by 6.1.2450 WORD</b></li>
<br>Ficl stores the first FICL_STRING_MAX-1 chars in the destination buffer.
(The extra character is the trailing space required by the standard. Yuck.)
<li>
<b>u greater than or equal to the number of bits in a cell (6.1.1805 LSHIFT,
6.1.2162 RSHIFT)</b></li>
<br>Depends on target process or and C runtime library implementations
of the << and >> operators on unsigned values. For I386, the processor
appears to shift modulo the number of bits in a cell.
<li>
<b>word not defined via 6.1.1000 CREATE (6.1.0550 >BODY, 6.1.1250 DOES>)</b></li>
<br><b>words improperly used outside 6.1.0490 <# and 6.1.0040 #> (6.1.0030
#, 6.1.0050 #S, 6.1.1670 HOLD, 6.1.2210 SIGN)</b>
<br>Don't. <tt>CREATE</tt> reserves a field in words it builds for <tt>DOES></tt>to
fill in. If you use <tt>DOES></tt> on a word not made by <tt>CREATE</tt>,
it will overwrite the first cell of its parameter area. That's probably
not what you want. Likewise, pictured numeric words assume that there is
a string under construction in the VM's scratch buffer. If that's not the
case, results may be unpleasant.</ul>
<h3>
Locals Implementation-defined options</h3>
<ul>
<li>
<b>maximum number of locals in a definition (13.3.3 Processing locals,
13.6.2.1795 LOCALS|)</b></li>
<br>Default is 16. Change by redefining FICL_MAX_LOCALS, defined in sysdep.h</ul>
<h3>
Locals Ambiguous conditions</h3>
<ul>
<li>
<b>executing a named local while in interpretation state (13.6.1.0086 (LOCAL))</b></li>
<br>Locals can be found in interpretation state while in the context of
a definition under construction. Under these circumstances, locals behave
correctly. Locals are not visible at all outside the scope of a definition.
<li>
<b>name not defined by VALUE or LOCAL (13.6.1.2295 TO)</b></li>
<br>See the CORE ambiguous conditions, above (no change)</ul>
<h3>
Programming Tools Implementation-defined options</h3>
<ul>
<li>
<b>source and format of display by 15.6.1.2194 SEE</b></li>
<br>SEE de-compiles definitions from the dictionary. Because Ficl words
are threaded by their header addresses, it is very straightforward to print
the name and other characteristics of words in a definition. Primitives
are so noted. Colon definitions are decompiled, but branch target labels
are not reconstructed. Literals and string literals are so noted, and their
contents displayed.</ul>
<h3>
Search Order Implementation-defined options</h3>
<ul>
<li>
<b>maximum number of word lists in the search order (16.3.3 Finding definition
names, 16.6.1.2197 SET-ORDER)</b> </li>
<br>Defaults to 16. Can be changed by redefining FICL_DEFAULT_VOCS, declared
in sysdep.h
<li>
<b>minimum search order (16.6.1.2197 SET-ORDER, 16.6.2.1965 ONLY)</b> </li>
<br>Equivalent to <tt>FORTH-WORDLIST 1 SET-ORDER</tt></ul>
<h3>
Search Order Ambiguous conditions</h3>
<ul>
<li>
<b>changing the compilation word list (16.3.3 Finding definition names)</b></li>
<br>Ficl stores a link to the current definition independently of the compile
wordlist while it is being defined, and links it into the compile wordlist
only after the definition completes successfully. Changing the compile
wordlist mid-definition will cause the definition to link into the <i>new</i>
compile wordlist.
<li>
<b>search order empty (16.6.2.2037 PREVIOUS)</b></li>
<br>Ficl prints an error message if the search order underflows, and resets
the order to its default state.
<li>
<b>too many word lists in search order (16.6.2.0715 ALSO)</b></li>
<br>Ficl prints an error message if the search order overflows, and resets
the order to its default state.</ul>
</td>
</tr>
</table>
<h2>
<hr WIDTH="100%"><a NAME="links"></a>For more information</h2>
<ul>
<li>
<a href="http://www.taygeta.com/ficl.html">Web home of ficl</a></li>
<li>
<b><a href="ftp://ftp.taygeta.com/pub/Forth/Compilers/native/misc/ficl203/ficl203.zip">Download
ficl 2.03</a></b></li>
<li>
<a href="ficlddj.pdf">Manuscript of Ficl article for January 1999 Dr. Dobb's
Journal</a></li>
<li>
<a href="jwsforml.pdf">1998 FORML Conference paper</a></li>
<li>
<a href="http://www.taygeta.com/forthlit.html">Forth literature</a></li>
<ul>
<li>
<a href="http://www.softsynth.com/pforth/pf_tut.htm">Phil Burk's Forth
Tutorial</a></li>
<li>
<a href="http://www.taygeta.com/forth/dpans.html">Draft Proposed American
National Standard for Forth</a></li>
</ul>
<li>
<a href="http://www.forth.org">Forth Interest Group</a></li>
</ul>
<h2>
<a NAME="includesficl"></a>Some software that uses ficl</h2>
<ul>
<li>
<a href="http://www.freebsd.org/">FreeBSD</a> boot loader</li>
<li>
<a href="http://www.pagesz.net/~sessoms/debuffer/">Palm Pilot Debuffer</a>
(Eric Sessoms)</li>
<li>
<a href="http://www.swcp.com/~jchavez/osmond.html">Mac PC Board Layout
tool</a> (J Chavez)</li>
<li>
<a href="http://www.netcomsystems.com">NetCom Systems</a> ML7710 (Martin
Usher)</li>
</ul>
<h2>
<hr WIDTH="100%"></h2>
<table BORDER=0 CELLSPACING=3 COLS=1 WIDTH="600" >
<tr>
<td>
<h2>
<a NAME="lawyerbait"></a>DISCLAIMER OF WARRANTY and LICENSE</h2>
<i>Ficl is freeware. Use it in any way that you like, with the understanding
that the code is not supported.</i>
<p>Any third party may reproduce, distribute, or modify the ficl software
code or any derivative works thereof without any compensation or license,
provided that the original author information and this disclaimer text
are retained in the source code files. The ficl software code is provided
on an "as is" basis without warranty of any kind, including, without limitation,
the implied warranties of merchantability and fitness for a particular
purpose and their equivalents under the laws of any jurisdiction.
<p>I am interested in hearing from anyone who uses ficl. If you have a
problem, a success story, a defect, an enhancement request, or if you would
like to contribute to the ficl release (yay!), please <a href="mailto:john_sadler@alum.mit.edu">send
me email</a>. </td>
</tr>
</table>
</body>
</html>