1: \ definitions needed for interpreter only
2:
3: \ Copyright (C) 1995-2000,2004,2005,2007,2009,2010 Free Software Foundation, Inc.
4:
5: \ This file is part of Gforth.
6:
7: \ Gforth is free software; you can redistribute it and/or
8: \ modify it under the terms of the GNU General Public License
9: \ as published by the Free Software Foundation, either version 3
10: \ of the License, or (at your option) any later version.
11:
12: \ This program is distributed in the hope that it will be useful,
13: \ but WITHOUT ANY WARRANTY; without even the implied warranty of
14: \ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15: \ GNU General Public License for more details.
16:
17: \ You should have received a copy of the GNU General Public License
18: \ along with this program. If not, see http://www.gnu.org/licenses/.
19:
20: \ \ Revision-Log
21:
22: \ put in seperate file 14sep97jaw
23:
24: \ \ input stream primitives 23feb93py
25:
26: require ./basics.fs \ bounds decimal hex ...
27: require ./io.fs \ type ...
28: require ./nio.fs \ . <# ...
29: require ./errore.fs \ .error ...
30: require kernel/version.fs \ version-string
31:
32: has? new-input 0= [IF]
33: : tib ( -- c-addr ) \ core-ext t-i-b
34: \G @i{c-addr} is the address of the Terminal Input Buffer.
35: \G OBSOLESCENT: @code{source} superceeds the function of this word.
36: >tib @ ;
37:
38: Defer source ( -- c-addr u ) \ core
39: \ used by dodefer:, must be defer
40: \G @i{c-addr} is the address of the input buffer and @i{u} is the
41: \G number of characters in it.
42:
43: : (source) ( -- c-addr u )
44: tib #tib @ ;
45: ' (source) IS source
46: [THEN]
47:
48: : (word) ( addr1 n1 char -- addr2 n2 )
49: dup >r skip 2dup r> scan nip - ;
50:
51: \ (word) should fold white spaces
52: \ this is what (parse-white) does
53:
54: \ parse 23feb93py
55:
56: : parse ( char "ccc<char>" -- c-addr u ) \ core-ext
57: \G Parse @i{ccc}, delimited by @i{char}, in the parse
58: \G area. @i{c-addr u} specifies the parsed string within the
59: \G parse area. If the parse area was empty, @i{u} is 0.
60: >r source >in @ over min /string ( c-addr1 u1 )
61: over swap r> scan >r
62: over - dup r> IF 1+ THEN >in +!
63: [ has? new-input [IF] ]
64: 2dup input-lexeme!
65: [ [THEN] ] ;
66:
67: \ name 13feb93py
68:
69: [IFUNDEF] (name) \ name might be a primitive
70:
71: : (name) ( -- c-addr count ) \ gforth
72: source 2dup >r >r >in @ /string (parse-white)
73: [ has? new-input [IF] ]
74: 2dup input-lexeme!
75: [ [THEN] ]
76: 2dup + r> - 1+ r> min >in ! ;
77: \ name count ;
78: [THEN]
79:
80: : name-too-short? ( c-addr u -- c-addr u )
81: dup 0= -&16 and throw ;
82:
83: : name-too-long? ( c-addr u -- c-addr u )
84: dup lcount-mask u> -&19 and throw ;
85:
86: \ \ Number parsing 23feb93py
87:
88: \ (number?) number 23feb93py
89:
90: hex
91: const Create bases 0A , 10 , 2 , 0A ,
92: \ 10 16 2 10
93:
94: \ !! protect BASE saving wrapper against exceptions
95: : getbase ( addr u -- addr' u' )
96: 2dup s" 0x" string-prefix? >r
97: 2dup s" 0X" string-prefix? r> or
98: base @ &34 < and if
99: hex 2 /string
100: endif
101: over c@ [char] # - dup 4 u<
102: IF
103: cells bases + @ base ! 1 /string
104: ELSE
105: drop
106: THEN ;
107:
108: : sign? ( addr u -- addr1 u1 flag )
109: over c@ [char] - = dup >r
110: IF
111: 1 /string
112: THEN
113: r> ;
114:
115: : ?dnegate ( d1 f -- d2 )
116: if
117: dnegate
118: then ;
119:
120: has? os 0= [IF]
121: : x@+/string ( addr u -- addr' u' c )
122: over c@ >r 1 /string r> ;
123: [THEN]
124:
125: : s'>unumber? ( addr u -- ud flag )
126: \ convert string "C" or "C'" to character code
127: dup 0= if
128: false exit
129: endif
130: x@+/string 0 s" '" 2rot string-prefix? ;
131:
132: : s>unumber? ( c-addr u -- ud flag ) \ gforth
133: \G converts string c-addr u into ud, flag indicates success
134: dpl on
135: over c@ '' = if
136: 1 /string s'>unumber? exit
137: endif
138: base @ >r getbase sign?
139: over if
140: >r 0. 2swap
141: BEGIN ( d addr len )
142: dup >r >number dup
143: WHILE \ there are characters left
144: dup r> -
145: WHILE \ the last >number parsed something
146: dup 1- dpl ! over c@ [char] . =
147: WHILE \ the current char is '.'
148: 1 /string
149: REPEAT THEN \ there are unparseable characters left
150: 2drop rdrop false
151: ELSE
152: rdrop 2drop r> ?dnegate true
153: THEN
154: ELSE
155: drop 2drop 0. false THEN
156: r> base ! ;
157:
158: \ ouch, this is complicated; there must be a simpler way - anton
159: : s>number? ( addr u -- d f ) \ gforth
160: \G converts string addr u into d, flag indicates success
161: sign? >r
162: s>unumber?
163: 0= IF
164: rdrop false
165: ELSE \ no characters left, all ok
166: r> ?dnegate
167: true
168: THEN ;
169:
170: : s>number ( addr len -- d )
171: \ don't use this, there is no way to tell success
172: s>number? drop ;
173:
174: : snumber? ( c-addr u -- 0 / n -1 / d 0> )
175: s>number? 0=
176: IF
177: 2drop false EXIT
178: THEN
179: dpl @ dup 0< IF
180: nip
181: ELSE
182: 1+
183: THEN ;
184:
185: : (number?) ( string -- string 0 / n -1 / d 0> )
186: dup >r count snumber? dup if
187: rdrop
188: else
189: r> swap
190: then ;
191:
192: : number ( string -- d )
193: (number?) ?dup 0= abort" ?" 0<
194: IF
195: s>d
196: THEN ;
197:
198: \ \ Comments ( \ \G
199:
200: : ( ( compilation 'ccc<close-paren>' -- ; run-time -- ) \ thisone- core,file paren
201: \G ** this will not get annotated. The alias in glocals.fs will instead **
202: \G It does not work to use "wordset-" prefix since this file is glossed
203: \G by cross.fs which doesn't have the same functionalty as makedoc.fs
204: [char] ) parse 2drop ; immediate
205:
206: : \ ( compilation 'ccc<newline>' -- ; run-time -- ) \ thisone- core-ext,block-ext backslash
207: \G ** this will not get annotated. The alias in glocals.fs will instead **
208: \G It does not work to use "wordset-" prefix since this file is glossed
209: \G by cross.fs which doesn't have the same functionalty as makedoc.fs
210: [ has? file [IF] ]
211: blk @
212: IF
213: >in @ c/l / 1+ c/l * >in !
214: EXIT
215: THEN
216: [ [THEN] ]
217: source >in ! drop ; immediate
218:
219: : \G ( compilation 'ccc<newline>' -- ; run-time -- ) \ gforth backslash-gee
220: \G Equivalent to @code{\} but used as a tag to annotate definition
221: \G comments into documentation.
222: POSTPONE \ ; immediate
223:
224: has? ec [IF]
225: AVariable forth-wordlist
226: : find-name ( c-addr u -- nt | 0 ) \ gforth
227: \g Find the name @i{c-addr u} in the current search
228: \g order. Return its @i{nt}, if found, otherwise 0.
229: forth-wordlist (f83find) ;
230: [ELSE]
231: \ \ object oriented search list 17mar93py
232:
233: \ word list structure:
234:
235: struct
236: cell% field find-method \ xt: ( c_addr u wid -- nt )
237: cell% field reveal-method \ xt: ( nt wid -- ) \ used by dofield:, must be field
238: cell% field rehash-method \ xt: ( wid -- ) \ re-initializes a "search-data" (hashtables)
239: cell% field hash-method \ xt: ( wid -- ) \ initializes ""
240: \ \ !! what else
241: end-struct wordlist-map-struct
242:
243: struct
244: cell% field wordlist-map \ pointer to a wordlist-map-struct
245: cell% field wordlist-id \ linked list of words (for WORDS etc.)
246: cell% field wordlist-link \ link field to other wordlists
247: cell% field wordlist-extend \ wordlist extensions (eg bucket offset)
248: end-struct wordlist-struct
249:
250: has? f83headerstring [IF]
251: : f83find ( addr len wordlist -- nt / false )
252: wordlist-id @ (f83find) ;
253: [ELSE]
254: : f83find ( addr len wordlist -- nt / false )
255: wordlist-id @ (listlfind) ;
256: [THEN]
257:
258: : initvoc ( wid -- )
259: dup wordlist-map @ hash-method perform ;
260:
261: \ Search list table: find reveal
262: Create f83search ( -- wordlist-map )
263: ' f83find A, ' drop A, ' drop A, ' drop A,
264:
265: here f83search A, NIL A, NIL A, NIL A,
266: AValue forth-wordlist \ variable, will be redefined by search.fs
267:
268: AVariable lookup forth-wordlist lookup !
269: \ !! last is user and lookup?! jaw
270: AVariable current ( -- addr ) \ gforth
271: \G @code{Variable} -- holds the @i{wid} of the compilation word list.
272: AVariable voclink forth-wordlist wordlist-link voclink !
273: \ lookup AValue context ( -- addr ) \ gforth
274: Defer context ( -- addr ) \ gforth
275: \G @code{context} @code{@@} is the @i{wid} of the word list at the
276: \G top of the search order.
277:
278: ' lookup is context
279: forth-wordlist current !
280:
281: : (search-wordlist) ( addr count wid -- nt | false )
282: dup wordlist-map @ find-method perform ;
283:
284: : search-wordlist ( c-addr count wid -- 0 | xt +-1 ) \ search
285: \G Search the word list identified by @i{wid} for the definition
286: \G named by the string at @i{c-addr count}. If the definition is
287: \G not found, return 0. If the definition is found return 1 (if
288: \G the definition is immediate) or -1 (if the definition is not
289: \G immediate) together with the @i{xt}. In Gforth, the @i{xt}
290: \G returned represents the interpretation semantics. ANS Forth
291: \G does not specify clearly what @i{xt} represents.
292: (search-wordlist) dup if
293: (name>intn)
294: then ;
295:
296: : find-name ( c-addr u -- nt | 0 ) \ gforth
297: \g Find the name @i{c-addr u} in the current search
298: \g order. Return its @i{nt}, if found, otherwise 0.
299: lookup @ (search-wordlist) ;
300: [THEN]
301:
302: \ \ header, finding, ticks 17dec92py
303:
304: \ The constants are defined as 32 bits, but then erased
305: \ and overwritten by the right ones
306:
307: has? f83headerstring [IF]
308: \ to save space, Gforth EC limits words to 31 characters
309: \ also, there's no predule concept in Gforth EC
310: $80 constant alias-mask
311: $40 constant immediate-mask
312: $20 constant restrict-mask
313: $1f constant lcount-mask
314: [ELSE]
315: \ 32-bit systems cannot generate large 64-bit constant in the
316: \ cross-compiler, so we kludge it by generating a constant and then
317: \ storing the proper value into it (and that's another kludge).
318: $80000000 constant alias-mask
319: 1 bits/char 1 - lshift
320: -1 cells allot bigendian [IF] c, 0 1 cells 1- times
321: [ELSE] 0 1 cells 1- times c, [THEN]
322: $40000000 constant immediate-mask
323: 1 bits/char 2 - lshift
324: -1 cells allot bigendian [IF] c, 0 1 cells 1- times
325: [ELSE] 0 1 cells 1- times c, [THEN]
326: $20000000 constant restrict-mask
327: 1 bits/char 3 - lshift
328: -1 cells allot bigendian [IF] c, 0 1 cells 1- times
329: [ELSE] 0 1 cells 1- times c, [THEN]
330: $10000000 constant prelude-mask
331: 1 bits/char 4 - lshift
332: -1 cells allot bigendian [IF] c, 0 1 cells 1- times
333: [ELSE] 0 1 cells 1- times c, [THEN]
334: $0fffffff constant lcount-mask
335: 1 bits/char 4 - lshift 1 -
336: -1 cells allot bigendian [IF] c, -1 1 cells 1- times
337: [ELSE] -1 1 cells 1- times c, [THEN]
338: [THEN]
339:
340: \ higher level parts of find
341:
342: : flag-sign ( f -- 1|-1 )
343: \ true becomes 1, false -1
344: 0= 2* 1+ ;
345:
346: : ticking-compile-only-error ( ... -- )
347: -&2048 throw ;
348:
349: : compile-only-error ( ... -- )
350: -&14 throw ;
351:
352: : (cfa>int) ( cfa -- xt )
353: [ has? compiler [IF] ]
354: dup interpret/compile?
355: if
356: interpret/compile-int @
357: then
358: [ [THEN] ] ;
359:
360: : (x>int) ( cfa w -- xt )
361: \ get interpretation semantics of name
362: restrict-mask and [ has? rom [IF] ] 0= [ [THEN] ]
363: if
364: drop ['] compile-only-error
365: else
366: (cfa>int)
367: then ;
368:
369: has? f83headerstring [IF]
370: : name>string ( nt -- addr count ) \ gforth name-to-string
371: \g @i{addr count} is the name of the word represented by @i{nt}.
372: cell+ count lcount-mask and ;
373:
374: : ((name>)) ( nfa -- cfa )
375: name>string + cfaligned ;
376:
377: : (name>x) ( nfa -- cfa w )
378: \ cfa is an intermediate cfa and w is the flags cell of nfa
379: dup ((name>))
380: swap cell+ c@ dup alias-mask and 0=
381: IF
382: swap @ swap
383: THEN ;
384: [ELSE]
385: : name>string ( nt -- addr count ) \ gforth name-to-string
386: \g @i{addr count} is the name of the word represented by @i{nt}.
387: cell+ dup cell+ swap @ lcount-mask and ;
388:
389: : ((name>)) ( nfa -- cfa )
390: name>string + cfaligned ;
391:
392: : (name>x) ( nfa -- cfa w )
393: \ cfa is an intermediate cfa and w is the flags cell of nfa
394: dup ((name>))
395: swap cell+ @ dup alias-mask and 0=
396: IF
397: swap @ swap
398: THEN ;
399: [THEN]
400:
401: : name>int ( nt -- xt ) \ gforth name-to-int
402: \G @i{xt} represents the interpretation semantics of the word
403: \G @i{nt}. If @i{nt} has no interpretation semantics (i.e. is
404: \G @code{compile-only}), @i{xt} is the execution token for
405: \G @code{ticking-compile-only-error}, which performs @code{-2048 throw}.
406: (name>x) (x>int) ;
407:
408: : name?int ( nt -- xt ) \ gforth name-question-int
409: \G Like @code{name>int}, but perform @code{-2048 throw} if @i{nt}
410: \G has no interpretation semantics.
411: (name>x) restrict-mask and [ has? rom [IF] ] 0= [ [THEN] ]
412: if
413: ticking-compile-only-error \ does not return
414: then
415: (cfa>int) ;
416:
417: : (name>comp) ( nt -- w +-1 ) \ gforth
418: \G @i{w xt} is the compilation token for the word @i{nt}.
419: (name>x) >r
420: [ has? compiler [IF] ]
421: dup interpret/compile?
422: if
423: interpret/compile-comp @
424: then
425: [ [THEN] ]
426: r> immediate-mask and [ has? rom [IF] ] 0= [ [THEN] ] flag-sign
427: ;
428:
429: : (name>intn) ( nfa -- xt +-1 )
430: (name>x) tuck (x>int) ( w xt )
431: swap immediate-mask and [ has? rom [IF] ] 0= [ [THEN] ] flag-sign ;
432:
433: [IFDEF] prelude-mask
434: : name>prelude ( nt -- xt )
435: dup cell+ @ prelude-mask and if
436: [ -1 cells ] literal + @
437: else
438: drop ['] noop
439: then ;
440: [THEN]
441:
442: const Create ??? 0 , 3 , char ? c, char ? c, char ? c,
443: \ ??? is used by dovar:, must be created/:dovar
444:
445: [IFDEF] forthstart
446: \ if we have a forthstart we can define head? with it
447: \ otherwise leave out the head? check
448:
449: : one-head? ( addr -- f )
450: \G heuristic check whether addr is a name token; may deliver false
451: \G positives; addr must be a valid address
452: dup dup aligned <>
453: if
454: drop false exit \ heads are aligned
455: then
456: dup cell+ @ alias-mask and 0= >r
457: name>string dup $20 $1 within if
458: rdrop 2drop false exit \ realistically the name is short
459: then
460: over + cfaligned over - 2dup bounds ?do \ should be a printable string
461: i c@ bl < if
462: 2drop unloop rdrop false exit
463: then
464: loop
465: + r> if \ check for valid aliases
466: @ dup forthstart here within
467: over ['] noop ['] lit-execute 1+ within or
468: over dup aligned = and
469: 0= if
470: drop false exit
471: then
472: then \ check for cfa - must be code field or primitive
473: dup @ tuck 2 cells - = swap
474: docol: ['] lit-execute @ 1+ within or ;
475:
476: : head? ( addr -- f )
477: \G heuristic check whether addr is a name token; may deliver false
478: \G positives; addr must be a valid address; returns 1 for
479: \G particularly unsafe positives
480: \ we follow the link fields and check for plausibility; two
481: \ iterations should catch most false addresses: on the first
482: \ iteration, we may get an xt, on the second a code address (or
483: \ some code), which is typically not in the dictionary.
484: \ we added a third iteration for working with code and ;code words.
485: 3 0 do
486: dup one-head? 0= if
487: drop false unloop exit
488: endif
489: dup @ dup 0= if
490: 2drop 1 unloop exit
491: else
492: dup rot forthstart within if
493: drop false unloop exit
494: then
495: then
496: loop
497: drop true ;
498:
499: : >head-noprim ( cfa -- nt ) \ gforth to-head-noprim
500: \ also heuristic
501: dup forthstart - max-name-length @
502: [ has? float [IF] ] float+ [ [ELSE] ] cell+ [ [THEN] ] cell+ min
503: cell max cell ?do ( cfa )
504: dup i - dup @ [ alias-mask lcount-mask or ] literal
505: [ 1 bits/char 3 - lshift 1 - 1 bits/char 1 - lshift or
506: -1 cells allot bigendian [IF] c, -1 1 cells 1- times
507: [ELSE] -1 1 cells 1- times c, [THEN] ]
508: and ( cfa len|alias )
509: swap + cell+ cfaligned over alias-mask + =
510: if ( cfa )
511: dup i - cell - dup head?
512: if
513: nip unloop exit
514: then
515: drop
516: then
517: cell +loop
518: drop ??? ( wouldn't 0 be better? ) ;
519:
520: [ELSE]
521:
522: : >head-noprim ( cfa -- nt ) \ gforth to-head-noprim
523: $25 cell do ( cfa )
524: dup i - dup @ [ alias-mask lcount-mask or ] literal
525: [ 1 bits/char 3 - lshift 1 - 1 bits/char 1 - lshift or
526: -1 cells allot bigendian [IF] c, -1 1 cells 1- times
527: [ELSE] -1 1 cells 1- times c, [THEN] ]
528: and ( cfa len|alias )
529: swap + cell + cfaligned over alias-mask + =
530: if ( cfa ) i - cell - unloop exit
531: then
532: cell +loop
533: drop ??? ( wouldn't 0 be better? ) ;
534:
535: [THEN]
536:
537: cell% 2* 0 0 field >body ( xt -- a_addr ) \ core to-body
538: \G Get the address of the body of the word represented by @i{xt} (the
539: \G address of the word's data field).
540: drop drop
541:
542: cell% -2 * 0 0 field body> ( xt -- a_addr )
543: drop drop
544:
545: has? standardthreading has? compiler and [IF]
546:
547: ' @ alias >code-address ( xt -- c_addr ) \ gforth
548: \G @i{c-addr} is the code address of the word @i{xt}.
549:
550: : >does-code ( xt -- a_addr ) \ gforth
551: \G If @i{xt} is the execution token of a child of a @code{DOES>} word,
552: \G @i{a-addr} is the start of the Forth code after the @code{DOES>};
553: \G Otherwise @i{a-addr} is 0.
554: dup @ dodoes: = if
555: cell+ @
556: else
557: drop 0
558: endif ;
559:
560: has? prims [IF]
561: : flash! ! ;
562: : flashc! c! ;
563: [THEN]
564:
565: has? flash [IF] ' flash! [ELSE] ' ! [THEN]
566: alias code-address! ( c_addr xt -- ) \ gforth
567: \G Create a code field with code address @i{c-addr} at @i{xt}.
568:
569: : any-code! ( a-addr cfa code-addr -- )
570: \ for implementing DOES> and ;ABI-CODE, maybe :
571: \ code-address is stored at cfa, a-addr at cfa+cell
572: over ! cell+ ! ;
573:
574: : does-code! ( a-addr xt -- ) \ gforth
575: \G Create a code field at @i{xt} for a child of a @code{DOES>}-word;
576: \G @i{a-addr} is the start of the Forth code after @code{DOES>}.
577: [ has? flash [IF] ]
578: dodoes: over flash! cell+ flash!
579: [ [ELSE] ]
580: dodoes: any-code!
581: [ [THEN] ] ;
582:
583: 2 cells constant /does-handler ( -- n ) \ gforth
584: \G The size of a @code{DOES>}-handler (includes possible padding).
585:
586: [THEN]
587:
588: : sfind ( c-addr u -- 0 / xt +-1 ) \ gforth-obsolete
589: find-name dup
590: if ( nt )
591: state @
592: if
593: (name>comp)
594: else
595: (name>intn)
596: then
597: then ;
598:
599: : find ( c-addr -- xt +-1 | c-addr 0 ) \ core,search
600: \G Search all word lists in the current search order for the
601: \G definition named by the counted string at @i{c-addr}. If the
602: \G definition is not found, return 0. If the definition is found
603: \G return 1 (if the definition has non-default compilation
604: \G semantics) or -1 (if the definition has default compilation
605: \G semantics). The @i{xt} returned in interpret state represents
606: \G the interpretation semantics. The @i{xt} returned in compile
607: \G state represented either the compilation semantics (for
608: \G non-default compilation semantics) or the run-time semantics
609: \G that the compilation semantics would @code{compile,} (for
610: \G default compilation semantics). The ANS Forth standard does
611: \G not specify clearly what the returned @i{xt} represents (and
612: \G also talks about immediacy instead of non-default compilation
613: \G semantics), so this word is questionable in portable programs.
614: \G If non-portability is ok, @code{find-name} and friends are
615: \G better (@pxref{Name token}).
616: dup count sfind dup
617: if
618: rot drop
619: then ;
620:
621: \ ticks in interpreter
622:
623: : (') ( "name" -- nt ) \ gforth
624: parse-name name-too-short?
625: find-name dup 0=
626: IF
627: drop -&13 throw
628: THEN ;
629:
630: : ' ( "name" -- xt ) \ core tick
631: \g @i{xt} represents @i{name}'s interpretation
632: \g semantics. Perform @code{-14 throw} if the word has no
633: \g interpretation semantics.
634: (') name?int ;
635:
636: has? compiler 0= [IF] \ interpreter only version of IS and TO
637:
638: : IS ' >body ! ;
639: ' IS Alias TO
640:
641: [THEN]
642:
643: \ \ the interpreter loop mar92py
644:
645: \ interpret 10mar92py
646:
647: Defer parser1 ( c-addr u -- ... xt)
648: \ "... xt" is the action to be performed by the text-interpretation of c-addr u
649:
650: : parser ( c-addr u -- ... )
651: \ text-interpret the word/number c-addr u, possibly producing a number
652: parser1 execute ;
653:
654: has? ec [IF]
655: ' (name) Alias parse-name
656: : no.extensions 2drop -&13 throw ;
657: ' no.extensions Alias compiler-notfound1
658: ' no.extensions Alias interpreter-notfound1
659: [ELSE]
660: Defer parse-name ( "name" -- c-addr u ) \ gforth
661: \G Get the next word from the input buffer
662: ' (name) IS parse-name
663:
664: ' parse-name alias parse-word ( -- c-addr u ) \ gforth-obsolete
665: \G old name for @code{parse-name}
666:
667: ' parse-name alias name ( -- c-addr u ) \ gforth-obsolete
668: \G old name for @code{parse-name}
669:
670: : no.extensions ( addr u -- )
671: 2drop -&13 throw ;
672:
673: has? recognizer 0= [IF]
674: Defer compiler-notfound1 ( c-addr count -- ... xt )
675: Defer interpreter-notfound1 ( c-addr count -- ... xt )
676:
677: ' no.extensions IS compiler-notfound1
678: ' no.extensions IS interpreter-notfound1
679: [THEN]
680:
681: Defer before-word ( -- ) \ gforth
682: \ called before the text interpreter parses the next word
683: ' noop IS before-word
684: [THEN]
685:
686: has? backtrace [IF]
687: : interpret1 ( ... -- ... )
688: rp@ backtrace-rp0 !
689: BEGIN
690: ?stack [ has? EC 0= [IF] ] before-word [ [THEN] ] parse-name dup
691: WHILE
692: parser1 execute
693: REPEAT
694: 2drop ;
695:
696: : interpret ( ?? -- ?? ) \ gforth
697: \ interpret/compile the (rest of the) input buffer
698: backtrace-rp0 @ >r
699: ['] interpret1 catch
700: r> backtrace-rp0 !
701: throw ;
702: [ELSE]
703: : interpret ( ... -- ... )
704: BEGIN
705: ?stack [ has? EC 0= [IF] ] before-word [ [THEN] ] parse-name dup
706: WHILE
707: parser1 execute
708: REPEAT
709: 2drop ;
710: [THEN]
711:
712: \ interpreter 30apr92py
713:
714: [IFDEF] prelude-mask
715: : run-prelude ( nt|0 -- nt|0 )
716: \ run the prelude of the name identified by nt (if present). This
717: \ is used in the text interpreter and similar stuff.
718: dup if
719: dup name>prelude execute
720: then ;
721: [THEN]
722:
723: has? recognizer 0= [IF]
724: \ not the most efficient implementations of interpreter and compiler
725: : interpreter1 ( c-addr u -- ... xt )
726: 2dup find-name [ [IFDEF] prelude-mask ] run-prelude [ [THEN] ] dup
727: if
728: nip nip name>int
729: else
730: drop
731: 2dup 2>r snumber?
732: IF
733: 2rdrop ['] noop
734: ELSE
735: 2r> interpreter-notfound1
736: THEN
737: then ;
738:
739: ' interpreter1 IS parser1
740: [THEN]
741:
742: \ \ Query Evaluate 07apr93py
743:
744: has? file 0= [IF]
745: : sourceline# ( -- n ) 1 ;
746: [ELSE]
747: has? new-input 0= [IF]
748: Variable #fill-bytes
749: \G number of bytes read via (read-line) by the last refill
750: [THEN]
751: [THEN]
752:
753: has? new-input 0= [IF]
754: : input-start-line ( -- ) >in off ;
755: : refill ( -- flag ) \ core-ext,block-ext,file-ext
756: \G Attempt to fill the input buffer from the input source. When
757: \G the input source is the user input device, attempt to receive
758: \G input into the terminal input device. If successful, make the
759: \G result the input buffer, set @code{>IN} to 0 and return true;
760: \G otherwise return false. When the input source is a block, add 1
761: \G to the value of @code{BLK} to make the next block the input
762: \G source and current input buffer, and set @code{>IN} to 0;
763: \G return true if the new value of @code{BLK} is a valid block
764: \G number, false otherwise. When the input source is a text file,
765: \G attempt to read the next line from the file. If successful,
766: \G make the result the current input buffer, set @code{>IN} to 0
767: \G and return true; otherwise, return false. A successful result
768: \G includes receipt of a line containing 0 characters.
769: [ has? file [IF] ]
770: blk @ IF 1 blk +! true EXIT THEN
771: [ [THEN] ]
772: tib /line
773: [ has? file [IF] ]
774: loadfile @ ?dup
775: IF (read-line) throw #fill-bytes !
776: ELSE
777: [ [THEN] ]
778: sourceline# 0< IF 2drop false EXIT THEN
779: accept eof @ 0=
780: [ has? file [IF] ]
781: THEN
782: 1 loadline +!
783: [ [THEN] ]
784: swap #tib !
785: input-start-line ;
786:
787: : query ( -- ) \ core-ext
788: \G Make the user input device the input source. Receive input into
789: \G the Terminal Input Buffer. Set @code{>IN} to zero. OBSOLESCENT:
790: \G superceeded by @code{accept}.
791: [ has? file [IF] ]
792: blk off loadfile off
793: [ [THEN] ]
794: refill drop ;
795: [THEN]
796:
797: \ save-mem extend-mem
798:
799: has? os [IF]
800: : save-mem ( addr1 u -- addr2 u ) \ gforth
801: \g copy a memory block into a newly allocated region in the heap
802: swap >r
803: dup allocate throw
804: swap 2dup r> -rot move ;
805:
806: : free-mem-var ( addr -- )
807: \ addr is the address of a 2variable containing address and size
808: \ of a memory range; frees memory and clears the 2variable.
809: dup 2@ drop dup
810: if ( addr mem-start )
811: free throw
812: 0 0 rot 2!
813: else
814: 2drop
815: then ;
816:
817: : extend-mem ( addr1 u1 u -- addr addr2 u2 )
818: \ extend memory block allocated from the heap by u aus
819: \ the (possibly reallocated) piece is addr2 u2, the extension is at addr
820: over >r + dup >r resize throw
821: r> over r> + -rot ;
822: [THEN]
823:
824: \ EVALUATE 17may93jaw
825:
826: has? file 0= has? new-input 0= and [IF]
827: : push-file ( -- ) r>
828: tibstack @ >r >tib @ >r #tib @ >r
829: >tib @ tibstack @ = IF r@ tibstack +! THEN
830: tibstack @ >tib ! >in @ >r >r ;
831:
832: : pop-file ( throw-code -- throw-code )
833: r>
834: r> >in ! r> #tib ! r> >tib ! r> tibstack ! >r ;
835: [THEN]
836:
837: has? new-input 0= [IF]
838: : evaluate ( c-addr u -- ) \ core,block
839: \G Save the current input source specification. Store @code{-1} in
840: \G @code{source-id} and @code{0} in @code{blk}. Set @code{>IN} to
841: \G @code{0} and make the string @i{c-addr u} the input source
842: \G and input buffer. Interpret. When the parse area is empty,
843: \G restore the input source specification.
844: [ has? file [IF] ]
845: s" *evaluated string*" loadfilename>r
846: [ [THEN] ]
847: push-file #tib ! >tib !
848: input-start-line
849: [ has? file [IF] ]
850: blk off loadfile off -1 loadline !
851: [ [THEN] ]
852: ['] interpret catch
853: pop-file
854: [ has? file [IF] ]
855: r>loadfilename
856: [ [THEN] ]
857: throw ;
858: [THEN]
859:
860: \ \ Quit 13feb93py
861:
862: Defer 'quit
863:
864: has? os [IF]
865: Defer .status
866: [ELSE]
867: [IFUNDEF] bye
868: : (bye) ( 0 -- ) \ back to DOS
869: drop 5 emit ;
870:
871: : bye ( -- ) 0 (bye) ;
872: [THEN]
873: [THEN]
874:
875: : prompt state @ IF ." compiled" EXIT THEN ." ok" ;
876:
877: : (quit) ( -- )
878: \ exits only through THROW etc.
879: BEGIN
880: [ has? ec [IF] ] cr [ [ELSE] ]
881: .status ['] cr catch if
882: [ has? OS [IF] ] >stderr [ [THEN] ]
883: cr ." Can't print to stdout, leaving" cr
884: \ if stderr does not work either, already DoError causes a hang
885: 2 (bye)
886: endif [ [THEN] ]
887: refill WHILE
888: interpret prompt
889: REPEAT
890: bye ;
891:
892: ' (quit) IS 'quit
893:
894: \ \ DOERROR (DOERROR) 13jun93jaw
895:
896: has? os [IF]
897: 8 Constant max-errors
898: 5 has? file 2 and + Constant /error
899: Variable error-stack 0 error-stack !
900: max-errors /error * cells allot
901: \ format of one cell:
902: \ source ( c-addr u )
903: \ last parsed lexeme ( c-addr u )
904: \ line-number
905: \ Loadfilename ( addr u )
906:
907: : error> ( -- c-addr1 u1 c-addr2 u2 line# [addr u] )
908: -1 error-stack +!
909: error-stack dup @
910: /error * cells + cell+
911: /error cells bounds DO
912: I @
913: cell +LOOP ;
914:
915: : >error ( c-addr1 u1 c-addr2 u2 line# [addr u] -- )
916: error-stack dup @ dup 1+
917: max-errors 1- min error-stack !
918: /error * cells + cell+
919: /error 1- cells bounds swap DO
920: I !
921: -1 cells +LOOP ;
922:
923: : input-error-data ( -- c-addr1 u1 c-addr2 u2 line# [addr u] )
924: \ error data for the current input, to be used by >error or .error-frame
925: source over >r save-mem over r> -
926: input-lexeme 2@ >r + r> sourceline#
927: [ has? file [IF] ] sourcefilename [ [THEN] ] ;
928:
929: : dec. ( n -- ) \ gforth
930: \G Display @i{n} as a signed decimal number, followed by a space.
931: \ !! not used...
932: base @ decimal swap . base ! ;
933:
934: : dec.r ( u n -- ) \ gforth
935: \G Display @i{u} as a unsigned decimal number in a field @i{n}
936: \G characters wide.
937: base @ >r decimal .r r> base ! ;
938:
939: : hex. ( u -- ) \ gforth
940: \G Display @i{u} as an unsigned hex number, prefixed with a "$" and
941: \G followed by a space.
942: \ !! not used...
943: [char] $ emit base @ swap hex u. base ! ;
944:
945: : -trailing ( c_addr u1 -- c_addr u2 ) \ string dash-trailing
946: \G Adjust the string specified by @i{c-addr, u1} to remove all
947: \G trailing spaces. @i{u2} is the length of the modified string.
948: BEGIN
949: dup
950: WHILE
951: 1- 2dup + c@ bl <>
952: UNTIL 1+ THEN ;
953:
954: DEFER DOERROR
955:
956: has? backtrace [IF]
957: Defer dobacktrace ( -- )
958: ' noop IS dobacktrace
959: [THEN]
960:
961: : .error-string ( throw-code -- )
962: dup -2 =
963: IF "error @ ?dup IF count type THEN drop
964: ELSE .error
965: THEN ;
966:
967: [IFUNDEF] umin
968: : umin ( u1 u2 -- u )
969: 2dup u>
970: if
971: swap
972: then
973: drop ;
974: [THEN]
975:
976: Defer mark-start
977: Defer mark-end
978:
979: :noname ." >>>" ; IS mark-start
980: :noname ." <<<" ; IS mark-end
981:
982: : part-type ( addr1 u1 u -- addr2 u2 )
983: \ print first u characters of addr1 u1, addr2 u2 is the rest
984: over umin 2 pick over type /string ;
985:
986: : .error-line ( c-addr1 u1 c-addr2 u2 -- )
987: \ print error in line c-addr1 u1, where the error-causing lexeme
988: \ is c-addr2 u2
989: >r 2 pick - part-type ( c-addr3 u3 R: u2 )
990: mark-start r> part-type mark-end ( c-addr4 u4 )
991: type ;
992:
993: : .error-frame ( throwcode addr1 u1 addr2 u2 n2 [addr3 u3] -- throwcode )
994: \ addr3 u3: filename of included file - optional
995: \ n2: line number
996: \ addr2 u2: parsed lexeme (should be marked as causing the error)
997: \ addr1 u1: input line
998: error-stack @
999: IF ( throwcode addr1 u1 n0 n1 n2 [addr2 u2] )
1000: [ has? file [IF] ] \ !! unbalanced stack effect
1001: over IF
1002: cr ." in file included from "
1003: type ." :"
1004: 0 dec.r 2drop 2drop
1005: ELSE
1006: 2drop 2drop 2drop drop
1007: THEN
1008: [ [THEN] ] ( throwcode addr1 u1 n0 n1 n2 )
1009: ELSE ( throwcode addr1 u1 n0 n1 n2 [addr2 u2] )
1010: [ has? file [IF] ]
1011: cr type ." :"
1012: [ [THEN] ] ( throwcode addr1 u1 n0 n1 n2 )
1013: dup 0 dec.r ." : " 5 pick .error-string
1014: IF \ if line# non-zero, there is a line
1015: cr .error-line
1016: ELSE
1017: 2drop 2drop
1018: THEN
1019: THEN ;
1020:
1021: : (DoError) ( throw-code -- )
1022: [ has? os [IF] ]
1023: >stderr
1024: [ [THEN] ]
1025: input-error-data .error-frame
1026: error-stack @ 0 ?DO
1027: error>
1028: .error-frame
1029: LOOP
1030: drop
1031: [ has? backtrace [IF] ]
1032: dobacktrace
1033: [ [THEN] ]
1034: normal-dp dpp ! ;
1035:
1036: ' (DoError) IS DoError
1037:
1038: [ELSE]
1039: : dec. base @ >r decimal . r> base ! ;
1040: : DoError ( throw-code -- )
1041: cr source drop >in @ type ." <<< "
1042: dup -2 = IF "error @ type drop EXIT THEN
1043: .error ;
1044: [THEN]
1045:
1046: : quit ( ?? -- ?? ) \ core
1047: \G Empty the return stack, make the user input device
1048: \G the input source, enter interpret state and start
1049: \G the text interpreter.
1050: rp0 @ rp! handler off clear-tibstack
1051: [ has? new-input 0= [IF] ] >tib @ >r [ [THEN] ]
1052: BEGIN
1053: [ has? compiler [IF] ]
1054: [compile] [
1055: [ [THEN] ]
1056: \ stack depths may be arbitrary here
1057: ['] 'quit CATCH dup
1058: WHILE
1059: <# \ reset hold area, or we may get another error
1060: DoError
1061: \ stack depths may be arbitrary still (or again), so clear them
1062: clearstacks
1063: [ has? new-input [IF] ] clear-tibstack
1064: [ [ELSE] ] r@ >tib ! r@ tibstack !
1065: [ [THEN] ]
1066: REPEAT
1067: drop [ has? new-input [IF] ] clear-tibstack
1068: [ [ELSE] ] r> >tib !
1069: [ [THEN] ] ;
1070:
1071: \ \ Cold Boot 13feb93py
1072:
1073: : gforth ( -- )
1074: ." Gforth " version-string type
1075: ." , Copyright (C) 1995-2011 Free Software Foundation, Inc." cr
1076: ." Gforth comes with ABSOLUTELY NO WARRANTY; for details type `license'"
1077: [ has? os [IF] ]
1078: cr ." Type `bye' to exit"
1079: [ [THEN] ] ;
1080:
1081: defer bootmessage ( -- ) \ gforth
1082: \G Hook (deferred word) executed right after interpreting the OS
1083: \G command-line arguments. Normally prints the Gforth startup
1084: \G message.
1085:
1086: has? file [IF]
1087: defer process-args
1088: [THEN]
1089:
1090: ' gforth IS bootmessage
1091:
1092: has? os [IF]
1093: Defer 'cold ( -- ) \ gforth tick-cold
1094: \G Hook (deferred word) for things to do right before interpreting the
1095: \G OS command-line arguments. Normally does some initializations that
1096: \G you also want to perform.
1097: ' noop IS 'cold
1098: [THEN]
1099:
1100: : cold ( -- ) \ gforth
1101: [ has? backtrace [IF] ]
1102: rp@ backtrace-rp0 !
1103: [ [THEN] ]
1104: [ has? file [IF] ]
1105: os-cold
1106: [ [THEN] ]
1107: [ has? os [IF] ]
1108: set-encoding-fixed-width
1109: 'cold
1110: [ [THEN] ]
1111: [ has? file [IF] ]
1112: process-args
1113: loadline off
1114: [ [THEN] ]
1115: bootmessage
1116: quit ;
1117:
1118: has? new-input 0= [IF]
1119: : clear-tibstack ( -- )
1120: [ has? glocals [IF] ]
1121: lp@ forthstart 7 cells + @ -
1122: [ [ELSE] ]
1123: [ has? os [IF] ]
1124: r0 @ forthstart 6 cells + @ -
1125: [ [ELSE] ]
1126: sp@ cell+
1127: [ [THEN] ]
1128: [ [THEN] ]
1129: dup >tib ! tibstack ! #tib off
1130: input-start-line ;
1131: [THEN]
1132:
1133: : boot ( path n **argv argc -- )
1134: [ has? no-userspace 0= [IF] ]
1135: main-task up!
1136: [ [THEN] ]
1137: [ has? os [IF] ]
1138: os-boot
1139: [ [THEN] ]
1140: [ has? rom [IF] ]
1141: ram-shadow dup @ dup -1 <> >r u> r> and IF
1142: ram-shadow 2@ ELSE
1143: ram-mirror ram-size THEN ram-start swap move
1144: [ [THEN] ]
1145: sp@ sp0 !
1146: [ has? peephole [IF] ]
1147: \ only needed for greedy static superinstruction selection
1148: \ primtable prepare-peephole-table TO peeptable
1149: [ [THEN] ]
1150: [ has? new-input [IF] ]
1151: current-input off
1152: [ [THEN] ]
1153: clear-tibstack
1154: 0 0 includefilename 2!
1155: rp@ rp0 !
1156: [ has? floating [IF] ]
1157: fp@ fp0 !
1158: [ [THEN] ]
1159: [ has? os [IF] ]
1160: handler off
1161: ['] cold catch dup -&2049 <> if \ broken pipe?
1162: DoError cr
1163: endif
1164: [ [ELSE] ]
1165: cold
1166: [ [THEN] ]
1167: [ has? os [IF] ]
1168: 1 (bye) \ !! determin exit code from throw code?
1169: [ [THEN] ]
1170: ;
1171:
1172: has? os [IF]
1173: : bye ( -- ) \ tools-ext
1174: [ has? file [IF] ]
1175: script? 0= IF cr THEN
1176: [ [ELSE] ]
1177: cr
1178: [ [THEN] ]
1179: 0 (bye) ;
1180: [THEN]
1181:
1182: \ **argv may be scanned by the C starter to get some important
1183: \ information, as -display and -geometry for an X client FORTH
1184: \ or space and stackspace overrides
1185:
1186: \ 0 arg contains, however, the name of the program.
1187:
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>