Diff for /gforth/blocks.fs between versions 1.4 and 1.40

version 1.4, 1995/04/20 09:42:45 version 1.40, 2003/01/22 10:59:54
Line 1 Line 1
 \ A simple immplementation of the blocks wordset.   \ A less simple implementation of the blocks wordset. 
   
 \ This implementation uses only a single buffer and will therefore be a  \ Copyright (C) 1995,1996,1997,1998,2000 Free Software Foundation, Inc.
 \ little slow. An efficient implementation would use mmap on OSs that  
   \ This file is part of Gforth.
   
   \ Gforth is free software; you can redistribute it and/or
   \ modify it under the terms of the GNU General Public License
   \ as published by the Free Software Foundation; either version 2
   \ of the License, or (at your option) any later version.
   
   \ This program is distributed in the hope that it will be useful,
   \ but WITHOUT ANY WARRANTY; without even the implied warranty of
   \ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   \ GNU General Public License for more details.
   
   \ You should have received a copy of the GNU General Public License
   \ along with this program; if not, write to the Free Software
   \ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111, USA.
   
   
   \ A more efficient implementation would use mmap on OSs that
 \ provide it and many buffers on OSs that do not provide mmap.  \ provide it and many buffers on OSs that do not provide mmap.
   
   \ Now, the replacement algorithm is "direct mapped"; change to LRU
   \ if too slow. Using more buffers helps, too.
   
 \ I think I avoid the assumption 1 char = 1 here, but I have not tested this  \ I think I avoid the assumption 1 char = 1 here, but I have not tested this
   
 \ 1024 constant chars/block \ mandated by the standard  \ 1024 constant chars/block \ mandated by the standard
   
 create block-buffer chars/block chars allot  require struct.fs
   
 variable buffer-block 0 buffer-block ! \ the block currently in the buffer  
 variable block-fid 0 block-fid ! \ the file id of the current block file  
 variable buffer-dirty buffer-dirty off  
   
   struct
       cell%               field buffer-block   \ the block number
       cell%               field buffer-fid     \ the block's fid
       cell%               field buffer-dirty   \ the block dirty flag
       char% chars/block * field block-buffer   \ the data
       cell% 0 *           field next-buffer
   end-struct buffer-struct
   
   Variable block-buffers
   Variable last-block
   
   $20 Value buffers
   
   \ limit block files to 2GB; gforth <0.6.0 erases larger block files on
   \ 32-bit systems
   $200000 Value block-limit
   
   User block-fid
   User block-offset ( -- addr ) \ gforth
   \G User variable containing the number of the first block (default
   \G since 0.5.0: 0).  Block files created with Gforth versions before
   \G 0.5.0 have the offset 1.  If you use these files you can: @code{1
   \G offset !}; or add 1 to every block number used; or prepend 1024
   \G characters to the file.
   0 block-offset !  \ store 1 here fore 0.4.0 compatibility
   
   ' block-offset alias offset \ !! eliminate this?
   
   : block-cold ( -- )
       block-fid off  last-block off
       buffer-struct buffers * %alloc dup block-buffers ! ( addr )
       buffer-struct %size buffers * erase ;
   
   ' block-cold INIT8 chained
   
   block-cold
   
   Defer flush-blocks ( -- ) \ gforth
   
   : open-blocks ( c-addr u -- ) \ gforth
   \g Use the file, whose name is given by @i{c-addr u}, as the blocks file.
       try ( c-addr u )
           2dup open-fpath-file throw
           rot close-file throw  2dup file-status throw bin open-file throw
           >r 2drop r>
       recover ( c-addr u ior )
           >r 2dup file-status nip 0= r> and throw \ does it really not exist?
           r/w bin create-file throw
       endtry
       block-fid @ IF
           flush-blocks block-fid @ close-file throw
       THEN
       block-fid ! ;
   
   : use ( "file" -- ) \ gforth
       \g Use @i{file} as the blocks file.
       name open-blocks ;
   
 \ the file is opened as binary file, since it either will contain text  \ the file is opened as binary file, since it either will contain text
 \ without newlines or binary data  \ without newlines or binary data
 : get-block-fid ( -- fid )  : get-block-fid ( -- wfileid ) \ gforth
       \G Return the file-id of the current blocks file. If no blocks
       \G file has been opened, use @file{blocks.fb} as the default
       \G blocks file.
     block-fid @ 0=      block-fid @ 0=
     if      if
         s" blocks.fb" r/w bin open-file 0<>          s" blocks.fb" open-blocks
         if  
             s" blocks.fb" r/w bin create-file throw  
         then  
         block-fid !  
     then      then
     block-fid @ ;      block-fid @ ;
   
 : block-position ( u -- )  : block-position ( u -- ) \ block
     \ positions the block file to the start of block u  \G Position the block file to the start of block @i{u}.
     1- chars/block chars um* get-block-fid reposition-file throw ;      dup block-limit u>= -35 and throw
       offset @ - chars/block chars um* get-block-fid reposition-file throw ;
 : update ( -- )  
     buffer-dirty on ;  : update ( -- ) \ block
       \G Mark the state of the current block buffer as assigned-dirty.
 : save-buffers ( -- )      last-block @ ?dup IF  buffer-dirty on  THEN ;
     buffer-dirty @ buffer-block @ 0<> and  
   : save-buffer ( buffer -- ) \ gforth
       >r
       r@ buffer-dirty @ r@ buffer-block @ 0<> and
     if      if
         buffer-block @ block-position          r@ buffer-block @ block-position
         block-buffer chars/block get-block-fid write-file throw          r@ block-buffer chars/block  r@ buffer-fid @  write-file throw
         buffer-dirty off          r@ buffer-fid @ flush-file throw
     endif ;          r@ buffer-dirty off 
       endif
 : empty-buffers ( -- )      rdrop ;
     0 buffer-block ! ;  
   : empty-buffer ( buffer -- ) \ gforth
 : flush ( -- )      buffer-block off ;
   
   : save-buffers  ( -- ) \ block
       \G Transfer the contents of each @code{update}d block buffer to
       \G mass storage, then mark all block buffers as assigned-clean.
       block-buffers @
       buffers 0 ?DO dup save-buffer next-buffer LOOP drop ;
   
   : empty-buffers ( -- ) \ block-ext
       \G Mark all block buffers as unassigned; if any had been marked as
       \G assigned-dirty (by @code{update}), the changes to those blocks
       \G will be lost.
       block-buffers @
       buffers 0 ?DO dup empty-buffer next-buffer LOOP drop ;
   
   : flush ( -- ) \ block
       \G Perform the functions of @code{save-buffers} then
       \G @code{empty-buffers}.
     save-buffers      save-buffers
     empty-buffers ;      empty-buffers ;
   
 : block ( u -- a-addr )  ' flush IS flush-blocks
     dup 0= -35 and throw  
     dup buffer-block @ <>  : get-buffer ( u -- a-addr ) \ gforth
       0 buffers um/mod drop buffer-struct %size * block-buffers @ + ;
   
   : block ( u -- a-addr ) \ gforthman- block
       \G If a block buffer is assigned for block @i{u}, return its
       \G start address, @i{a-addr}. Otherwise, assign a block buffer
       \G for block @i{u} (if the assigned block buffer has been
       \G @code{update}d, transfer the contents to mass storage), read
       \G the block into the block buffer and return its start address,
       \G @i{a-addr}.
       dup offset @ u< -35 and throw
       dup get-buffer >r
       dup r@ buffer-block @ <>
       r@ buffer-fid @ block-fid @ <> or
     if      if
         save-buffers          r@ save-buffer
         dup block-position          dup block-position
         block-buffer chars/block get-block-fid read-file throw          r@ block-buffer chars/block get-block-fid read-file throw
         \ clear the rest of the buffer if the file is too short          \ clear the rest of the buffer if the file is too short
         block-buffer over chars + chars/block rot - blank          r@ block-buffer over chars + chars/block rot chars - blank
         buffer-block !              r@ buffer-block !
           get-block-fid r@ buffer-fid !
     else      else
         drop          drop
     then      then
     block-buffer ;      r> dup last-block ! block-buffer ;
   
 : buffer ( u -- a-addr )  : buffer ( u -- a-addr ) \ block
       \G If a block buffer is assigned for block @i{u}, return its
       \G start address, @i{a-addr}. Otherwise, assign a block buffer
       \G for block @i{u} (if the assigned block buffer has been
       \G @code{update}d, transfer the contents to mass storage) and
       \G return its start address, @i{a-addr}.  The subtle difference
       \G between @code{buffer} and @code{block} mean that you should
       \G only use @code{buffer} if you don't care about the previous
       \G contents of block @i{u}. In Gforth, this simply calls
       \G @code{block}.
     \ reading in the block is unnecessary, but simpler      \ reading in the block is unnecessary, but simpler
     block ;      block ;
   
 User scr 0 scr !  User scr ( -- a-addr ) \ block-ext s-c-r
       \G @code{User} variable -- @i{a-addr} is the address of a cell containing
 : list ( u -- )      \G the block number of the block most recently processed by
       \G @code{list}.
   0 scr !
   
   \ nac31Mar1999 moved "scr @" to list to make the stack comment correct
   : updated?  ( n -- f ) \ gforth
       \G Return true if @code{updated} has been used to mark block @i{n}
       \G as assigned-dirty.
       buffer
       [ 0 buffer-dirty 0 block-buffer - ] Literal + @ ;
   
   : list ( u -- ) \ block-ext
       \G Display block @i{u}. In Gforth, the block is displayed as 16
       \G numbered lines, each of 64 characters.
     \ calling block again and again looks inefficient but is necessary      \ calling block again and again looks inefficient but is necessary
     \ in a multitasking environment      \ in a multitasking environment
     dup scr !      dup scr !
     ." Screen " u. cr      ." Screen " u.
       scr @ updated?  0= IF ." not "  THEN  ." modified     " cr
     16 0      16 0
     ?do      ?do
         i 2 .r space scr @ block i 64 * chars + 64 type cr          i 2 .r space scr @ block i 64 * chars + 64 type cr
     loop ;      loop ;
   
 : (source)  ( -- addr len )  [IFDEF] current-input
   :noname  2 <> -12 and throw >in ! blk ! ;
                                 \ restore-input
   :noname  blk @ >in @ 2 ;      \ save-input
   :noname  2 ;                  \ source-id "*a block*"
   :noname  1 blk +! 1 loadline +! true ;      \ refill
   :noname  blk @ block chars/block ;  \ source
   
   Create block-input   A, A, A, A, A,
   
   : load  ( i*x n -- j*x ) \ block
       \G Save the current input source specification. Store @i{n} in
       \G @code{BLK}, set @code{>IN} to 0 and interpret. When the parse
       \G area is exhausted, restore the input source specification.
       block-input 0 new-tib dup loadline ! blk !  s" * a block*" loadfilename 2!
       ['] interpret catch pop-file throw ;
   [ELSE]
   : (source)  ( -- c-addr u )
   blk @ ?dup    blk @ ?dup
   IF    block chars/block    IF    block chars/block
   ELSE  tib #tib @    ELSE  tib #tib @
   THEN ;    THEN ;
   
 ' (source) IS source  ' (source) IS source ( -- c-addr u ) \ core
   \G @i{c-addr} is the address of the input buffer and @i{u} is the
 : load ( i*x n -- j*x )  \G number of characters in it.
   push-file  
   dup loadline ! blk ! >in off ( ['] ) interpret ( catch )  : load ( i*x n -- j*x ) \ block
   pop-file ( throw ) ;      \G Save the current input source specification. Store @i{n} in
       \G @code{BLK}, set @code{>IN} to 0 and interpret. When the parse
 : thru ( i*x n1 n2 -- j*x )      \G area is exhausted, restore the input source specification.
   1+ swap 0 ?DO  I load  LOOP ;      s" * a block*" loadfilename>r
       push-file
 : +load ( i*x n -- j*x )  blk @ + load ;      dup loadline ! blk ! >in off ['] interpret catch
       pop-file
 : +thru ( i*x n1 n2 -- j*x )      r>loadfilename
   1+ swap 0 ?DO  I +load  LOOP ;      throw ;
   [THEN]
   
   : thru ( i*x n1 n2 -- j*x ) \ block-ext
       \G @code{load} the blocks @i{n1} through @i{n2} in sequence.
       1+ swap ?DO  I load  LOOP ;
   
   : +load ( i*x n -- j*x ) \ gforth
       \G Used within a block to load the block specified as the
       \G current block + @i{n}.
       blk @ + load ;
   
   : +thru ( i*x n1 n2 -- j*x ) \ gforth
       \G Used within a block to load the range of blocks specified as the
       \G current block + @i{n1} thru the current block + @i{n2}.
       1+ swap ?DO  I +load  LOOP ;
   
   : --> ( -- ) \ gforthman- gforth chain
       \G If this symbol is encountered whilst loading block @i{n},
       \G discard the remainder of the block and load block @i{n+1}. Used
       \G for chaining multiple blocks together as a single loadable
       \G unit.  Not recommended, because it destroys the independence of
       \G loading.  Use @code{thru} (which is standard) or @code{+thru}
       \G instead.
       refill drop ; immediate
   
   : block-included ( a-addr u -- ) \ gforth
       \G Use within a block that is to be processed by @code{load}. Save
       \G the current blocks file specification, open the blocks file
       \G specified by @i{a-addr u} and @code{load} block 1 from that
       \G file (which may in turn chain or load other blocks). Finally,
       \G close the blocks file and restore the original blocks file.
       block-fid @ >r block-fid off open-blocks
       1 load block-fid @ close-file throw flush
       r> block-fid ! ;
   
   \ thrown out because it may provide unpleasant surprises - anton
   \ : include ( "name" -- )
   \     name 2dup dup 3 - /string s" .fb" compare
   \     0= IF  block-included  ELSE  included  THEN ;
   
 get-current environment-wordlist set-current  get-current environment-wordlist set-current
 true constant block  true constant block
 true constant block-ext  true constant block-ext
 set-current  set-current
   
   : bye ( -- ) \ tools-ext
     \G Return control to the host operating system (if any).
     ['] flush catch drop bye ;

Removed from v.1.4  
changed lines
  Added in v.1.40


FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>