Diff for /gforth/blocks.fs between versions 1.23 and 1.33

version 1.23, 1999/03/29 22:52:27 version 1.33, 2000/09/23 15:46:51
Line 1 Line 1
 \ A less simple implementation of the blocks wordset.   \ A less simple implementation of the blocks wordset. 
   
 \ Copyright (C) 1995,1996,1997,1998 Free Software Foundation, Inc.  \ Copyright (C) 1995,1996,1997,1998,2000 Free Software Foundation, Inc.
   
 \ This file is part of Gforth.  \ This file is part of Gforth.
   
Line 16 Line 16
   
 \ You should have received a copy of the GNU General Public License  \ You should have received a copy of the GNU General Public License
 \ along with this program; if not, write to the Free Software  \ along with this program; if not, write to the Free Software
 \ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.  \ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111, USA.
   
   
 \ A more efficient implementation would use mmap on OSs that  \ A more efficient implementation would use mmap on OSs that
Line 45  Variable last-block Line 45  Variable last-block
 $20 Value buffers  $20 Value buffers
   
 User block-fid  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-cold ( -- )
     block-fid off  last-block off      block-fid off  last-block off
Line 55  User block-fid Line 64  User block-fid
   
 block-cold  block-cold
   
 Defer flush-blocks  Defer flush-blocks ( -- ) \ gforth
   
 : open-blocks ( addr u -- ) \ gforth  : open-blocks ( c-addr u -- ) \ gforth
     \g Use the file, whose name is given by @var{addr u}, as the blocks file.      \g Use the file, whose name is given by @i{c-addr u}, as the blocks file.
     2dup open-fpath-file 0<>      2dup open-fpath-file 0<>
     if      if
         r/w bin create-file throw          r/w bin create-file throw
Line 70  Defer flush-blocks Line 79  Defer flush-blocks
     block-fid ! ;      block-fid ! ;
   
 : use ( "file" -- ) \ gforth  : use ( "file" -- ) \ gforth
     \g Use @var{file} as the blocks file.      \g Use @i{file} as the blocks file.
     name open-blocks ;      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 ) \ gforth  : 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" open-blocks          s" blocks.fb" open-blocks
Line 83  Defer flush-blocks Line 95  Defer flush-blocks
     block-fid @ ;      block-fid @ ;
   
 : block-position ( u -- ) \ block  : block-position ( u -- ) \ block
     \G Position the block file to the start of block @var{u}.      \G Position the block file to the start of block @i{u}.
     1- chars/block chars um* get-block-fid reposition-file throw ;      offset @ - chars/block chars um* get-block-fid reposition-file throw ;
   
 : update ( -- ) \ block  : update ( -- ) \ block
       \G Mark the state of the current block buffer as assigned-dirty.
     last-block @ ?dup IF  buffer-dirty on  THEN ;      last-block @ ?dup IF  buffer-dirty on  THEN ;
   
 : save-buffer ( buffer -- ) \ gforth  : save-buffer ( buffer -- ) \ gforth
Line 103  Defer flush-blocks Line 116  Defer flush-blocks
     buffer-block off ;      buffer-block off ;
   
 : save-buffers  ( -- ) \ block  : 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 @      block-buffers @
     buffers 0 ?DO  dup save-buffer  next-buffer  LOOP  drop ;      buffers 0 ?DO dup save-buffer next-buffer LOOP drop ;
   
 : empty-buffers ( -- ) \ block  : 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 @      block-buffers @
     buffers 0 ?DO  dup empty-buffer  next-buffer  LOOP  drop ;      buffers 0 ?DO dup empty-buffer next-buffer LOOP drop ;
   
 : flush ( -- ) \ block  : flush ( -- ) \ block
       \G Perform the functions of @code{save-buffers} then
       \G @code{empty-buffers}.
     save-buffers      save-buffers
     empty-buffers ;      empty-buffers ;
   
 ' flush IS flush-blocks  ' flush IS flush-blocks
   
 : get-buffer ( n -- a-addr ) \ gforth  : get-buffer ( u -- a-addr ) \ gforth
     buffers mod buffer-struct %size * block-buffers @ + ;      0 buffers um/mod drop buffer-struct %size * block-buffers @ + ;
   
 : block ( u -- a-addr ) \ block- block  : block ( u -- a-addr ) \ gforthman- block
   \G @var{u} identifies a block number. Assign a block buffer to @var{u},      \G If a block buffer is assigned for block @i{u}, return its
   \G make it the current block buffer and return its start      \G start address, @i{a-addr}. Otherwise, assign a block buffer
   \G address, @var{a-addr}.      \G for block @i{u} (if the assigned block buffer has been
     dup 0= -35 and throw      \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 get-buffer >r
     dup r@ buffer-block @ <>      dup r@ buffer-block @ <>
     r@ buffer-fid @ block-fid @ <> or      r@ buffer-fid @ block-fid @ <> or
Line 141  Defer flush-blocks Line 164  Defer flush-blocks
     r> dup last-block ! block-buffer ;      r> dup last-block ! block-buffer ;
   
 : buffer ( u -- a-addr ) \ block  : 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 ( -- a-addr ) \ block-ext  User scr ( -- a-addr ) \ block-ext s-c-r
     \G USER VARIABLE @var{a-addr} is the address of a cell containing      \G @code{User} variable -- @i{a-addr} is the address of a cell containing
     \G the block number of the block most recently processed by      \G the block number of the block most recently processed by
     \G @code{LIST}.      \G @code{list}.
     0 scr !  0 scr !
   
   \ nac31Mar1999 moved "scr @" to list to make the stack comment correct
 : updated?  ( n -- f ) \ gforth  : updated?  ( n -- f ) \ gforth
     scr @ buffer      \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 + @ ;      [ 0 buffer-dirty 0 block-buffer - ] Literal + @ ;
   
 : list ( u -- ) \ block  : 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.      ." Screen " u.
     updated?  0= IF ." not "  THEN  ." modified     " cr      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
Line 172  User scr ( -- a-addr ) \ block-ext Line 209  User scr ( -- a-addr ) \ block-ext
   THEN ;    THEN ;
   
 ' (source) IS source ( -- c-addr u ) \ core  ' (source) IS source ( -- c-addr u ) \ core
 \G @var{c-addr} is the address of the input buffer and @var{u} is the  \G @i{c-addr} is the address of the input buffer and @i{u} is the
 \G number of characters in it.  \G number of characters in it.
   
 : load ( i*x n -- j*x ) \ block  : load ( i*x n -- j*x ) \ block
   push-file      \G Save the current input source specification. Store @i{n} in
   dup loadline ! blk ! >in off ['] interpret catch      \G @code{BLK}, set @code{>IN} to 0 and interpret. When the parse
   pop-file throw ;      \G area is exhausted, restore the input source specification.
       loadfilename# @ >r
 : thru ( i*x n1 n2 -- j*x ) \ block      2 loadfilename# ! \ "\a block/"
   1+ swap ?DO  I load  LOOP ;      push-file
       dup loadline ! blk ! >in off ['] interpret catch
 : +load ( i*x n -- j*x ) \ block      pop-file
       r> loadfilename# !
       throw ;
   
   : 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 ;      blk @ + load ;
   
 : +thru ( i*x n1 n2 -- j*x ) \ block  : +thru ( i*x n1 n2 -- j*x ) \ gforth
   1+ swap ?DO  I +load  LOOP ;      \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}.
 : --> ( -- ) \ block- block      1+ swap ?DO  I +load  LOOP ;
     \G If this symbol is encountered whilst loading block @var{n},  
     \G discard the remainder of the block and load block @var{n+1}. Used  : --> ( -- ) \ gforthman- gforth chain
     \G for chaining multiple blocks together as a single loadable unit.      \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      refill drop ; immediate
   
 : block-included ( addr u -- ) \ gforth  : 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      block-fid @ >r block-fid off open-blocks
     1 load block-fid @ close-file throw flush      1 load block-fid @ close-file throw flush
     r> block-fid ! ;      r> block-fid ! ;

Removed from v.1.23  
changed lines
  Added in v.1.33


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