[BACK]Return to BUGS CVS log [TXT] [DIR] Up to [gforth] / gforth

Diff for /gforth/BUGS between versions 1.12 and 1.57

version 1.12, 1995/01/25 18:53:42 version 1.57, 2005/12/27 09:35:45
Line 1 Line 1
 name> does not take the same argument as e.g. .name. Remedy: add cell+  You can find more recent bug reports at
 before name>, but adapt all uses.  anton 23apr94  http://savannah.gnu.org/bugs/?group=gforth
   
 revealing the same name several times (e.g., by using recursive)  gforth.el: indentation does not work right on the first line of a
 results in redefined messages.  anton 28jul94  buffer. anton 27jan95
   
 [IF] is case-sensitive.  anton 2aug94  Conditional compilation continues after the file ends. This is allowed
   by the standard (through an ambiguous condition), but the compiler
 if blocks.fb does not exist, 1 block creates the file, but cannot  should at least produce a warning.  anton 27jan95
 read-file from it. Only if the file-id has been created with  
 open-file, not create-file, read-file works. - anton 6aug94  QUERY and TIB may behave differently than some may expect. TIB always
   points to the current SOURCE, and QUERY puts its result there.  anton
 etags.fs crashes one of my applications (gs.fs). anton 12jan95  28nov96
   
 f. suppresses all digits when it prints 0:  open-path-file expands "./" into the sourcefilename. It should either
 0e0 f. .  ok  not expand "./" or provide a mechanism that allows the application to
 There's also one other problem with f.:  determine what "./" should expand to. anton 16jun98
 1e-20                    f. 0.00000000000000000001000000000000001  ok  
 -20e0 falog              f. 0.00000000000000000001000000000000001  ok  gforth.el: Typing <ret> does not work when tabs separate words in a
 0.00000000000000000001e0 f. 0.00000000000000000001000000000000001  ok  line, and there is a \-comment at the end of the line.  anton 19feb00
 All this happens under Slackware Linux. On the DecStation I get a  
 similar error in the other direction.  anton 17jan95  SEE does not work when the control structure is too complex (e.g.,
   load http://www.complang.tuwien.ac.at/forth/pentomino.fs and then do
 not all aliases are in the etags file. Bug in etags.fs? anton 24jan95  SEE NEXT-PIECE). anton 5mar2000
   
 emacs often finds the wrong tag. anton 24jan95  Include cannot handle lines longer than 255 characters.  anton 4sep00
   
 source after a locals definition is ignored, e.g., in  Errors happening during a LOAD do not report the offending word and
 { a } ." xxx"  its context (e.g., the 64-byte line). anton 8sep00
 the `." xxx"' is not executed.  anton 24jan95  
   Our ecvt routine apparently does not work correctly for Infs and NaNs.
   Try "ac_cv_func_ecvt=no ./configure; make" and then in Gforth: "1e 0e
   f/ f. 0e 0e f/ f.". anton 25sep00
   
   Our ecvt routine does not round correctly, e.g., 0.25->0.3. Marcel
   Hendrix 3oct00 <8rdcmd$j96$1@news.IAEhv.nl>
   
   Newline has only LF (instead of CRLF) in DOS.  Bruce Hoyt 25oct2000
   <39f7b14b$2@clear.net.nz>
   
   F. does not print trailing zeroes (e.g., "10 SET-PRECISION 125e f.")
   anton 31may01
   
   Block 0 does not work as it should: "0 block drop update save-buffers"
   does not write to the blocks file.  "0 block 1024 dump" seems to give
   the previous contents of the buffer.  Travis Bemann 10jul2001
   <3b4b4f57$0$42883$272ea4a1@news.execpc.com>
   
   When accessing a block beyond the end of the block file, the result is
   filled with spaces (this is also documented).  However, when accessing
   a previously unwritten block before the end of the block file, we will
   get a block full of zeroes on most (all?) OSs.  This inconsistency
   should be eliminated and the documentation fixed.  anton 14jul2001
   
   Path handling may have a bug:
   strace $FORTH 'fpath= .|~+/gforth s" /home/anton/threading-variations/gforth/mach32b.fs" include gforth/kernel/main.fs bye'
   tries to open gforth/kernel/gforth/kernel/aliases.fs, then
   gforth/gforth/kernel/aliases.fs, but not gforth/kernel/aliases.fs.
   anton 3 sep 2001
   
   gforth-native does not deal correctly with inline arguments in static
   superinstructions: when using lit_lit, the wrong literals are put into
   the native code (probably due to wrong assumptions about argument
   offsets). 6 nov 2003
   
   finish-code changes the stack for gforth-fast with --enable-force-reg
   (probably for gforth-native too).  Apparently ecx (spTOS) is neither
   saved by engine, nor by finish_code().
   
   Wrong decompilation:
   create foo
   5 constant bar
   : xxx foo bar ;
   simple-see xxx
   16 nov 2005
   
   Debian's gforth's etags.fs crashes (probably due to path problems)
   on b3:
   gforth etags.fs
   16 nov 2005
   
   -----
   Copyright (C) 1995,1996,1997,1998,2000,2003 Free Software Foundation, Inc.
   
   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.
Removed from v.1.12  
changed lines
  Added in v.1.57

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