Diff for /gforth/BUGS between versions 1.8 and 1.57

version 1.8, 1994/09/05 17:36:15 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
 gforth.el: doing a forth-fill-paragraph on the following piece of code  28nov96
 uncomments the first comment line. This may be a bug in  
 fill-paragraph, but documentation says that a paragraph can start  open-path-file expands "./" into the sourcefilename. It should either
 without prefix (hanging indentation), so I guess it's all right. -  not expand "./" or provide a mechanism that allows the application to
 anton 19aug94  determine what "./" should expand to. anton 16jun98
   
 : bb-shortest-paths recursive { bb bb-from start-path -- }  gforth.el: Typing <ret> does not work when tabs separate words in a
     \ compute the  line, and there is a \-comment at the end of the line.  anton 19feb00
     \ paths from bb-from through bb, start-path is the path from  
     \ bb-from to bb  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
   SEE NEXT-PIECE). anton 5mar2000
 etags.fs: The TAGS file for the kernal is strange: there are no tags  
 for some aliases; many constants and variables appear twice. anton  Include cannot handle lines longer than 255 characters.  anton 4sep00
 2sep94  
   Errors happening during a LOAD do not report the offending word and
 error-causing word indication no longer works well. Cause: relies on  its context (e.g., the 64-byte line). anton 8sep00
 name being HERE. Future trouble (once CATCH really restores the input  
 stream): it also relies on SOURCE delivering the line that caused the  Our ecvt routine apparently does not work correctly for Infs and NaNs.
 error. anton 2sep94  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.1.8  
changed lines
  Added in v.1.57


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