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

Diff for /gforth/BUGS between versions 1.5 and 1.52

version 1.5, 1994/08/10 17:22:33 version 1.52, 2003/02/03 08:59:28
Line 1 Line 1
 name> does not take the same argument as e.g. .name. Remedy: add cell+  gforth.el: indentation does not work right on the first line of a
 before name>, but adapt all uses.  anton 23apr94  buffer. anton 27jan95
   
 revealing the same name several times (e.g., by using refcursive)  Conditional compilation continues after the file ends. This is allowed
 results in redefined messages.  anton 28jul94  by the standard (through an ambiguous condition), but the compiler
   should at least produce a warning.  anton 27jan95
   
 [IF] is case-sensitive.  anton 2aug94  QUERY and TIB may behave differently than some may expect. TIB always
   points to the current SOURCE, and QUERY puts its result there.  anton
   28nov96
   
 gforth.el: fill-paragraph should either not work, or it should make sure that comments stay comments and code stays code.  anton 2aug94  open-path-file expands "./" into the sourcefilename. It should either
   not expand "./" or provide a mechanism that allows the application to
   determine what "./" should expand to. anton 16jun98
   
 if blocks.fb does not exist, 1 block creates the file, but cannot  gforth.el: Typing <ret> does not work when tabs separate words in a
 read-file from it. Only if the file-id has been created with  line, and there is a \-comment at the end of the line.  anton 19feb00
 open-file, not create-file, read-file works. - anton 6aug94  
   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
   
   Include cannot handle lines longer than 255 characters.  anton 4sep00
   
   Errors happening during a LOAD do not report the offending word and
   its context (e.g., the 64-byte line). anton 8sep00
   
   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>
   
   SIGPIPE can make Gforth hard to stop (e.g., 
   gforth -m 2M wordfreq.fs -e bye|head
   anton 30may01
   
   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
Removed from v.1.5  
changed lines
  Added in v.1.52

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