efs 1.11 (beta) available

sandy@ibm550.sissa.it
Thu, 1 Sep 1994 01:01:09 +0200


**********************************************************************
* Note that anubis.ethz.ch is currently down for a hardware upgrade, *
* so the efs 1.11 files do not yet exist on it.  I'll be putting     *
* them there when it comes back up.                                  *
**********************************************************************

efs-1.11 is now available at

Switzerland:

   /anonymous@anubis.ethz.ch:/sandy/efs/efs-1.11.tar.gz

Massachusetts, USA:

  /anonymous@alpha.gnu.ai.mit.edu:/efs/efs-1.11.tar.gz

California, USA:

  /anonymous@ftp.hmc.edu:/pub/emacs/packages/efs/efs-1.11.tar.gz
  (After a new release, will be updated at 2:00 PST)

For HP people, it's on the HP internet:

  /anonymous@cuckoo.hpl.hp.com:pub/tools/efs-1.11.tar.gz

I.P. addresses are:  alpha.gnu.ai.mit.edu = 128.52.46.18
                     anubis.ethz.ch = 129.132.61.1
                     ftp.hmc.edu = 134.173.32.18

Included is dired-7.4. A dired only distribution is available as

   /anonymous@anubis.ethz.ch:/sandy/dired/dired-7.4.tar.gz

efs 1.11 is still beta test code.

Version 2.12 of reporter.el is back being bundled with efs.  Older
versions do not deal with references to unbound variables.

Also, Emacs 18 users need comint.el to run efs.  We have put copies of
comint V2.03 on all the efs distribution sites, as well.  Note that
comint V2.03 is for Emacs 18 only.  Emacs 19 users neither need it,
nor should they install it.

Changes from 1.10 to 1.11:

dired:

1.  New variable dired-print-command-alist.  It replaces
    efs-print-command, and allows print commands for files to be chosen
    based on the file name syntax.

2.  *Dired Background Shell Output* buffer is now in comint-mode.
    Also, background shell commands now run with pty's instead of
    pipes.  These two changes mean that you can type responses to
    confirmation prompts in the buffer.

3.  Fixed bug in dired-report-bug.  It was inadvertently dumping the
    efs state.

4.  Since dired-do-uucode looks at the header of each file to
    determine what the output file will be, I now have it also
    decide whether to encode or decode at the same point.  It used to
    make this decision based on whether the file name ended with ".uu"
    or not.

5.  Fixed bug where dired might incorrectly insert a file line into
    wildcard buffers, when updating dired buffers.  Bug found by
    Jost Krieger.  Thanks.

6.  Fixed bug where dired would lose the file line marker when making
    backups of the file.

7.  Changed "C-x d" so that if you have a wildcard buffer on /tmp/f*
    and do "C-x d /tmp", a non-wildcard buffer for /tmp would get
    created, instead of just making the f* buffer current.  This is
    independent of the setting of dired-find-subdir.  That only
    applies to whether dired will go to a buffer which contains the 
    subdir as a proper subdirectory.

8.  Fixed bug when using "1 s" to add "R" to listing switches.


efs:

1.  Fixed bug where efs-dired.el was clobbering the supported prefix
    arg for dired-find-file-other-window.

2.  Fixed up MULE support with some more patches from Ishikawa Ichiro.

3.  Support for NOS/VE now.  Thanks to Jost Krieger for helping with
    this.

4.  dired-do-compress can now do in-place compression of remote files,
    if the FTP server supports SITE EXEC, and the desired compression
    program is in the FTP exec path of the server.  efs should be able
    to determine automatically if this will work, and will use the
    previous GET-compress-file-PUT sequence if it doesn't.  Both
    methods of compressing work asynchronously.

5.  efs-dynix.el is no longer.  This only supported a hacked up server
    at the University of Lancaster anyway.  It seems that this server
    has since been replaced.

6.  Renaming files from one host to another now works for directories.
    Of course, what really happens is a series of copies and
    deletions.  From dired, this works asynchronously.  Note that
    regardless of how many files are actually involved, this is
    treated as an atomic operation.  This means that it occupies one
    slot in the FTP command queue.  If a request is made for a
    synchronous file transfer (eg. find-file), efs waits until the
    currently executing entry in the queue completes, before
    processing the synchronous command ahead of the next waiting
    command in the queue.  This means that renaming a large directory
    tree from one host to another may tie up the FTP connection for a
    while.

7.  Running remote shells, efs-dired.el now sends the "-l USER"
    argument, if efs-remote-shell-takes-user is non-nil.  Set this
    according to the capabilities of your operating system.  An
    attempt is made to guess at what this should be.  Let us know if
    the logic in the default setting for this variable is incomplete.
    It probably is.

8.  Michael Sperber patched a bug where efs.el was munging the
    minor-mode-alist.  Thanks.

9.  Fixed bug that efs-report-bug was referencing the non-existent
    variable efs-smart-gateway.

10. For gateway users, tuned the logic that efs uses to determine the
    most efficient route to transfer data between two hosts.  It now
    treats efs-gateway-host as living in both your local domain and
    the outside world.  This means that even if you are using a
    sockified gateway FTP (and hence have efs-gateway-type set to nil)
    it still pays to set efs-gateway-host.

11. Rewrote the .netrc file parser.  We support # as a comment char.
    The # must be at the beginning of a word or surrounded by white
    space.  Things like "foo#bar" are not comments.  This allows
    references to port numbers in the .netrc file in a way so that efs can
    provide port number completion for hosts using the nonstandard port.
    The text from # to end of line is treated as a comment.