HP3000-L Archives

December 2000, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Ken Hirsch <[log in to unmask]>
Reply To:
Ken Hirsch <[log in to unmask]>
Date:
Thu, 28 Dec 2000 15:26:37 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (85 lines)
There's a big difference.  Even on Unix, there's a difference.

On Unix, there are hard links and symbolic links.  When the number of hard
links to a file goes to zero and the file is not open by an process, the
file is deleted and the space it used is freed.  Hard links may not point
across file systems (just like you can't rename across volumes under MPE),
and hard links to directories are not permitted (except for . and ..).

You can use symbolic links across filesystems and to directories.  Symbolic
links are only used when a process opens them.  A symbolic link may point to
a non-existent file, which is never true of  a hard link.

On MPE/iX, there is only one hard link per file.  One nice thing this allows
is that you can get the name of an open file (from the file handle).  On
Unix, all the hard links are equivalent, so a file's name is indeterminate
if there's more than one link.




----- Original Message -----
From: "Dave Darnell" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Thursday, December 28, 2000 3:03 PM
Subject: Re: MPE misdesign


> Is it not true that all files (except those created with maybe an FOPEN of
> the form that allows a temporary unnamed file) are created with a link,
> otherwise we wouldn't be able to reference the file by name?  That would
> mean that logically, there is no difference (other than link name) between
> the link created by merely creating the file, and the link created with
the
> NEWLINK command?
>
> Maybe an expert on the internals can give us a tutorial on the data
> structures involved.
>
> -d
>
> > -----Original Message-----
> > From: Johnson, Tracy [mailto:[log in to unmask]]
> > Sent: Thursday, December 28, 2000 12:50 PM
> > To: [log in to unmask]
> > Subject: Re: MPE misdesign
> >
> >
> > Once I came across the incongruity that PURGELINK will purge
> > a file whether it is a link or not.
> >
> > When I want to purge a link ONLY, I use MPEX and qualify
> > whether the file is a link using the logical "ISLINK"
> > parameter (defaults to TRUE) as follows:
> >
> > PURGELINK fileset.groupset.accountset(ISLINK)
> >
> > Tracy Johnson
> > MSI Schaevitz Sensors
> >
> > > Ted suggests:
> > >
> > > > Pssst, HP.  Fix this instead!
> > > >
> > > >  Here's my official complaint/warning/whatever about how
> > > :PURGE deals with
> > > >  links.  A :PURGE of a link does *not* purge the link.  It
> > > purges the file
> > > > the
> > > >  link points to.  It's not at all obvious to me why one
> > > would want it to do
> > > >  this and I'd rather not think too long about my recent
> > > experience regarding
> > > >  why one would NOT want it to do this!
> > > >
> > > >  For anyone not terribly familiar with links, be warned --
> > > :PURGE is not the
> > > >  same as :RM.HPBIN.SYS (or :PURGELINK for that matter).
> > > >
> > > >  For anyone else, can you give me some understanding of
> > why this was
> > > designed
> > > >  this way?
> >
>

ATOM RSS1 RSS2