|
Joe, As I understand it, a hard link is a directory entry that points to another directory. A symbolic link is a file containing the path to the other directory. Create hard links by not specifying -s on an ln or *HARD on crtlnk. You can use the -H option to prevent copies from chasing your symbolic links something like: cp -RH dir1 dir2 or QSH CMD('cp -RH dir1 dir2') David Morris >>> joepluta@xxxxxxxxxxxxxxxxx 7/8/2003 6:04:56 PM >>> > From: David Morris > > In your P.S. I read that you are having problems with JAR and ZIP > files. > Since I have used symbolic links quite a bit in the past and haven't > run > into these problems, I thought I would check it out. You are right that > > there appears to be some issues but it appears that a hard link did > work > in my tests. I also found that a symbolic link to a directory always > works > (which is what I must have always done or something broke). I did > find an example in IBM's 4.5 manual saying you should create symbolic > links to the JDK jars you add to the ext directory, implying that it > worked > at one point or that the ext directory/JDK jars are somehow different. When you say a hard link "works", what do you mean? For example, I create a directory (dir1) and in it I place a symbolic link to another directory - (symdir1 which is a link to subdir1). I then copy the first directory to another name (dir1 to dir2). What happens is that dir2 has a directory called symdir1, which is NOT a symbolic link, but is instead a real directory, and it contains real copies of all the contents of the original subdir1. And this makes sense, to a degree. There should be an option for a copy to not dereference symbolic links, but I am no link expert, so maybe I just missed that one. However, this can get VERY ugly. In my setup, I have a folder which contains a link to its parent. By definition, this folder can now never be copied because copying the folder will dereference the link, which will copy the contents of the parent which will copy the original folder which will dereference the link... and so on. Unfortunately, ZIP and JAR both dereference links, and so this sort of recursive link causes the same problem - an endless loop until something blows up. Since my workaround for the hardcoded J2EE file structure requires two links to the same folder, one containing the other, you can probably see that I have a recursive setup. That means I cannot save that folder. Which is no big deal; it's just a couple of links which can be created at installation time, except that another weakness in the J2EE deployment model is that there is no provision for post-installation scripts. Oh well, someday it may grow up to be a real industrial strength deployment model, but until then I will have to work around it. Joe
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.