bug-cvs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

AW: "cvs move" command


From: Schoenebeck Christian
Subject: AW: "cvs move" command
Date: Wed, 15 Jan 2003 10:46:16 +0100

>-----Ursprüngliche Nachricht-----
>Von: Kaz Kylheku [mailto:kaz@ashi.footprints.net] 
>Gesendet: Mittwoch, 15. Januar 2003 02:41
>An: bug-cvs@gnu.org
>Betreff: Re: "cvs move" command
>
>A fellow by the name of David D. ``Zoo'' Kuhn presented some 
>ideas in 1994 that resemble Meta-CVS. Here is a link to the 
>November 7, 1994 message archived at geocrawler:
>
>http://www.geocrawler.com/mail/msg.php3?msg_id=2110252

His ideas are almost exactly what I had in mind.

>Here are quotes:
>
>``The mapping between the repository filename and the name the 
>user sees it
>  as is handled in the PDDB.  For example, I might create a 
>file "foobar".
>  A unique name is generated (say "AAA123" and the information 
>is created
>  in the repository using that new name.  The PDDB maps "foobar" into
>  AAA123.  Additional information about the file (owner,
>permissions,etc)
>  are also stored in the PDDB.''

Yes, the working file's name and it's filename in the repository must be
mapped and thus independent, but instead of unmeaning names like "AAA123" I
would propose to use combinations like
<initial-working-file-name><creation-time>, just to ease manual
manipulations in the repository, as this is needed sometimes.

>
>Aha!
>
>``Renaming a file becomes quite simple if we don`t move it to another
>  directory.  The filename mapping needs to change, but the 
>file contents
>  doesn`t change at all.  Renaming a directory uses the exact same
>  principles (and mechanism).''
>
>Aha II!
>
>``Moving a file to another directory becomes a bit of a 
>challenge, since we
>  now have a mapping that spans a single directory.  So some means of
>  linking these PDDB`s must be developed.  I could actually envision a
>  repository structure where EVERY file is stored in the same flat
>  directory.''

My thougts. The only thing I'm concerned about is that the ext2 file system
would have it's problems with this, as it's performance slows down with a
high amount of files in a single directory. But that's a system specific
problem and guess this doesn't account to the new ext3 file system.

I don't know if this is already been mentioned, but I would create a control
file for each directory containing the file mappings, permissiona, etc. of
the files and only the files in the specific directory, instead of one huge
mapping/control file for all files, but I guess this idea suggests itself.

>Note that I found this information sometime I put the first 
>release of Meta-CVS out last January. I started digging to see 
>if anyone ever proposed something similar for CVS, just to see 
>how much time had elapsed since that suggestion. A little over 
>seven years, not bad. ;) 

Yeah, I guess it's time to get implemented :)

Regards, Cuse




reply via email to

[Prev in Thread] Current Thread [Next in Thread]