I'm converting from CVS w/Eclipse.
In CVS, all my Java files started with the following CVS keywords (with
ASCII keyword substitution enabled):
/*
* $Id$
* $Log$
*/
After each update and check-in the file would contain an added copy of the
expanded keywords, here's an actual extract from a file in my CVS repository:
/**
* $Id: AsciiFileReader.java,v 1.4.2.2.6.1.10.8.2.1 2005/01/25 19:11:28
davidm Exp $
* $Log: AsciiFileReader.java,v $
* Revision 1.4.2.2.6.1.10.8.2.1 2005/01/25 19:11:28 davidm
* [MAINTENANCE] merge from Rel 2.6
*
* Revision 1.4.2.2.6.1.10.9 2005/01/05 22:20:08 adrianp
* [BUG00743] Check for log file rollover in the call to ready()
*
*/
Notice that the substitutions are cumulative and are actually present in
the file as a chronological archive.
My questions:
1) How do I enable svn:keyword substitution through Subclipse? (I
don't see anything in Preferences)
2) How do I enable it automatically (as the default) on all new files
as well (for the project) ?
3) How to have Subversion add the substitution as new lines (as CVS
does) in the file and become a permanent part of the checked-in file?
As far as I can tell from the SVN book, the svn:keyword values are actually
properties that SVN magically substitutes in my file right after checking
an update into the repository. New updates override the previous values
(rather than append).
This is a major difference with CVS that is getting no PR !! The CVS
substitution facility generates an incredibly useful history of the updates
embedded right in the file.
How can I accomplish the same using Subclipse/Subversion?
-TIA
David
Received on Mon Feb 28 03:55:47 2005