Hi folks !
(I hope I won't put oil on the fire and ignate the thread much)
(Maybe I missed some arguments, but I've almost read the whole thread)
Some people says "we shouldn't put revtag at the end of url", why ?
Some people says "it because Subversion is not 'path/file' versioneed but 'directory versioned' ...
Who cares ! Subversion is Repostitory Versionned !
(each RevNo is unique on the whole repository, even with branches !)
if we took that previously sent script :
find src -name \*.java |
while read javafile; do
svn diff http://svn.com/repo;123/$javafile \
http://svn.com/repo;456/$javafile
done
and this one :
find src -name \*.java |
while read javafile; do
svn diff http://svn.com/repo/${javafile}\@\@\123 \
http://svn.com/repo/${javafile}\@\@\456
done
(oupps ! it think I turn out that script with Perl in mind, doesn't matter ...:-)
it will still give you the same result, according the implementation.
Some other says, "we all know the root of the repositories we work on" ...
Wrong ! You never can assume what a SysAdm can do on server side ...
Let assume the famous URL mentionned previously :
svn co http://svn.collab.net/repos/svn/clients/rapidsvn/trunk rapidsvn
Can people say what is the root of the repository ?
(Remember, for those who follow both list : RapidSVN should have seperate repository soon if it not already done)
A SysAdm can decide to "svnadmin dump", filter the dump to replace paths, and than recreate new repository with Location changed in Apache, and then all Clients will be still working correctly, but those "middle revs" will be fooled : it is not the same root anymore ...
my two centimes !
---------------------------------
Lèche-vitrine ou lèche-écran ? Yahoo! Magasinage.
Received on Wed Aug 6 17:24:28 2003