Thank you Ryan
I appreciate the correction about "HEAD" Version; I sure did have it
I will try your suggestions, which implies I will have to install it on a
LINUX machine, or install a PERL interpreter in a windows machine.
Right now I am running it local in a Windows machine, simply because there
is where all the development is happening -- I (will try to get it running
on a LAMP server which I have available for this kind of things.
From: Ryan Schmidt [mailto:firstname.lastname@example.org]
Sent: Thursday, April 19, 2007 6:33 PM
To: Mail List
Subject: Re: Adding existing project to new install
On Apr 19, 2007, at 14:57, Mail List wrote:
> I have been working on a project for the last 6 weeks.
> I have saved every week a complete copy of the progress made in
> separate folders each named by the week the work was completed, and
> then the folder gets duplicated under the new week name.
> Now I found out about subversion and installed, I want to add these
> folder collection to the repository ( 99% of the content is the
> same, there are just the few changes made during each work week)
> I would like to
> Feb_week1 to be the head version
> Feb_week2 - Be imported / added as Version 2
> Feb_week3 - Be imported / added as Version 3
> Feb_week4 - Be imported / added as Version 4
> I want to keep the HEAD version INTACT in case I have to revert
> back to it, but also I would like to add the work done after the
> head version in to the system.
First of all, you are misunderstanding what "HEAD" means. HEAD is the
most-recent version of the repository, not the first version.
You can accomplish your task by importing Feb_week1 as usual with the
"svn import" command, then using the "svn_load_dirs.pl" script to
load the remaining weeks on top of the first one. The
svn_load_dirs.pl script will also "tag" your imported versions with a
name you give it, so that you can refer to these stages by name
later. You should also tag the version you import initially.
svn_load_dirs.pl is included with the Subversion source distribution
(but not with the binary distributions). Here is its read-me:
And here's the script itself:
(Replace "@SVN_BINDIR@" in the script with the prefix where
Subversion is installed on your system.)
Here's how you would import your directories:
### Create a repository.
svnadmin create /path/to/myrepo
### Set up empty directories for the project.
svn mkdir \
-m "Creating directory for myproject"
svn mkdir \
-m "Creating branches directory for myproject"
svn mkdir \
-m "Creating tags directory for myproject"
### Import the initial version of the project as the trunk and tag it.
svn import \
-m "Importing Feb_week1 into trunk"
svn copy \
-m "Tagging trunk as Feb_week1"
### Import and tag subsequent versions using svn_load_dirs.pl
svn_load_dirs.pl -t Feb_week2 $PROJECT trunk tags/Feb_week2
svn_load_dirs.pl -t Feb_week3 $PROJECT trunk tags/Feb_week3
svn_load_dirs.pl -t Feb_week4 $PROJECT trunk tags/Feb_week4
At the end of all this, your repository will have this structure:
The tags represent the snapshots of your work at different points in
time. At this point the trunk is identical to the Feb_week4 tag, but
now you will check out a working copy of the trunk and continue to
improve it, committing after every logical change (more often than
once a week, please!). When you reach another point that you would
like to remember with a snapshot, you can make a new tag from the
trunk again using "svn copy" as above.
Any questions? Please ask! You can also read more about
svn_load_dirs.pl and this strategy in the book chapter on Vendor
Branches (because this strategy is most often used to import sources
from 3rd-party vendors):
To reply to the mailing list, please use your mailer's Reply To All
To unsubscribe, e-mail: email@example.com
For additional commands, e-mail: firstname.lastname@example.org
Received on Fri Apr 20 13:46:25 2007