Ruby script to make a release from SVN trunk.

Assumes a repository structure of:

/path/to/Trunk
/path/to/Releases/YYYY-MM-DD--XX

Should be easy enough to modify to suit your needs. Comments welcome :)

Example usage:

vi /usr/local/bin/release (and dump snippet into it)
chmod +x /usr/local/bin/release
cd /path/to/checked-out-trunk
release

#!/usr/bin/ruby

today = Time.now.strftime('%Y-%m-%d')

if `svn info`.to_a[1] !~ /^URL: (.*)\/Trunk$/
        puts "Repository invalid. Must be in Trunk."
        exit
end

repo_url = $1

repo = repo_url.split('/').last

release_url = repo_url + "/Releases"

if `svn ls #{release_url}`.to_a.last =~ /#{today}--(\d+)/
        num = $1.to_i + 1
else
        num = 1
end

release = "#{today}--#{num}"

cmd = "svn copy #{repo_url}/Trunk #{release_url}/#{release} -m 'copy #{repo} trunk to release #{release}'"
puts cmd
puts

print "Execute command? "

if gets.chomp.upcase != 'Y'
        puts "Nothing done."
        exit
end

`#{cmd}`
puts "Copied as #{release}."
puts "Execute on the remote machine:"
puts "svn switch #{release_url}/#{release}"

Bash script to add all new files to Subversion

for i in `svn st | grep ? | awk -F "      " '{print $2}'`; do svn add $i; done

Just run this code from inside your repository. This will do a bulk action like "svn add" on each file marked with the given filter (like '?') when doing "svn status"

Beware: this code might break with multi word filenames...

usage :
# ./svn_bulk ? add ---> this shows a preview of the actions
# ./svn_bulk ? add 1 ---> action !

#!/usr/bin/ruby
sign, action, doit = ARGV
sign = '\?' if sign == '?'
list = IO.popen("svn st")
list.each {|i|
if (i =~ /^#{sign}\s*(.*)/) 
cmd = "svn #{action} '"+$1+"'"
print cmd + "\n"
system(cmd) if doit
end
}

Taken and slightly adapted from www.robgonda.com/blog

svnadmin dump /path/to/repos > proj.dump
cat proj.dump | svndumpfilter exclude somefolder > cleanproj.dump
STOP SVN services
BACKUP /path/to/repos/conf /path/to/repos/hooks (all custom configuration for this repository)
DELETE /path/to/repos
svnadmin create /path/to/repos
RESTORE /path/to/repos/conf /path/to/repos/hooks
svnadmin load /path/to/repos < cleanproj.dump
RESTART SVN services

This is still a work in progress. The objective is to script the entire initial svn hassle that needs to be done with a rails project. The first import, then the first checkout, and then dealing with all the files that need to be ignored. Based heavily on:

http://wiki.rubyonrails.org/rails/pages/HowtoUseRailsWithSubversion

In the future I may add capistrano setup and such.

This worked the last time I tried it, but I don't expect it to work the next time.

#!/bin/bash
set -v                         # verbose output

## USAGE: configure the following variables, and execute in your rails root, ie the directory with
## config/ and app/
## NOTE: This script assumes that your directory of your rails root has the same name as both your 
## application, and your svn repo.
##
## This script also assumes that it's ok to make a backup in the parent directory of your
## rails root.

username="jonshea"               # CHANGE ME!!!!!!!
svn_url="http://jonshea.com/svn/" # CHANGE ME!!!!!

## This is still a work in progress. The objective is to script the entire initial svn hassle that 
## needs to be done with a rails project. The first import, then the first checkout, and then 
## dealing with all the files that need to be ignored. Based heavily on: 
##
## http://wiki.rubyonrails.org/rails/pages/HowtoUseRailsWithSubversion
##
## It designed only for freshly created applications, that haven't really be changed at all from the
## original generation script

app_dir=`pwd`
app_name=`basename $app_dir`
svn_url_to_your_repository=${svn_url}${app_name} # Assumes your repo has the same name as your app
echo $app_name
echo $svn_url_to_your_repository

## Do the initial import
svn import . ${svn_url_to_your_repository}/trunk -m "First Import" --username $username



cd ..                           # Back out a directory from the root

## We're going to make a backup of your app. If one is already there, then remove it.
test -d ./pre_svn_backup_$app_name || rm -rf pre_svn_backup_$app_name
mkdir ../pre_svn_backup_$app_name
mv -f $app_dir ./pre_svn_backup_${app_name} # Move the rails app to the backup dir.

mkdir $app_dir # recreate the application directory
cd $app_dir
svn checkout ${svn_url_to_your_repository}/trunk . # Check out the subversion repo to the app directory

## This section cleans up the svn repo, so that you're not versioning things that shouldn't be versioned.
svn remove log/*
svn commit -m 'removing all log files from subversion'
svn propset svn:ignore "*.log" log/
svn update log/
svn commit -m 'Ignoring all files in /log/ ending in .log'

svn remove tmp/
svn commit -m 'removing the temp directory from subversion'
svn propset svn:ignore "*" tmp/
svn update tmp/
svn commit -m 'Ignore the whole tmp/ directory, might not work on subdirectories?'

svn move config/database.yml config/database.example
svn commit -m 'Moving database.yml to database.example to provide a template for anyone who checks out the code'
svn propset svn:ignore "database.yml" config/
svn update config/
svn commit -m 'Ignoring database.yml'

svn move public/dispatch.rb public/dispatch.rb.example
cp public/dispatch.rb.example public/dispatch.rb
svn move public/dispatch.cgi public/dispatch.cgi.example
cp public/dispatch.cgi.example public/dispatch.cgi
svn move public/dispatch.fcgi public/dispatch.fcgi.example
cp public/dispatch.fcgi.example public/dispatch.fcgi
svn commit -m 'Moving dispatch.(star) to dispatch.(star).example to provide a template.'

svn propedit svn:ignore public/ dispatch.rb
svn propedit svn:ignore public/ dispatch.cgi
svn propedit svn:ignore public/ dispatch.fcgi
svn update public/
svn commit -m 'Ignoring dispatch.* files'

svn propedit svn:ignore db/ *.sqlite
svn propedit svn:ignore db/ *.sqlite3
svn commit -m 'Ignore database files'




#cap --apply-to $app_dir $app_name
#svn add config/deploy.rb     
#svn add lib/tasks/capistrano.rake

exit 0

I have all of my projects in a directory like ~/PROJECTS. Each day when I come to work, I want to ensure that everything on my workstation is up to date. Rather than cd into each directory and issue "svn update" by hand, I kludged the following shell script together. Comments appreciated.

 for X in `find . -maxdepth 2 -type d -name ".svn" | cut -d / -f 2`; do svn update ./$X; done  

svn --username anon --password anon co http://macromates.com/svn/Bundles/trunk/Bundles /Users/ned/Library/Application\ Support/TextMate/Bundles

Installing Subversion for local use is general an easy install, but allowing remote access to your svn repository over SSH can be problomatic dependent upon your OS and the means taken to install.

For Darwinports and Fink on OS X the install location has to be added to users $PATHs, but there are extra steps outlined here for use of the svn+ssh means of access:

http://subversion.tigris.org/faq.html#ssh-svnserve-location

A much easier alternate is to sym link the svn binaries to a place on the default PATH (used by the SSH login):

#For Darwinports
ln -s /opt/local/bin/sv* /usr/bin/

#For Fink
ln -s /sw/bin/sv* /usr/bin/

This links all the binaries at once.

svn add `svn st | grep '\?' | tr '\?' ' ' | xargs`

If you use the same repository for mostly everything, you'll occasionally find it amusing that your brand new project is at r1800 or that that old project not touched since r60 now shares r2385 with everyone else.

Well, amusing != useful, so here's what I use when I'm curious about how many commits a certain project took me:

svn log --stop-on-copy | grep -e "--------" | wc -l

find . -name .svn -print0 | xargs -0 rm -rf

Update: Thankyou iburrell

For some odd reason tortoise SVN loves to commit empty commit sets. So I managed to get the following pre-commit script working under FreeBSD, seeing that the pre-commit.tmpl does not work as expected.

#!/usr/local/bin/python
"""
Subversion pre-commit hook which currently checks that the commit contains
a commit message to avoid commiting empty changesets which tortoisesvn seems
to have a habbit of committing.

Based on http://svn.collab.net/repos/svn/branches/1.2.x/contrib/hook-scripts/commit-block-joke.py
and hooks/pre-commit.tmpl

Hacked together by Jacques Marneweck 

$Id$
"""

import sys, os, string

SVNLOOK='/usr/local/bin/svnlook'

def main(repos, txn):
    log_cmd = '%s log -t "%s" "%s"' % (SVNLOOK, txn, repos)
    log_msg = os.popen(log_cmd, 'r').readline().rstrip('\n')

    if len(log_msg) < 10:
        sys.stderr.write ("Please enter a commit message which details what has changed during this commit.\n")
        sys.exit(1)
    else:
        sys.exit(0)

if __name__ == '__main__':
    if len(sys.argv) < 3:
        sys.stderr.write("Usage: %s REPOS TXN\n" % (sys.argv[0]))
    else:
        main(sys.argv[1], sys.argv[2])

This saves considerable time, and it's recursive:

svn add --force *

You can run --force to make an SVN export overwrite a folder rather than write into it

cd /usr/local/etc/
svn export --force --username user@textdrive.com --password apassword https://svn.textdrive.com/repos/setup/usr/local/etc/apache2

In a shell, cd to the directory from which you'd like to serve the site, and rename or delete any '.htaccess' or 'index.php' files, and any directories called 'images' or 'files', that may be there.

Checkout the current code:

svn co http://svn.textpattern.com/current/ .

(Note the dot at the end). To complete the installation, just load http://thesiteurl/textpattern/ in a browser, fill in your database info, and when presented with the config.php block, copy it, go back to the shell client and:

nano textpattern/config.php

Paste in the code block, save and exit (ctrl-x, then y, then return).

Back to the browser, click through, add personal details, and that, as they say, is it.

Whenever you want to update to the latest code, cd to the same directory and:

svn update

...and the latest code will be seamlessly integrated with your installation.