Fetch svn file renames/moves

svn_get_renames.py

#!/usr/bin/env python2.7
import argparse
from difflib import SequenceMatcher
import os
import subprocess
import tempfile


def main(old_url, new_url, threshold):
    OLD_URL, OLD_REV = old_url.split('@')
    NEW_URL, NEW_REV = new_url.split('@')

    command = 'svn diff --old {}@{} --new {}@{} --summarize'
    command = command.format(OLD_URL, OLD_REV, NEW_URL, NEW_REV)
    stdout = subprocess.check_output(command, shell=True)
    stdout = stdout.split('\n')
    added = [std.strip('A').strip() for std in stdout if std.startswith('A')]
    removed = [std.strip('D').strip() for std in stdout if std.startswith('D')]
    
    
    pairing = {}
    # Those files removed to search
    removed_search = removed
    for afile in added:
        fh_a = tempfile.NamedTemporaryFile()
        fh_a.close()
        command = 'svn export -q {}@{} {}'.format(afile, NEW_REV, fh_a.name)
        subprocess.call(command, shell=True)
    
        if not os.path.isfile(fh_a.name):
            continue
        with open(fh_a.name, 'r') as added:
            added_text = added.read()
    
            found = False
            for ind in range(len(removed_search)):
                dfile = removed_search[ind]
                fh_d = tempfile.NamedTemporaryFile()
                fh_d.close()
                command = 'svn export -q {}@{} {}'.format(dfile, OLD_REV, fh_d.name)
                subprocess.call(command, shell=True)
    
                if not os.path.isfile(fh_d.name):
                    continue
    
                with open(fh_d.name, 'r') as deleted:
                    deleted_text = deleted.read()
                    m = SequenceMatcher(None, deleted_text, added_text)
                    if m.ratio() > threshold:
                        pairing[dfile] = afile
                        found = True
                        break
            if found:
                # Remove the match so that it doesn't contribute to the next
                # search.
                removed_search.pop(ind)
    
    
    #for ind in range(len(stdout)):
    #    change = stdout[ind].split()
    #    if change[-1] in pairing.keys():
    #        continue
    #    if change[-1] in pairing.values():
    #        continue
    #    print chane
    for key, value in pairing.iteritems():
        print '{}\t->\t{}'.format(key,value)


if __name__ == '__main__':
    msg = ('List all files that are identified as being renamed/moved using a '
           'heuristic for determining percentage similarity.')
    parser = argparse.ArgumentParser(description=msg)
    parser.add_argument('old', help='OLD-URL[@OLDREV]')
    parser.add_argument('new', help='NEW-URL[@NEWREV]')
    msg = ('Threshold, defaults to 0.7 when not specified.  Similarities '
           'between files greater than this fraction are flagged as a '
           'rename/move.')
    parser.add_argument('--threshold', type=float, default=.7, help=msg)
    args = parser.parse_args()
    main(args.old, args.new, args.threshold)