Quick-n-dirty finder for duplicated classes in JAR-files

find-jdups.py

#!/usr/bin/python

# Usage sample:
#  python find-jdups.py dir1 dir2 ... | grep -v = | sort | uniq | awk -F/ '{ print $NF }'

import os
import sys
import zipfile

pathlist = dict()

def store_path(key, val):
    if not key in pathlist:
        pathlist[key] = [val]
        return
    pl = pathlist[key]
    if pl[0] != val:
        pl.append(val);

def read_jar(jarpath):
    jar = zipfile.ZipFile(jarpath)
    dir = jar.namelist()
    for n in dir:
        if '.class' in n:
            store_path(n, jarpath)

def read_dir(dir):
    for root, dirs, files in os.walk(dir):
        for name in files:
            if '.jar' in name:
                read_jar(os.path.join(root, name))

def print_dups():
    for cls in pathlist:
        jars = pathlist[cls]
        if len(jars) > 1:
            print("%s = %d" % (cls, len(jars)))
            for j in jars:
                print("\t%s" % (j))

for n in range(1, len(sys.argv)):
    read_dir(sys.argv[n])
print_dups()