idup.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 # find duplicate images given a hamming distance threshold.
 # employs dhash to do the heavy lifting.
-# does not recurse into "./_duplicate" so you can dump things there if you wish.
+# doesn't recurse into "./_duplicate/" so you can dump things there if you wish.
 # dependencies: pillow, dhash
 
 import sys, os, os.path, pickle
@@ -11,12 +11,15 @@ import dhash
 def lament(*args, **kwargs):
     print(*args, file=sys.stderr, **kwargs)
 
-def result(diff, ours, theirs): # TODO: rename
-    print("{}\t{}\t{}".format(diff, ours, theirs))
+def result(diff, p1, p2): # TODO: rename
+    print("{}\t{}\t{}".format(diff, p1, p2))
 
 dbname = "idup.db"
 exts = ".jpeg .jpg .png".split()
 
+rootpath = "."
+ignore_dir = os.path.join(rootpath, "_duplicate")
+
 """verbosity:
     -1: only unrecoverable errors.
     0: include failures.
@@ -34,9 +37,7 @@ args = sys.argv[1:]
 
 threshold = int(args[0])
 
-ignore_dir = os.path.join(rootpath, "_duplicate")
-
-paths = {}
+paths = {} # path to hash mapping.
 
 if os.path.exists(dbname) and os.path.getsize(dbname) > 0:
     with open(dbname, "rb") as f:
@@ -55,7 +56,7 @@ for path in paths.keys():
 paths = existing
 
 def compare_hash(h1, h2):
-    # hashes are in byte format, so we have to convert them to integers.
+    # hashes are in byte strings, so we have to convert them to integers.
     i1 = int.from_bytes(h1, byteorder="big")
     i2 = int.from_bytes(h2, byteorder="big")
     # return the hamming distance.