From 7feaddceb111be033d0f773621c3bb8b13cf0fed Mon Sep 17 00:00:00 2001
From: Lars Wirzenius <liw@liw.fi>
Date: Sun, 18 Apr 2010 14:21:56 +1200
Subject: Improve progress reporting.

---
 dupfiles | 54 ++++++++++++++++++++++++++++++++----------------------
 1 file changed, 32 insertions(+), 22 deletions(-)

diff --git a/dupfiles b/dupfiles
index cc1ff47..93287f5 100755
--- a/dupfiles
+++ b/dupfiles
@@ -77,32 +77,42 @@ class DuplicateFileFinder(object):
     # groups. If a group becomes just a single file, that file is not
     # a duplicate and can be dropped completely.
     def duplicates(self):
-        total_files = sum(len(x) for x in self.by_size.itervalues())
-        done_files = 0
+        skip = [size for size in self.by_size if len(self.by_size[size]) == 1]
+        for size in skip:
+            del self.by_size[size]
+
+        total_bytes = sum(len(tuples) * size
+                          for size, tuples in self.by_size.iteritems())
+        done_bytes = 0
         result = []
-        for tuples in self.by_size.itervalues():
-            if len(tuples) > 1:
-                by_checksum = dict()
-                for dev, ino, pathname in tuples:
-                    checksum = self.file_checksum(pathname)
-                    if checksum not in by_checksum:
-                        by_checksum[checksum] = set()
-                    by_checksum[checksum].add(pathname)
-                    done_files += 1
-                    self.duplicates_progress(done_files, total_files)
-                for names in by_checksum.itervalues():
-                    if len(names) > 1:
-                        result.append(names)
-            else:
-                done_files += 1
-                self.duplicates_progress(done_files, total_files)
+        for size, tuples in self.by_size.iteritems():
+            by_checksum = dict()
+            for dev, ino, pathname in tuples:
+                checksum = self.file_checksum(pathname)
+                if checksum not in by_checksum:
+                    by_checksum[checksum] = set()
+                by_checksum[checksum].add(pathname)
+                done_bytes += size
+                self.duplicates_progress(done_bytes, total_bytes)
+            for names in by_checksum.itervalues():
+                if len(names) > 1:
+                    result.append(names)
         self.progress.finished()
         return result
 
-    def duplicates_progress(self, done_files, total_files):
-        self.progress.write('%d/%d (%.1f%%) files done' %
-                            (done_files, total_files, 
-                                100.0 * float(done_files) / total_files))
+    def duplicates_progress(self, done, total):
+        self.progress.write('%s/%s (%.1f%%) bytes done' %
+                            (self.human_size(done), self.human_size(total), 
+                             100.0 * float(done) / float(total)))
+
+    def human_size(self, size):
+        tab = [(1024**3, 'GiB'),
+               (1024**2, 'MiB'),
+               (1024**1, 'KiB')]
+        for limit, unit in tab:
+            if size >= limit:
+                return '%.1f %s' % (float(size) / float(limit), unit)
+        return '0 B'
 
     def file_checksum(self, pathname):
         return hashlib.md5(file(pathname, 'rb').read()).digest()
-- 
cgit v1.2.1