fsdump/fsstats improvements

CHANGES.rst

@@ -5,6 +5,12 @@
 5.6.1 (unreleased)
 ==================
 
+- Readd transaction size information to ``fsdump`` output;
+  adapt `fsstats` to ``fsdump``'s exchanged order for ``size`` and ``class``
+  information in data records;
+  (fixes `#354 <https://github.com/zopefoundation/ZODB/issues/354>_).
+  Make ``fsdump`` callable via Python's ``-m`` command line option.
+
 - Fix UnboundLocalError when running fsoids.py script.
   See `issue 285 <https://github.com/zopefoundation/ZODB/issues/285>`_.
 

src/ZODB/FileStorage/fsdump.py

@@ -23,12 +23,14 @@ from ZODB.utils import u64, get_pickle_metadata
 def fsdump(path, file=None, with_offset=1):
     iter = FileIterator(path)
     for i, trans in enumerate(iter):
+        size = trans._tend - trans._tpos
         if with_offset:
-            print(("Trans #%05d tid=%016x time=%s offset=%d" %
-                  (i, u64(trans.tid), TimeStamp(trans.tid), trans._pos)), file=file)
+            print(("Trans #%05d tid=%016x size=%d time=%s offset=%d" %
+                  (i, u64(trans.tid), size,
+                   TimeStamp(trans.tid), trans._pos)), file=file)
         else:
-            print(("Trans #%05d tid=%016x time=%s" %
-                  (i, u64(trans.tid), TimeStamp(trans.tid))), file=file)
+            print(("Trans #%05d tid=%016x size=%d time=%s" %
+                  (i, u64(trans.tid), size, TimeStamp(trans.tid))), file=file)
         print(("    status=%r user=%r description=%r" %
               (trans.status, trans.user, trans.description)), file=file)
 
@@ -122,3 +124,7 @@ class Dumper(object):
 def main():
     import sys
     fsdump(sys.argv[1])
+
+
+if __name__ == "__main__":
+    main()

src/ZODB/scripts/fsstats.py

@@ -7,7 +7,7 @@ import six
 from six.moves import filter
 
 rx_txn = re.compile("tid=([0-9a-f]+).*size=(\d+)")
-rx_data = re.compile("oid=([0-9a-f]+) class=(\S+) size=(\d+)")
+rx_data = re.compile("oid=([0-9a-f]+) size=(\d+) class=(\S+)")
 
 def sort_byhsize(seq, reverse=False):
     L = [(v.size(), k, v) for k, v in seq]
@@ -50,7 +50,10 @@ class Histogram(dict):
         return mode
 
     def make_bins(self, binsize):
-        maxkey = max(six.iterkeys(self))
+        try:
+            maxkey = max(six.iterkeys(self))
+        except ValueError:
+            maxkey = 0
         self.binsize = binsize
         self.bins = [0] * (1 + maxkey / binsize)
         for k, v in six.iteritems(self):
@@ -104,7 +107,7 @@ def class_detail(class_size):
     # per class details
     for klass, h in sort_byhsize(six.iteritems(class_size), reverse=True):
         h.make_bins(50)
-        if len(filter(None, h.bins)) == 1:
+        if len(tuple(filter(None, h.bins))) == 1:
             continue
         h.report("Object size for %s" % klass, usebins=True)
 
@@ -146,7 +149,7 @@ def main(path=None):
             m = rx_data.search(line)
             if not m:
                 continue
-            oid, klass, size = m.groups()
+            oid, size, klass = m.groups()
             size = int(size)
 
             obj_size.add(size)