Rev 51: Add a progress indicator for chk pages. in http://bazaar.launchpad.net/%7Ebzr/bzr-groupcompress/trunk

[John Arbash Meinel]

At http://bazaar.launchpad.net/%7Ebzr/bzr-groupcompress/trunk

------------------------------------------------------------
revno: 51
revision-id: john at arbash-meinel.com-20090227035442-4sk2vcl343hg9afv
parent: john at arbash-meinel.com-20090227033445-gk1o5rr6t0u0d14i
committer: John Arbash Meinel <john at arbash-meinel.com>
branch nick: trunk
timestamp: Thu 2009-02-26 21:54:42 -0600
message:
  Add a progress indicator for chk pages.
  Fix a bug with handling signatures, which don't have a parent graph
-------------- next part --------------
=== modified file 'groupcompress.py'
--- a/groupcompress.py	2009-02-27 03:34:45 +0000
+++ b/groupcompress.py	2009-02-27 03:54:42 +0000
@@ -110,17 +110,18 @@
     """
     # gc-optimal ordering is approximately reverse topological,
     # properly grouped by file-id.
-    per_prefix_map = {'': []}
+    per_prefix_map = {}
     present_keys = []
     for item in parent_map.iteritems():
         key = item[0]
         if isinstance(key, str) or len(key) == 1:
-            per_prefix_map[''].append(item)
+            prefix = ''
         else:
-            try:
-                per_prefix_map[key[0]].append(item)
-            except KeyError:
-                per_prefix_map[key[0]] = [item]
+            prefix = key[0]
+        try:
+            per_prefix_map[prefix].append(item)
+        except KeyError:
+            per_prefix_map[prefix] = [item]
 
     for prefix in sorted(per_prefix_map):
         present_keys.extend(reversed(topo_sort(per_prefix_map[prefix])))
@@ -585,7 +586,8 @@
         keys = set(orig_keys)
         if not keys:
             return
-        if not self._index.has_graph and ordering == 'topological':
+        if (not self._index.has_graph
+            and ordering in ('topological', 'gc-optimal')):
             # Cannot topological order when no graph has been stored.
             ordering = 'unordered'
         # Cheap: iterate
@@ -604,7 +606,7 @@
             # Now group by source:
         elif ordering == 'gc-optimal':
             parent_map = dict((key, details[2]) for key, details in
-                locations.iteritems())
+                              locations.iteritems())
             for key in local_keys:
                 parent_map[key] = self._unadded_refs[key]
             # XXX: This only optimizes for the target ordering. We may need to

=== modified file 'repofmt.py'
--- a/repofmt.py	2009-02-27 03:04:49 +0000
+++ b/repofmt.py	2009-02-27 03:54:42 +0000
@@ -267,7 +267,7 @@
         stream = source_vf.get_record_stream(keys, 'gc-optimal', True)
         return _filter_inv_stream(stream), id_roots, p_id_roots
 
-    def _get_chk_stream(self, source_vf, keys, id_roots, p_id_roots):
+    def _get_chk_stream(self, source_vf, keys, id_roots, p_id_roots, pb=None):
         # We want to stream the keys from 'id_roots', and things they
         # reference, and then stream things from p_id_roots and things they
         # reference, and then any remaining keys that we didn't get to.
@@ -309,9 +309,18 @@
                 cur_keys = []
                 for prefix in sorted(keys_by_search_prefix):
                     cur_keys.extend(keys_by_search_prefix[prefix])
+        counter = 0
         for record in _get_referenced_stream(id_roots):
+            # We don't know how many total
+            counter += 1
+            if pb is not None:
+                pb.update('chk node', counter)
             yield record
         for record in _get_referenced_stream(p_id_roots):
+            # We don't know how many total
+            counter += 1
+            if pb is not None:
+                pb.update('chk node', counter)
             yield record
         if remaining_keys:
             trace.note('There were %d keys in the chk index, which'
@@ -320,6 +329,10 @@
             stream = source_vf.get_record_stream(remaining_keys, 'unordered',
                                                  True)
             for record in stream:
+                # We don't know how many total
+                counter += 1
+                if pb is not None:
+                    pb.update('chk node', counter)
                 yield record
 
     def _execute_pack_operations(self, pack_operations, _packer_class=Packer,
@@ -395,16 +408,20 @@
                         access=target_access,
                         delta=source_vf._delta)
                     stream = None
-                    if has_chk:
-                        if vf_name == 'inventories':
-                            stream, id_roots, p_id_roots = self._get_filtered_inv_stream(
-                                source_vf, keys)
-                        elif vf_name == 'chk_bytes':
-                            stream = self._get_chk_stream(source_vf, keys,
-                                                          id_roots, p_id_roots)
-                    if stream is None:
-                        stream = source_vf.get_record_stream(keys, 'gc-optimal', True)
-                    target_vf.insert_record_stream(stream)
+                    child_pb = ui.ui_factory.nested_progress_bar()
+                    try:
+                        if has_chk:
+                            if vf_name == 'inventories':
+                                stream, id_roots, p_id_roots = self._get_filtered_inv_stream(
+                                    source_vf, keys)
+                            elif vf_name == 'chk_bytes':
+                                stream = self._get_chk_stream(source_vf, keys,
+                                    id_roots, p_id_roots, pb=child_pb)
+                        if stream is None:
+                            stream = source_vf.get_record_stream(keys, 'gc-optimal', True)
+                        target_vf.insert_record_stream(stream)
+                    finally:
+                        child_pb.finished()
                 new_pack._check_references() # shouldn't be needed
             except:
                 pb.finished()