Rev 11: Add progress bars to make the wait slightly more bearable. in http://bazaar.launchpad.net/%7Ebzr/bzr-repodetails/trunk

[John Arbash Meinel]

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

------------------------------------------------------------
revno: 11
revision-id: john at arbash-meinel.com-20081202020220-xx65k70hbx2ai9ne
parent: john at arbash-meinel.com-20081202014502-40sghjisv1lflwhb
committer: John Arbash Meinel <john at arbash-meinel.com>
branch nick: trunk
timestamp: Mon 2008-12-01 20:02:20 -0600
message:
  Add progress bars to make the wait slightly more bearable.
-------------- next part --------------
=== modified file 'gather_stats.py'
--- a/gather_stats.py	2008-12-02 01:45:02 +0000
+++ b/gather_stats.py	2008-12-02 02:02:20 +0000
@@ -22,7 +22,7 @@
 
 import math
 
-from bzrlib import chk_map, repository
+from bzrlib import chk_map, repository, ui
 from bzrlib.inventory import CHKInventory
 from bzrlib.repofmt.pack_repo import (
     RepositoryFormatKnitPack1,
@@ -168,12 +168,14 @@
         repo.unlock()
 
 
-def _gather_object_vf_texts(objectstats, vf):
-    for text in _gather_and_iter_object_vf_texts(objectstats, vf):
+def _gather_object_vf_texts(objectstats, vf, kind=None):
+    for text in _gather_and_iter_object_vf_texts(objectstats, vf, kind=kind):
         pass
 
 
-def _gather_and_iter_object_vf_texts(objectstats, vf, keys=None):
+def _gather_and_iter_object_vf_texts(objectstats, vf, keys=None, kind=None):
+    if kind is None:
+        kind = '<unknown>'
     if keys is None:
         keys = vf.keys()
     objectstats.objects += len(keys)
@@ -183,12 +185,17 @@
         objectstats.add_compressed_size(detail[0][2])
     keys = sorted(keys)
     batch_size = 200
-    for offset in xrange(0, len(keys), batch_size):
-        batch = keys[offset:offset + batch_size]
-        for entry in vf.get_record_stream(batch, 'unordered', True):
-            bytes = entry.get_bytes_as('fulltext')
-            objectstats.add_raw_size(len(bytes))
-            yield bytes, entry.key
+    pb = ui.ui_factory.nested_progress_bar()
+    try:
+        for offset in xrange(0, len(keys), batch_size):
+            pb.update(kind, offset, len(keys))
+            batch = keys[offset:offset + batch_size]
+            for entry in vf.get_record_stream(batch, 'unordered', True):
+                bytes = entry.get_bytes_as('fulltext')
+                objectstats.add_raw_size(len(bytes))
+                yield bytes, entry.key
+    finally:
+        pb.finished()
 
 
 def _gather_chk_map(objectstats, chk_bytes, pending, internal_counter,
@@ -201,7 +208,8 @@
         done.update(pending)
         next = pending
         pending = set()
-        for bytes, key in _gather_and_iter_object_vf_texts(objectstats, chk_bytes, next):
+        for bytes, key in _gather_and_iter_object_vf_texts(objectstats,
+                            chk_bytes, next, kind='chk'):
             node = chk_map._deserialise(bytes, key)
             refs = node.refs()
             pending.update(refs)
@@ -220,7 +228,8 @@
     # first pass: the inventory objects yield chk dicts:
     pending = set()
     pending_parent_id = set()
-    for bytes, key in _gather_and_iter_object_vf_texts(objectstats, repo.inventories):
+    for bytes, key in _gather_and_iter_object_vf_texts(objectstats,
+                        repo.inventories, kind='inv'):
         inv = CHKInventory.deserialise(repo.chk_bytes, bytes, key)
         pending.add(inv.id_to_entry._root_node)
         if inv.parent_id_basename_to_file_id is not None:
@@ -240,21 +249,24 @@
     keys = repo.revisions.keys()
     result.revision_count = len(keys)
     if isinstance(repo._format, pack_rev_types):
-        _gather_object_vf_texts(result.revisions, repo.revisions)
+        _gather_object_vf_texts(result.revisions, repo.revisions, kind='revs')
     else:
         raise AssertionError("Don't know how to process %r" % repo)
     if isinstance(repo._format, pack_xml_inv_types):
-        _gather_object_vf_texts(result.inventories, repo.inventories)
+        _gather_object_vf_texts(result.inventories, repo.inventories,
+                                kind='inv')
     elif isinstance(repo._format, pack_chk_inv_types):
         _gather_chk_inv(result.inventories, repo)
     else:
         raise AssertionError("Don't know how to process %r" % repo)
     if isinstance(repo._format, pack_text_types):
-        _gather_object_vf_texts(result.texts, repo.texts)
+        _gather_object_vf_texts(result.texts, repo.texts,
+                                kind='texts')
     else:
         raise AssertionError("Don't know how to process %r" % repo)
     if isinstance(repo._format, pack_signature_types):
-        _gather_object_vf_texts(result.signatures, repo.signatures)
+        _gather_object_vf_texts(result.signatures, repo.signatures,
+                                kind='sigs')
     else:
         raise AssertionError("Don't know how to process %r" % repo)
     result.total = ObjectStats()