Rev 106: Remove an unwanted pdb. in http://bzr.arbash-meinel.com/branches/bzr/history_db/tip_numbering
John Arbash Meinel
john at arbash-meinel.com
Fri Apr 16 21:40:29 BST 2010
At http://bzr.arbash-meinel.com/branches/bzr/history_db/tip_numbering
------------------------------------------------------------
revno: 106
revision-id: john at arbash-meinel.com-20100416204005-wqi8zd5joa5nmqnm
parent: john at arbash-meinel.com-20100416202747-smljmr1l2waogbdw
committer: John Arbash Meinel <john at arbash-meinel.com>
branch nick: tip_numbering
timestamp: Fri 2010-04-16 15:40:05 -0500
message:
Remove an unwanted pdb.
Big improvement on MySQL:
Stats:
{'_insert_node_calls': 68814,
'gdfo hit': 1181,
'gdfo miss': 615292,
'is interesting': 592684,
'nodes_expanded': 17059,
'not interesting imported': 1853,
'not interesting is mainline': 4877,
'not interesting known imported': 317801,
'pushed': 661498,
'ranges_inserted': 28,
'revs_in_ranges': 2778,
'step mainline': 1950421,
'step mainline added': 56149882,
'step mainline cache missed': 6260,
'step mainline cached': 1944161,
'step mainline unknown': 1950421,
'total_nodes_inserted': 661498}
real 4m18.383s
I would guess that at least one improvement is that we avoid walking
the graph 2 times. Once to build up what is interesting, and a second
time to actually number stuff. The code is certainly a lot simpler.
On average, we seem to step the mainline 114 times per branch, evaluating
28 nodes-per-step, or 3291 revs per branch. And each branch seems
to add 39 nodes. 3k nodes seems pretty high.
-------------- next part --------------
=== modified file 'history_db.py'
--- a/history_db.py 2010-04-16 20:17:11 +0000
+++ b/history_db.py 2010-04-16 20:40:05 +0000
@@ -199,7 +199,6 @@
merge_sorted = self._import_tip(self._branch_tip_rev_id)
if not expand_all:
return
- import pdb; pdb.set_trace()
# We know all the other imports are going to be incremental
self._incremental = True
self._stats['nodes_expanded'] += 0 # create an entry
More information about the bazaar-commits
mailing list