[Bug 3782] New: test-convert-cvs-synthetic.t sometimes fails due to varying order of changesets

[mercurial-bugs at selenic.com]

http://bz.selenic.com/show_bug.cgi?id=3782

          Priority: normal
            Bug ID: 3782
                CC: mercurial-devel at selenic.com
          Assignee: bugzilla at selenic.com
           Summary: test-convert-cvs-synthetic.t sometimes fails due to
                    varying order of changesets
          Severity: bug
    Classification: Unclassified
                OS: Linux
          Reporter: thomas at intevation.de
          Hardware: PC
            Status: UNCONFIRMED
           Version: 2.5-rc
         Component: convert
           Product: Mercurial

While running tests I noticed this today on two different machines (Debian
squeeze with python2.6 and Debian sarge with python2.4):

--- /home/thomas/hg/repos/tah/tests/test-convert-cvs-synthetic.t 
+++ /home/thomas/hg/repos/tah/tests/test-convert-cvs-synthetic.t.err 
@@ -144,8 +144,8 @@
   sorting...
   converting...
   8 add file1 on trunk
-  7 add file2
-  6 MERGE from v1_0: add file2
+  7 MERGE from v1_0: add file2
+  6 add file2
   5 file file3 was initially added on branch v1_1.
   4 add file3, file4 on branch v1_1
   3 add file5 on v1_2
@@ -168,9 +168,9 @@
   | | |
   o | |  3 file file3 was initially added on branch v1_1.
   |/ /
-  | o  2 MERGE from v1_0: add file2
-  |/
-  | o  1 add file2
+  +---o  2 add file2
+  | |
+  | o  1 MERGE from v1_0: add file2
   |/
   o  0 add file1 on trunk

@@ -191,8 +191,8 @@
   sorting...
   converting...
   8 add file1 on trunk
-  7 add file2
-  6 MERGE from v1_0: add file2
+  7 MERGE from v1_0: add file2
+  6 add file2
   5 file file3 was initially added on branch v1_1.
   4 add file3, file4 on branch v1_1
   3 add file5 on v1_2
@@ -215,9 +215,9 @@
   | | |
   o | |  3 file file3 was initially added on branch v1_1.
   |/ /
-  | o  2 MERGE from v1_0: add file2
-  |/
-  | o  1 add file2
+  +---o  2 add file2
+  | |
+  | o  1 MERGE from v1_0: add file2
   |/
   o  0 add file1 on trunk

It does not always happen, so I bisected it with running the test up to three
times:

hg bisect -r && hg bisect -g 2.4 && hg bisect -b 2.5-rc && hg bisect -c 'nice
./run-tests.py test-convert-cvs-synthetic.t && nice ./run-tests.py
test-convert-cvs-synthetic.t && nice ./run-tests.py
test-convert-cvs-synthetic.t'

The first bad revision is:
changeset:   18261:1b7b5975793f
user:        Frank Kingswood <frank at kingswood-consulting.co.uk>
date:        Tue Jan 08 20:11:20 2013 +0000
summary:     cvsps: use commitids (when present) to detect changesets


This is just a sorting problem, so not urgent in itself, but as it can cause
failed tests, it should be corrected before the 2.5 release.

-- 
You are receiving this mail because:
You are on the CC list for the bug.