Hi!
I got some pretty strange results today when I tried to blame a file with "Include merge info" enabled. TortoiseBlame showed a lot of things that didn't make any sense. For example a large part of the file as modified by a developer that had only changed one line in that part of the file. (When viewing without merge info enabled, everything looked like I excpected, but then the information is of little value as I only see the merge-commits...) I then tried to use a text viewer to view the blames, and didn't get much wiser.
Here's an excerpt from the blame output, with path, author and code omitted:
--- begin ---
line rev rev date
...
G 1088 927 1048 2009-03-30 16:06:52
G 1089 927 1048 2009-03-30 16:06:52
G 1090 1 1 2009-01-21 10:37:28
G 1091 1031 1048 2009-04-01 13:18:44
G 1092 1031 1048 2009-04-01 13:18:44
G 1093 1031 1048 2009-04-01 13:18:44
G 1094 1031 1048 2009-04-01 13:18:44
G 1095 1031 1048 2009-04-01 13:18:44
G 1096 1 1 2009-01-21 10:37:28
G 1097 927 1048 2009-03-30 16:06:52
G 1098 1031 1 2009-04-01 13:18:44
G 1099 1031 1 2009-04-01 13:18:44
G 1100 1031 1 2009-04-01 13:18:44
G 1101 1031 1 2009-04-01 13:18:44
G 1102 1031 1 2009-04-01 13:18:44
G 1103 1031 1 2009-04-01 13:18:44
G 1104 1031 1 2009-04-01 13:18:44
G 1105 1 1048 2009-01-21 10:37:28
G 1106 1 1048 2009-01-21 10:37:28
G 1107 927 1 2009-03-30 16:06:52
G 1108 927 1 2009-03-30 16:06:52
G 1109 1 1048 2009-01-21 10:37:28
G 1110 927 1 2009-03-30 16:06:52
G 1111 927 1 2009-03-30 16:06:52
G 1112 927 1 2009-03-30 16:06:52
G 1113 927 1 2009-03-30 16:06:52
G 1114 1 1048 2009-01-21 10:37:28
G 1115 927 1 2009-03-30 16:06:52
G 1116 927 1 2009-03-30 16:06:52
G 1117 927 1 2009-03-30 16:06:52
--- end ---
Tortoise blame seems to use the left "rev" column as source for the blame info in this case.
As I'm interpreting the result, the left rev column shows the latest edit to a certain line, and the right rev column shown the latest merge affecting that line. Looking at line 1088 this makes sense. In revision 927 this line was edited on a branch and in revision 1048 it was merged back to trunk (where I'm doing the blame). But looking at line 1099 things look kinda strange strange. This line was edited in revision 1031 (a whitespace only change), and it was reintegrated in 1048, but the second column does not indicate any change. Line 1117 wasn't edited at all in revision 927!
Am I missinterpreting something here, or am I looking at an error in the blame function on the SVN side?
Hans-Emil
------------------------------------------------------
http://tortoisesvn.tigris.org/ds/viewMessage.do?dsForumId=757&dsMessageId=1574544
To unsubscribe from this discussion, e-mail: [dev-unsubscribe_at_tortoisesvn.tigris.org].
Received on 2009-04-07 12:09:13 CEST