I am trying use svndumpfilter. I exported the repository and I am now trying to run: svndumpfilter exclude foo < a.dmp > b.dmp
The problem is that the generated dump file, 'b.dmp', is inconsistent with the documented format (which is causing some other problems for me): https://svn.apache.org/repos/asf/subversion/trunk/notes/dump-load-format.txt The above documentation states that 'Node-path' will be the first header line of a Node record. However, this is an example of what I'm getting: Revision-number: 1 Prop-content-length: 186 Content-length: 186 K 8 svn:date V 27 2006-10-11T22:12:09.613653Z K 7 svn:log V 73 Share project 'CampaignDirectorPOC' into 'http://test-build:7080/svn/dev' K 10 svn:author V 17 tim.m...@par3.com PROPS-END Node-kind: dir Node-path: trunk Node-action: add Prop-content-length: 10 Content-length: 10 PROPS-END Note that in the above example it's 2nd. In other revisions from my filtered dump it is 3rd and otherwise all over the map. The order is correct in the original dump file. I've tried using svndumpfilter 1.8.5 on Linux and 1.8.8 on Windows. Has anyone else come across this? I couldn't find any documented known issues/bugs on the matter. Any suggestions? Thanks, Jonathan
