<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <meta content="text/html;charset=UTF-8" http-equiv="Content-Type"> </head> <body bgcolor="#ffffff" text="#000000"> John Szakmeister wrote: <blockquote cite="mid:a1c1f4751001120748j6c10df3bp89a1b3ebd40b1853@mail.gmail.com" type="cite"> <pre wrap="">On Tue, Jan 12, 2010 at 10:27 AM, John Arbash Meinel <a class="moz-txt-link-rfc2396E" href="mailto:john@arbash-meinel.com"><john@arbash-meinel.com></a> wrote: [snip] </pre> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="">In my case, the graph looks more like: A |\ | B | | | C | | | D |/ E So the parents for E is both A and D. But in the inventory (I assume), it knows that C is a (text) parent of D. It's complaining about the fact that C is missing. Does that make sense? -John </pre> </blockquote> <pre wrap="">Right, so after going to svn and then back to bzr, at the *revision* level you end up with: A | | ? # D something |/ E At the inventory level, some things probably reference any of B, C, D. </pre> </blockquote> <pre wrap=""> I understand that. I was trying to give you more information so that we could understand the inconsistent parent bit. Here's the output of 'bzr check': checked repository <bzrlib.transport.local.LocalTransport url=<a class="moz-txt-link-freetext" href="file:///Users/jszakmeister/tmp/ghost/trunk2/">file:///Users/jszakmeister/tmp/ghost/trunk2/</a>> format <RepositoryFormat2a> 3 revisions 2 file-ids 0 unreferenced text versions 1 ghost revisions <a class="moz-txt-link-abbreviated" href="mailto:john@szakmeister.net-20100109110920-2o7etdw3o064uaee">john@szakmeister.net-20100109110920-2o7etdw3o064uaee</a> 1 inconsistent parents * <a class="moz-txt-link-abbreviated" href="mailto:2@b41ffb15-e730-42e5-b8a5-20245dfbf914:trunk%2Ffile.txt">2@b41ffb15-e730-42e5-b8a5-20245dfbf914:trunk%2Ffile.txt</a> version <a class="moz-txt-link-abbreviated" href="mailto:john@szakmeister.net-20100109110920-2o7etdw3o064uaee">john@szakmeister.net-20100109110920-2o7etdw3o064uaee</a> has parents ('<a class="moz-txt-link-abbreviated" href="mailto:john@szakmeister.net-20100109110920-rsdetf6fzzfyopbd">john@szakmeister.net-20100109110920-rsdetf6fzzfyopbd</a>',) but should have () checked branch <a class="moz-txt-link-freetext" href="file:///Users/jszakmeister/tmp/ghost/trunk2/">file:///Users/jszakmeister/tmp/ghost/trunk2/</a> format Branch format 7 The parent it's referring to is a text-parent for file.txt. So my question is: should 'bzr check' be telling me this? And if so, then what does bzr-svn need to do to make this problem go away? Given what was said earlier in the thread, should 'bzr check' be modified to say that this is okay? Right now it exits with an error code. The place that I'm trying to reach here is that I'd very much like to use bzr for most of our development. But, we need to operate against svn repositories in many cases. So, I'd like to reach a happy place where everything works as it should, all checks pass, etc. -John </pre> </blockquote> <tt> </tt><tt>You should see mine: </tt><tt> Checking working tree at '/home/dmuir/www/book_dev'. Checking branch at '<a class="moz-txt-link-freetext" href="file:///home/dmuir/www/book_dev/">file:///home/dmuir/www/book_dev/</a>'. Checking repository at '<a class="moz-txt-link-freetext" href="file:///home/dmuir/www/book_dev/">file:///home/dmuir/www/book_dev/</a>'. checked repository <bzrlib.transport.local.LocalTransport url=<a class="moz-txt-link-freetext" href="file:///home/dmuir/www/book_dev/">file:///home/dmuir/www/book_dev/</a>> format <RepositoryFormat2a> 627 revisions 2885 file-ids 165 ghost revisions 1482 inconsistent parents checked branch <a class="moz-txt-link-freetext" href="file:///home/dmuir/www/book_dev/">file:///home/dmuir/www/book_dev/</a> format Branch format 7 David</tt> </body> </html>