replay: fetch full revision at most once per run (issue252) Before this change, hgsubversion was fetching full revisions from the first revision the project was created to the first revision containing converted data. Unfortunately, some projects exhibits such spans longer than 500 revisions, during which hgsubversion was uselessly scanning the whole tree. The fix is not technically perfect, we could record somewhere that while no data was converted we scanned the project already, instead of scanning once at every hgsubversion run until a revision is converted. But it should be good enough unless someone runs hgsubversion once for every target revision. One repository exhibiting this behaviour: svn://svn.zankasoftware.com/zanka

#!/bin/sh
mkdir temp || exit 1
cd temp
svnadmin create repo
svn co file://`pwd`/repo wc
pushd wc
mkdir -p project/trunk
svn add project
svn ci -m 'trunk'
cd project/trunk
echo a > a
mkdir narf
svn add a narf
svn ci -m 'file and empty dir'
popd
svnadmin dump repo > ../empty_dir_in_trunk_not_repo_root.svndump
echo 'dump in empty_dir_in_trunk_not_repo_root.svndump'
echo 'you can probably delete temp now'