svnwrap: use custom StringIO class in get_file() The wrappers were calling ra.get_file() with a cStringIO object. Empirically, svn 1.7.5 is writing 16kB blocks to the stream object, and cStringIO reallocates its internal buffer and doubles its size whenever it is filled. With large committed files this requires two large memory blocks at the same time. SimpleStringIO implements the mimimum StringIO interface used by ra.get_file() but instead stores all the blocks and "join" them at the end. It means more fragmentation but requires only one large block, without overallocation. Also, 16kB blocks should be friendly to most allocators. In practice, this simple change let me convert a revision containing multiple moderately large files, the largest being around 450MB, with a 32-bits Windows setup, python 2.7, swig svn 1.7.5, in stupid mode, while it was previously aborting with "not enough memory". The same revision still fails in replay mode.

#!/bin/sh
mkdir temp
cd temp
svnadmin create repo
svn co file://`pwd`/repo wc
cd wc
mkdir branches trunk tags
mkdir tags/versions
mkdir tags/blah
svn add *
svn ci -m 'btt'
cd trunk
for a in alpha beta gamma delta ; do
    echo $a > $a
    svn add $a
done
svn ci -m 'Add files.'

echo 'foo bar' > 'foo bar'
svn add 'foo bar'
svn ci -m 'Add files.'

mkdir 'blah blah'
echo 'another file' > 'blah blah/another file'
svn add 'blah blah'
svn ci -m 'Add files.'

cd ..
svn up
svn cp trunk branches/dev_branch
svn ci -m 'Make a branch'
cd ../..

svnadmin dump temp/repo > spaces-in-path.svndump
echo
echo 'Complete.'
echo 'You probably want to clean up temp now.'
echo 'Dump in spaces-in-path.svndump'
exit 0