revlog: return lazy set from findcommonmissing
When computing the commonmissing, it greedily computes the entire set
immediately. On a large repo where the majority of history is irrelevant, this
causes a significant slow down.
Replacing it with a lazy set makes amend go from 11 seconds to 8.7 seconds.
#!/bin/sh
#
# Use this script to generate startrev.svndump
#
mkdir temp
cd temp
mkdir project-orig
cd project-orig
mkdir trunk
mkdir branches
mkdir tags
cd ..
svnadmin create svn-repo
svnurl=file://`pwd`/svn-repo
svn import project-orig $svnurl -m "init projA"
svn co $svnurl project
cd project
echo a > trunk/a
echo b > trunk/b
svn add trunk/a trunk/b
svn ci -m createab
svn rm trunk/b
svn ci -m removeb
svn up
echo a >> trunk/a
svn ci -m changeaa
# Branch
svn up
svn copy trunk branches/branch1
echo a >> branches/branch1/a
svn ci -m "branch, changeaaa"
echo a >> branches/branch1/a
echo c > branches/branch1/c
svn add branches/branch1/c
svn ci -m "addc,changeaaaa"
svn up
cd ..
svnadmin dump svn-repo > ../startrev.svndump