# HG changeset patch
# User mpm@selenic.com
# Date 1115155633 28800
# Node ID 273ce12ad8f155317b2c078ec75a4eba507f1fba
# Parent  9117c6561b0bd7792fa13b50d28239d51b78e51f
Update README to discuss remote pull, rsync, and the hg repo
add a .hgignore file



diff -r 9117c6561b0b -r 273ce12ad8f1 .hgignore
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/.hgignore	Tue May 03 13:27:13 2005 -0800
@@ -0,0 +1,1 @@
+.*~
diff -r 9117c6561b0b -r 273ce12ad8f1 README
--- a/README	Tue May 03 13:16:10 2005 -0800
+++ b/README	Tue May 03 13:27:13 2005 -0800
@@ -69,6 +69,10 @@
 
 Network support (highly experimental):
 
+ # pull the self-hosting hg repo
+ foo$ hg init
+ foo$ hg merge http://selenic.com/hg/
+
  # export your .hg directory as a directory on your webserver
  foo$ ln -s .hg ~/public_html/hg-linux 
 
@@ -76,5 +80,10 @@
  bar$ hg merge http://foo/~user/hg-linux
 
  This is just a proof of concept of grabbing byte ranges, and is not
- expected to perform well.
+ expected to perform well. Fixing this needs some pipelining to reduce
+ the number of round trips. See zsync for a similar approach.
 
+ Another approach which does perform well right now is to use rsync.
+ Simply rsync the remote repo to a read-only local copy and then do a
+ local pull.
+