clone: only use stream when we understand the revlog format

clone: only use stream when we understand the revlog format

This patch fixes issues with stream cloning in the presense of parentdelta,
lwcopy and similar additions that change the interpretation of the revlog
format, or the format itself.

Currently, the stream capability is sent like this:
stream=<version of changelog>

But the client doesn't actually check the version number; also, it only checks
the changelog and it doesn't capture the interpretation-changes and
flag-changes in parentdelta and lwcopy.

This patch removes the 'stream' capability whenever we use a non-basic revlog
format, to prevent old clients from receiving incorrect data. In those cases,
a new capability called 'streamreqs' is added instead. Instead of a revlog
version, it comes with a list of revlog-format relevant requirements, which
are a subset of the repository requirements, excluding things that are not
relevant for stream.

New clients use this to determine whether or not they can stream. Old clients
only look for the 'stream' capability, as always. New servers will still send
this when serving old repositories.

--- a/mercurial/localrepo.py	Wed Sep 15 10:44:10 2010 +0200
+++ b/mercurial/localrepo.py	Wed Sep 15 11:06:22 2010 +0200
@@ -1784,7 +1784,7 @@
             return newheads - oldheads + 1
 
 
-    def stream_in(self, remote):
+    def stream_in(self, remote, requirements):
         fp = remote.stream_out()
         l = fp.readline()
         try:
@@ -1829,6 +1829,13 @@
         self.ui.status(_('transferred %s in %.1f seconds (%s/sec)\n') %
                        (util.bytecount(total_bytes), elapsed,
                         util.bytecount(total_bytes / elapsed)))
+
+        # new requirements = old non-format requirements + new format-related
+        # requirements from the streamed-in repository
+        requirements.update(set(self.requirements) - self.supportedformats)
+        self._applyrequirements(requirements)
+        self._writerequirements()
+
         self.invalidate()
         return len(self.heads()) + 1
 
@@ -1847,8 +1854,17 @@
         # and format flags on "stream" capability, and use
         # uncompressed only if compatible.
 
-        if stream and not heads and remote.capable('stream'):
-            return self.stream_in(remote)
+        if stream and not heads:
+            # 'stream' means remote revlog format is revlogv1 only
+            if remote.capable('stream'):
+                return self.stream_in(remote, set(('revlogv1',)))
+            # otherwise, 'streamreqs' contains the remote revlog format
+            streamreqs = remote.capable('streamreqs')
+            if streamreqs:
+                streamreqs = set(streamreqs.split(','))
+                # if we support it, stream in and adjust our requirements
+                if not streamreqs - self.supportedformats:
+                    return self.stream_in(remote, streamreqs)
         return self.pull(remote, heads)
 
     def pushkey(self, namespace, key, old, new):

--- a/mercurial/wireproto.py	Wed Sep 15 10:44:10 2010 +0200
+++ b/mercurial/wireproto.py	Wed Sep 15 11:06:22 2010 +0200
@@ -172,7 +172,13 @@
 def capabilities(repo, proto):
     caps = 'lookup changegroupsubset branchmap pushkey'.split()
     if _allowstream(repo.ui):
-        caps.append('stream=%d' % repo.changelog.version)
+        requiredformats = repo.requirements & repo.supportedformats
+        # if our local revlogs are just revlogv1, add 'stream' cap
+        if not requiredformats - set(('revlogv1',)):
+            caps.append('stream')
+        # otherwise, add 'streamreqs' detailing our local revlog format
+        else:
+            caps.append('streamreqs=%s' % ','.join(requiredformats))
     caps.append('unbundle=%s' % ','.join(changegroupmod.bundlepriority))
     return ' '.join(caps)