Differences From:

File src/schema.c part of check-in [ba486fec5a] - Add the unclustered table and the M lines to manifests. Process clusters in manifest_parse and manifest_crossref. The xfer process still does not yet use clusters though so it is still compatible. But that is about to change. by drh on 2007-09-03 01:28:57. [view]

To:

File src/schema.c part of check-in [48c4e69d2b] - Cluster-based synchronization appears to be working. by drh on 2007-09-09 17:51:16. Also file src/schema.c part of check-in [bbcb6326c9] - Pulled in the navbar and timeline changes. by aku on 2007-09-17 00:58:51. [view]

@@ -163,9 +163,10 @@
 @ );
 @ CREATE INDEX event_i1 ON event(mtime);
 @ CREATE INDEX event_i2 ON event(objid);
 @
-@ -- A record of phantoms
+@ -- A record of phantoms.  A phantom is a record for which we know the
+@ -- UUID but we do not (yet) know the file content.
 @ --
 @ CREATE TABLE phantom(
 @   rid INTEGER PRIMARY KEY         -- Record ID of the phantom
 @ );
@@ -172,8 +173,13 @@
 @
 @ -- Unclustered records.  An unclustered record is a record (including
 @ -- a cluster records themselves) that is not mentioned by some other
 @ -- cluster.
+@ --
+@ -- Phantoms are usually included in the unclustered table.  A new cluster
+@ -- will never be created that contains a phantom.  But another repository
+@ -- might send us a cluster that contains entries that are phantoms to
+@ -- us.
 @ --
 @ CREATE TABLE unclustered(
 @   rid INTEGER PRIMARY KEY         -- Record ID of the unclustered file
 @ );