From 0827cb483ef4cf74123313e8f2d0d2bd5ce8e46b Mon Sep 17 00:00:00 2001 From: Robert Newson Date: Mon, 20 Jun 2011 21:22:02 +0000 Subject: Fix spurious declarations of new merge conflicts This patch also adds extra tests of the key tree merging logic as well as edoc-formatted documentation for the module and a few of the merge functions. Closes COUCHDB-902. Thanks Paul Davis, Bob Dionne, Klaus Trainer. backported from trunk@1065471 Conflicts: src/couchdb/couch_key_tree.erl git-svn-id: https://svn.apache.org/repos/asf/couchdb/branches/1.1.x@1137789 13f79535-47bb-0310-9956-ffa450edef68 --- src/couchdb/couch_key_tree.erl | 48 ++++++++++++++++++++++++++++++++++++++---- 1 file changed, 44 insertions(+), 4 deletions(-) (limited to 'src/couchdb') diff --git a/src/couchdb/couch_key_tree.erl b/src/couchdb/couch_key_tree.erl index e5d3549f..48a76b1d 100644 --- a/src/couchdb/couch_key_tree.erl +++ b/src/couchdb/couch_key_tree.erl @@ -10,6 +10,41 @@ % License for the specific language governing permissions and limitations under % the License. +%% @doc Data structure used to represent document edit histories. + +%% A key tree is used to represent the edit history of a document. Each node of +%% the tree represents a particular version. Relations between nodes represent +%% the order that these edits were applied. For instance, a set of three edits +%% would produce a tree of versions A->B->C indicating that edit C was based on +%% version B which was in turn based on A. In a world without replication (and +%% no ability to disable MVCC checks), all histories would be forced to be +%% linear lists of edits due to constraints imposed by MVCC (ie, new edits must +%% be based on the current version). However, we have replication, so we must +%% deal with not so easy cases, which lead to trees. +%% +%% Consider a document in state A. This doc is replicated to a second node. We +%% then edit the document on each node leaving it in two different states, B +%% and C. We now have two key trees, A->B and A->C. When we go to replicate a +%% second time, the key tree must combine these two trees which gives us +%% A->(B|C). This is how conflicts are introduced. In terms of the key tree, we +%% say that we have two leaves (B and C) that are not deleted. The presense of +%% the multiple leaves indicate conflict. To remove a conflict, one of the +%% edits (B or C) can be deleted, which results in, A->(B|C->D) where D is an +%% edit that is specially marked with the a deleted=true flag. +%% +%% What makes this a bit more complicated is that there is a limit to the +%% number of revisions kept, specified in couch_db.hrl (default is 1000). When +%% this limit is exceeded only the last 1000 are kept. This comes in to play +%% when branches are merged. The comparison has to begin at the same place in +%% the branches. A revision id is of the form N-XXXXXXX where N is the current +%% revision. So each path will have a start number, calculated in +%% couch_doc:to_path using the formula N - length(RevIds) + 1 So, .eg. if a doc +%% was edit 1003 times this start number would be 4, indicating that 3 +%% revisions were truncated. +%% +%% This comes into play in @see merge_at/3 which recursively walks down one +%% tree or the other until they begin at the same revision. + -module(couch_key_tree). -export([merge/3, find_missing/2, get_key_leafs/2, get_full_key_paths/2, get/2]). @@ -31,6 +66,9 @@ merge(Paths, Path, Depth) -> {Merged, Conflicts} = merge(Paths, Path), {stem(Merged, Depth), Conflicts}. +%% @doc Merge a path with an existing list of paths, returning a new list of +%% paths. A return of conflicts indicates a new conflict was discovered in this +%% merge. Conflicts may already exist in the original list of paths. -spec merge([path()], path()) -> {[path()], conflicts | no_conflicts}. merge(Paths, Path) -> {ok, Merged, HasConflicts} = merge_one(Paths, Path, [], false), @@ -69,6 +107,7 @@ merge_at([{Key, Value, SubTree}|Sibs], Place, InsertTree) when Place > 0 -> {ok, Merged, Conflicts} -> {ok, [{Key, Value, Merged} | Sibs], Conflicts}; no -> + % first branch didn't merge, move to next branch case merge_at(Sibs, Place, InsertTree) of {ok, Merged, Conflicts} -> {ok, [{Key, Value, SubTree} | Merged], Conflicts}; @@ -111,11 +150,12 @@ merge_simple([{Key, V1, SubA} | NextA], [{Key, V2, SubB} | NextB]) -> Value = value_pref(V1, V2), {[{Key, Value, MergedSubTree} | MergedNextTree], Conflict1 or Conflict2}; merge_simple([{A, _, _} = Tree | Next], [{B, _, _} | _] = Insert) when A < B -> - {Merged, _} = merge_simple(Next, Insert), - {[Tree | Merged], true}; + {Merged, Conflict} = merge_simple(Next, Insert), + % if Merged has more branches than the input we added a new conflict + {[Tree | Merged], Conflict orelse (length(Merged) > length(Next))}; merge_simple(Ours, [Tree | Next]) -> - {Merged, _} = merge_simple(Ours, Next), - {[Tree | Merged], true}. + {Merged, Conflict} = merge_simple(Ours, Next), + {[Tree | Merged], Conflict orelse (length(Merged) > length(Next))}. find_missing(_Tree, []) -> []; -- cgit v1.2.3