couch-incarnate
v0.2.0
Published
CouchDB views in DB-form; for chaining map-reduce operations
Downloads
1
Readme
Couch Incarnate
A CouchDB tool for maintaining views of views (AKA chaining map-reduce operations).
Requirements
- Node.js (tried on 0.6.2)
- CouchDB (tried on 1.1.1)
- NPM - Node's package manager (https://github.com/isaacs/npm)
Installation
run:
npm install couch-incarnate
edit 'conf.json' file. Remove the "log" entry to write to stdout.
done! Now simply run:
incarnate
Incarnation - the basic concept
An incarnation is a CouchDB that is equivalent to a CouchDB view. Like a view, this DB is based on map and reduce functions, and each document in it is of the form '{key: ..., value: ...}'.
Once an incarnation is set up, you could add to it a design-doc with a view (or several), ending up with what is effectively a view of a view.
To be more precise, an incarnation is based on a source DB, a map-function, a reduce function, and a group-level.
Contents are equivalent to the following query, if instead of an incarnation, a normal view were used, with the same map and reduce functions:
GET /SOURCE_DB/_design/views/MAP_REDUCE_VIEW/?reduce=true&group_level=GROUP_LEVEL
Incarnation Updates
When trying to access an incarnation, it is first updated from the source DB, and only then does the actual access takes place.
This is much like the way views are queried without the 'stale' option.
Incarnator
An incarnator defines incarnations. All incarnations defined by it use the same source DB and map function, as follows:
{
"source": "SOURCE_DB_URL",
"map": "function (doc) { ... }",
"reduces": {
"REDUCE_1_NAME": {
"function": "function (key, values, rereduce) { ... }",
"group_levels": [ GROUP_LEVEL1, GROUP_LEVEL2, ... ]
},
"INC_2_NAME": {
...
},
...
}
}
With the above configuration, at least two incarnations would be created. Both would have the same map and reduce functions, but each would have a different group-level.
Example
Let's say your application holds documents in a DB called 'my_db', which look like this:
{
type: "user",
id: "blaster77",
join_date: "2011/05/29", // YYYY/MM/DD
}
Now say you want to have an incarnation holding the number of users that join each year. The incarnator configuration would look like this:
{
"source": "http://localhost:5984/my_db",
"map": "function (doc) { if (doc.type === "user") emit(doc.join_date.split('/')[0], null); }",
"reduces": {
"count": {
"function": "function (key, values, rereduce) { if (!rereduce) return {count: values.length}; var count = 0; for (var i = 0; i < values.length; i++) { count += values[i]; }; return count; }",
"group_levels": ['exact']
}
}
}
If you wanted also to have incarnations for the number of users that join each month, and each day, the incarnator configuration could look like this:
{
"source": "http://localhost:5984/my_db",
"map": "function (doc) { if (doc.type === "user") emit(doc.join_date.split('/'), null); }",
"reduces": {
"count": {
"function": "function (key, values, rereduce) { if (!rereduce) return {count: values.length}; var count = 0; for (var i = 0; i < values.length; i++) { count += values[i]; }; return count; }",
"group_levels": [1, 2, 'exact']
}
}
}
Incarnate server
An HTTP server that administrates and maintains incarnations
Configuration
Configuration file - ./conf :
{
"port": 4895,
"couch": "http://localhost:5984",
"log": {
"path": "/var/log/incarnate/incarnate.log"
}
}
API
Set incarnator
PUT /INCARNATOR_NAME
{
"source": "SOURCE_DB_URL",
"map": "function (doc) { ... }",
"reduces": {
"REDUCE_1_NAME": {
"reduce": "function (key, values, rereduce) { ... }",
"group_levels": [ GROUP_LEVEL1, GROUP_LEVEL2, ... ]
},
"REDUCE_2_NAME": {
...
},
...
}
}
If INCARNATOR_NAME exists, overwrite it.
Get incarnator status
GET /INCARNATOR_NAME
Delete incarnator
DELETE /INCARNATOR_NAME
Rename incarnator
MOVE /INCARNATOR_NAME
Destination: NEW_INCARNATOR_NAME
Access incarnation
METHOD_NAME /INCARNATOR_NAME/REDUCE_NAME/GROUP_LEVEL/[...]
Persistence
Persistence is aimed to be on par with CouchDB's.
Note: currently, unless running on Mac OS X, both CouchDB and Incarnate would provide absolute disk-persistence only if the disk has no cache on-board, or if the disk has a back-up power supply, in case of a power failure. See here for more: http://lwn.net/Articles/270891/
TODO
- tests. Does this thing really work?
- input validation
- _changes
Caveats
- not properly tested
- limited input validation
- limited error-handling
- only JS map and reduce functions are supported
- no SSL
- no continuous changes-feed for incarnations
Copyright
Copyright (c) 2011 Alon Keren [email protected]
You can redistribute this software and/or modify it under the terms of the GNU Affero General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more details.
A copy of the GNU Affero General Public License is in the file named 'LICENSE.txt'. It should be also available at: http://www.gnu.org/licenses/