Skip to content
This repository has been archived by the owner on Jan 21, 2022. It is now read-only.

Latest commit

 

History

History
94 lines (89 loc) · 6.82 KB

2.0b3.markdown

File metadata and controls

94 lines (89 loc) · 6.82 KB
Raw

2.0b3 / 2010-12-08

  • Added heavy test coverage
  • Tests/demos of all additivity of DBObject and Tuple
  • ++ additivity operator on MongoDBObject for lists of tuple pairs
  • Added test for $nor demonstrating additivity between a DSLed DBObject and a tuple
  • Added val aliasing for MongoDBObject -> DBObject and MongoDBList -> DBList (only for the 'object' singletons)
  • Added bareword DSL operator for $nor
  • Added new $not syntax, along with identical support for nested queries in $pull:
  • Added SLF4J-JCL Bindings as a +test+ dependency (so we can print logging while testing without forcing you to use an slf4j implementation yourself)
  • $elemMatch shouldn't have been a nestable, and only accept DBObject. Fixed.
  • Full test coverage on DSL ...nested query chains are heavily broken at the moment.
  • Migrated $each to a now functioning internal hook on $addToSet only exposed in certain circumstances
  • Various cleanups to docs and type constraints in BarewordOperators
  • Changed Geo related stuff to use ValidNumericType instead of Numeric as some Numerics are not valid, and we'll be using Valid+Type for auto conversion reg soon
  • GridFS' `files' should have returned a MongoCursor not a raw Java DBCursor. Closes #25
  • GridFS findOne wasn't returning Option[_] and detecting nulls like Collection, which has been corrected. Closes #26
  • Added a false $eq operator to emulate key = value easily in the DSL
  • Restored ability to chain DSL operators (First pass - may change)
  • Fixed RegEx and Binary $type tests
  • Differentiated GeoCoords to use two different Manifested Numerics for differing datatypes
  • Timestamp must take precedence over date to match properly
  • Changed precedence so DBList is first as it's also a dbobject and test failed
  • Added Joda DateTime to Type testing in $type
  • Updates and test coverage for $size and $exists
  • Added new implementations of $in, $nin, $all and $mod with tests. $mod now accepts non-Int numerics and of two differing types.
  • Stripped out support for implicit Product/Tuple conversions as they're buggy as hell and constantly interfere with other code.
  • Implemented new code and tests for $gt, $gte, $lt, $lte.
  • Added Date tests to $Ne and added testing for $lt
  • Valid Date and Numeric Type boundaries introduced and used instead of Numeric (since Char doesn't actually workwith Mongo and you can't double up type bounds)
  • Refactor of $ne to a new baseline, new unit test for DSLCoreOperators to test it
  • Checkpoint, ValidProduct won't work either
  • mid-refactor of core operators code, checkpoint for if we keep aritied tuple code
  • Removed crufty extra logging in distinctKeySet
  • Ported MapReduce test to Specs in new directory structure; needs more test methods
  • Removing GridFSSpec as it has been ported to Specs in the new modules
  • Changed sub$project names back to start with casbah- to reduce confusion with maven artifacts
  • Updated documentation buildout and migrated default build version to 2.8.1, removing 2.8.1rc4

2.0b2 / 2010-11-09

  • Added documentation build script
  • Moved loading of ConversionHelpers from Connection creation to instantiation of Commons' Implicits
    • This means conversions are ALWAYS loaded now for everyone
  • Added registered conversion for Product (tuples) which fixes open issues with serialization
  • Migrated Conversions code from core to commons, repackaging as com.mongodb.casbah.commons.conversions
    • Unit Test remains in Core as it uses core code to test loading/unloading which is fine for buildtime tests
  • Moved Logger from core to commons, package changed from com.mongodb.casbah.util.Logging to com.mongodb.casbah.commons.Logging

2.0b1 / 2010-11-05

  • Switched off of configgy to slf4j as akka did; fixed project build and released as 2.0b1
  • Collection is now inline with the 2.3 java driver
  • updated MongoDB for API compatibility with Java driver 2.3 and added some necessary docs.
  • added a WriteConcern helper object for Scala users w/ named & default args
  • Updated connection to match 2.3, added support for specifying writeconcern and slaveOK
  • Fixes #18 corrected pop implementation
  • Fixes #13 added findAndModify / findAndRemove
  • Fixes #19 Added full support for geospatial query. Needs some docs and tests
  • Updated to Java 2.3 driver #12
  • Fixes #2 ElemMatch works much like $not
  • Fixes #11 Context Bound and Byte Enum supported $type queries
  • Fixes #17 added $rename and fixed a bug caused by $or implementation
  • Fixes #1, added $each target and improved $addtoSet impl $each is a bit of an unfortunate stray in that it CAN be invoked elsewhere but should only work with $addToSet
  • Culled useless version javadoc cruft & added @see to all bareword query ops
  • Fixes #3 Resolved an issue where the $or wasn't being broken into individual documents as expected. Should be heavily tested.
  • Fixes #15 Wrap the return of operator calls in a DBObject constructor.
  • Fixed anchoring on CoreOperators - things like $mod and such needed to be callable
  • Fixes #3 dded $or operator with proper right-hand array creation.
  • Added @see linkage to each core operator's doc page
  • Fixes #10, added $slice operator
  • Removed 'batchSafely' methods as they are now deprecated and counterindicated
  • Renamed 'safely' methods to 'request' as they are NOT safe and 10gen convention is moving off 'safe' as a term.
  • Migration of codebase for 10gen ownership
    • Package changed from com.novus.* to com.mongodb.*
  • Cull casbah-mapper. Mapper now lives as an independent project at http://github.com/maxaf/casbah-mapper ...
  • bumped version of scala-time to the 0.2 release
  • stop spamming the logs when JodaDateTimeDeserializer is a NOOP
  • more thorough BigDecimal conversions & test
  • Fixes #24 adds DBList support via MongoDBList, following 2.8 collections
  • First pass at a DBList implementation
  • Forward port from 1.0.8.1 bugfix branch
    • Factory & Builder for MongoDBObject always return as DBObject
    • Unit test for reported issue
  • Added "safely" and "batchSafely" resource loaning methods on Collection & DB
    • Given an operation, uses write concern / durability on a single connection and throws an exception if anything goes wrong.
  • Adjusted boundaries on getAs and expand; the view-permitting Any was causing ambiguity issues at runtime with non AnyRefs (e.g. AnyVal).
  • Fixed an assumption in expand which could cause runtime failure
    • Updated MongoDBObject factory & builder to explicitly return a type; some pieces were assuming at runtime that it was a MongoDBObjectBuilder$anon1 which fucked things up beyond all recognition
    • New basic spec test for MongoDBObject, just validates expand behavior ATM
  • Beginning a massive refactor for submodules, which will become 1.1