Skip to content

Latest commit

 

History

History
100 lines (76 loc) · 4.76 KB

CHANGELOG.md

File metadata and controls

100 lines (76 loc) · 4.76 KB

Pull Requests: https://github.com/araddon/qlbridge/pulls?q=is%3Apr+is%3Aclosed

2017 Updates

  • Calculate Time Boundarys for Datemath expressions #183
  • Expression inliner for INCLUDE (referenced expressions) #182
  • Code Coverage improvements in value pkg #178
  • Cleanup the CREATE ... Statement #175
  • Exists() as part of column expression lex bug #167
  • Base64 Builtin functions #166
  • File based datasources support iterator instead of in-mem file list #165
  • Elasticsearch Sql -> ES search DSL generator #160
  • Completely revamp/improve performance of VM functions by removing reflect usage #148

Sept 2016

  • Expression Massive Revamp #125
    • Include and FilterQL moved into native Expresson parser/vm not separate dialect
    • Implement generic Expr type for usage as json representation of Node/Expressions

Aug 2016

  • Schema Improvements for internal schema mgmt #124
  • Column existence check #121
  • Filter QL improvements #114
  • Improve SQL Projections #123

July 2016

  • Handle Identity Quotes/Escaping and Left/Right better.
  • Support Dates in Between #109
  • New HasSuffix and HasPrefix functions #107
  • Dialect specific quote marks. Includes new SQL Rewriter #103
  • Function Registry #101
  • #100
    • SQL OrderBy
    • cleaner lex quoting
    • FilterQL Doc

June 1016

  • Schema Improvements #97
  • FilterQL Enhancments #96
  • Information Schema #91

May 2016

  • Filter Function #86

April 2016

  • sql SET @var = "stuff" #79
  • Better runtime close channel mgmt #80

March 27th 2016

  • Better internal Schema Query planning, system (SHOW, DESCRIBE) with schemadb #68
    • introspect csv files for types
    • convert SHOW, DESCRIBE into SELECT statements
    • better internal data-source registry

v 0.12 February 2016

  • Enable Distributed runtime by Executor interface #66
    • add support for WITH key=value pairs in sql dialect, ie SELECT title FROM article WITH distributed=true, node_ct=20
    • Support planner/executors to be swapped out with custom implementations, so upstream can implemented distributed planners.
      • expr, sql, plan support protobuf serialization
      • separate out the planner, executor so planning can occur on one master node, and send request (dag of plan tasks) to slave executor nodes.
      • support partitionable sources (run partion 1 on node 1, partition 2 on node 2, etc....)

v 0.11 December 2015

  • #61

    • convert IN statement from MultiArg -> BinaryNode w ArrayNode type
    • cleanup remove un-used interface methods (NodeType())
    • fingerprint() for filterql
    • fix negateable string for BinaryNode (LIKE)

  • #56

    • GroupBy implementation in execution engine
    • count, avg, sum functions for aggs
    • recognize aggregate functions (count, sum) without group-by in parser

v 0.10 December 2015

  • #54

    • implmennt FilterQL in vm. filterQL is a WHERE filter language with dsl nesting
    • context-wrapper, allow go structs to be used natively in vm including functions

  • #52

    • move datasource schema structures into schema pkg
    • new plan.context simplify interfaces and pass through ctx to runtime queries
    • implement literal query select 1;
    • implement @@session.max_allowed_packets type variables in both Global, Session contexts
    • new staticsource data type for simpler static-data returns
    • better support for SHOW FULL COLUMNS, SHOW CREATE TABLE
    • new Wrap interface on job builders allowing frontends to override query behavior