All notable changes to this project will be documented in this file. This change log follows the conventions of keepachangelog.com.
- Utility function to get lazy seq of all indexed IDs
- Persist index in temp file system for fast future initialization
- Eliminate reflective call when querying file
- Persist the processed index to disk for fast re-initialization
- Saves to temp filesystem folder (default)
- optionally a different folder to persist index between system reboots
- Uses filename, content and optionally regex-string to name the index file
0.4.0 - 2021-06-24
0.4.0
- simpler and smaller!
- Auto-infer
:doc-type
from db file extension (*.ndedn
-> :doc-type :edn
)
- This means you have to use either db extension
.ndedn
|.ndjson
or :doc-type :json
|:edn
- Defaults to
:json
if extension is unknown and :doc-type
isn't set
- Laying the groundwork for improved indexing performance via parallelization.
- Need more work to limit memory consumption for huge databases, before enabling it
- Rename core namespace to
nd-db.core
, and the library from luposlip/ndjson-db
to com.luposlip/nd-db
!
- Removed core.memoize and timbre (not used anymore)
- Updated clojure (-> 1.10.3)
0.3.0 - 2021-06-09
- Add support for the
.ndedn
file format, where all lines are well formed EDN documents.
0.2.2 - 2021-01-12
- Updated depencies (timbre, memoize, cheshire)
0.2.1 - 2019-08-29
- Updated depencies (clojure, cheshire, memoize)
0.2.0 - 2019-05-06
- Completely revamped API
clear-all-indices!!
-> clear-all-indexes!!
- Using timbre for logging
- Now using Apache License, Version 2.0 (instead of Eclipse Licence 2.0)
0.1.2 - 2019-04-01
- API for using default id-fn for querying by json name with type string or integer
- Added
clear-all-indices!!
and clear-index!
- Added documentation in README
0.1.1 - 2019-03-03
- Enhanced API for lazy/streaming usage
- Initial public release
- Example on how to query huge datasets