.. -*- coding: utf-8 -*- Version 0.3.0 ------------- Refactoring of code. :publication date: 2013/11/05 :expected date: n/a Ticket #183459 Add a logger ~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved Nazca alignment should have a logger rather than print Ticket #182030 Align from Files ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved It should be interesting to have an aligner object to align data directly from files. Ticket #183448 Autocast SparqlIO ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved SparqlIO autocast should be optional. Ticket #183470 Extract cdist in a function ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved Extract the "cdist" method of the BaseProcessing into a function, in ordre to use it in a similar way as scipy cdist(). Ticket #182023 Implement blockings ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved Implement standard blokcings techniques (e.g. NGram) Ticket #183415 Indice VS id ~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved The blocking technics should return both indices (relative identification of records) or ids (absolute identification of records). Ticket #183446 Key-based blocking safety belt ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved Add a safety belt for key blocking when key does not exist. Ticket #182033 Minhashing error with shape < 1 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved There is an error with Minhashing with a shape < 1. Ticket #183457 Normalization of the distance matrix ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved The distance matrix may be (optionally) normalized. Ticket #183461 Old API ~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved Test the old api for backward compat. Ticket #183439 Output results ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved Aligner should output some statistics on the alignments done. Ticket #182035 Pipeline aligners ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved With the new Object-based API, it should be interesting to have a way to pipeline aligners technics. Ticket #183468 Pipeline aligners unique ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved Pipeline aligners does not correctly handle unicity for now. Simplified them, and report to a 0.4.0 the full complexity of the pipeline (see https://www.logilab.org/ticket/183467) Ticket #182032 Pipeline blockings ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved With the new Object-based API, it should be interesting to have a way to pipeline blockings technics. Ticket #182001 Refactoring of Nazca API ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 1.000 :state: resolved IT could be intersting to change the function-based API of Nazca to an object-based on, in order to allow easy pipelining and avoid using functions with more than 10 parameters. Ticket #183444 Reference data ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved Add reference data for alignments (countries, states, ...) Ticket #183463 Return distance ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ :type: bug :load: 0.500 :state: resolved We should return the aligned pairs and the distance.