Skip to content

Latest commit

 

History

History
455 lines (318 loc) · 11.3 KB

CHANGELOG.md

File metadata and controls

455 lines (318 loc) · 11.3 KB

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[3.4.0] – 2024-03-06

Added

[3.3.1] – 2022-08-25

Changed

  • Fetch the data download URL from IATI data dump (rather than hardcoding it)

[3.3.0] – 2022-04-27

Changed

  • Changed exception raised if data is missing, from TypeError to FileNotFoundError

[3.2.1] – 2021-11-17

Fixed

  • Fixed unzipping bug, so that download correctly deals with folder structure of zip file

[3.2.0] – 2021-11-16

Changed

  • Change path to IATI data zip file
  • Drop support for python 3.5

[3.1.0] – 2021-10-05

Fixed

  • Include incomplete codelists in mapping file (#54)

Changed

  • Change path to IATI data zip file

[3.0.3] – 2021-02-17

Fixed

  • raw_xml should return a byte string

[3.0.1] – 2021-02-16

Added

  • Add Dataset.raw_xml

[3.0.0] – 2021-02-15

Fixed

  • Fix bug so that PublisherSet deals with missing XML files. (#52)

Changed

  • Drop support for python 2.7

[2.3.0] – 2020-12-16

Changed

  • Drop support for python 3.4 (again)
  • Add support for python 3.8

[2.2.12] – 2020-12-16

Changed

  • Unpin lxml requirement
  • Update codelist download locations

[2.2.11] – 2020-07-31

Fixed

  • Fix a longstanding python2.7 validator bug
  • Fix a python 2.7 bug introduced in v2.2.10

[2.2.10] – 2020-07-31

Fixed

  • Deal with missing XML files gracefully. (#52)

Changed

  • Don’t download metadata from gist (it’s now included in the IATI Data Dump zip file)

[2.2.9] – 2019-11-09

Changed

  • Update URL for IATI data dump metadata (again)

[2.2.8] – 2019-09-11

Added

  • Add a contributing guide

Fixed

  • Pin lxml, to maintain support for python 3.4

Changed

  • Update URL for IATI data dump metadata

[2.2.7] – 2019-07-25

Added

  • Add a code of conduct

Fixed

  • Fix travis deployment for dot com
  • Improve wording in installation docs

Changed

  • Change ownership of repo again

[2.2.4] – 2019-05-19

Added

  • Added some Organisation tests
  • Add support for python 3.4 and python 3.7

[2.2.3] – 2019-05-18

Fixed

  • Don’t use the root logger. Instead, use iatikit’s own logger, and default to NullHandler. (#46)

[2.2.2] – 2019-05-16

Added

  • API documentation for Organisation and OrganisationSet

Fixed

  • Make dataset.root return None if XML is invalid

[2.2.1] – 2019-05-06

Fixed

  • Make multiple modifiers work (#47)

[2.2.0] – 2019-05-06

Added

  • Added a dataset xpath filter

Fixed

  • Allow multiple filters of the same type to be specified (#45)
  • Prevent looping twice when calling all() (#44)

Removed

  • Remove (experimental) fast identifier search

[2.1.0] – 2019-04-21

Changed

  • Relax codelist item comparison (#41)

[2.0.8] – 2019-04-20

Added

  • First pass at modelling IATI organisations (#18)

[2.0.7] – 2019-04-20

Added

  • First pass at a humanitarian activity filter
  • Add a iatikit.download.standard() command
  • Add an optional path argument to iatikit.data()

[2.0.6] – 2019-04-04

Added

  • Add some Windows installation instructions (tl;dr: use Jupyter)

Fixed

  • Deal with null paths
  • Use huge_tree when parsing XML (to ensure big files still parse)

[2.0.5] – 2019-03-12

Fixed

  • Fix "metadoota" typo!

[2.0.4] – 2019-03-12

Added

  • (Re-)add metadata-only download

[2.0.3] – 2019-03-12

Fixed

  • Fix dataset name filter

[2.0.2] – 2019-03-12

Changed

  • Always use absolute paths to elements
  • Refactor to better handle missing data

[2.0.1] – 2019-02-27

Added

  • Add a dataset schema getter, and caching
  • Add line number and path to codelist validation errors
  • Add dataset.unminify_xml(), to ensure validation line numbers are meaningful
  • Add activity-level schema validation
  • Allow codelists to be filtered by name

Changed

  • show() uses dataset metadata
  • Pretty print XML
  • Rejig validation; make dataset.version fail on XML error
  • Make dataset.filetype return None on failure

Fixed

  • Make XSD Schema constructor a bit more robust, and improve error handling
  • Fix bug related to file and folder sort order
  • Fix incorrect XSD error message regarding element ordering
  • Allow datasets to be loaded from a StringIO
  • Handle missing metadata in old codelists

[2.0.0] – 2019-01-17

Added

  • Much better XSD and codelist validation messages

Fixed

  • Add checks to ensure filetype metadata from the registry is okay

Changed

  • Rename the project! "pyandi" -> "iatikit"

Removed

  • Drop validate_unique_ids. This should be part of ruleset validation

[1.5.7] – 2019-01-15

Added

  • Add an IATI schema downloader
  • Add IATI schema validation
  • Add a fast find activity search (#3)
  • Add CodelistItem and Sector comparison methods
  • Add a dataset.schema method
  • Make download.codelists() download codelist mappings, too
  • Add a CodelistMappings class; Add codelist validation (#31)
  • Add a Validator class, for storing the results of validation
  • Remove path args from lots of constructors; use pyandi.ini instead (#31)

Fixed

  • Lots of pylint-related fixes
  • Add missing _instance_class to PublisherSet
  • Don’t patch os.path.join in tests (#32)

Changed

  • Don’t instantiate schemas – use static classes instead

Removed

  • Remove download helpers (they’re not really useful)

[1.5.6] – 2019-01-08

Added

  • Add classifiers to setup

Changed

  • More import shuffling

Fixed

  • Fix pyversion badge
  • Fix typo in docs

[1.5.5] – 2019-01-08

Changed

  • Only declare requirements in setup
  • Run tests with python setup.py test
  • Run coverage by default when testing
  • Shuffle imports about
  • Remove support for python 3.4

[1.5.4] – 2019-01-03

Added

  • Add a __version__ string to the module (to help with debugging)
  • Add badges to docs
  • Use more custom exceptions
  • Add lots more tests

Fixed

  • URL-encode d-portal param
  • Ensure xml property returns a byte-string
  • where should not modify the original set object

Changed

  • Split codelist attributes and metadata
  • Rename codelist name filter to slug
  • Make codelist.complete return a boolean

[1.5.3] – 2019-01-02

Added

  • Various unit tests
  • Make DateType filter accept a datetime.date
  • Add path argument to Sector constructor
  • Add a "Deployment" section to the README

Changed

  • Require unicodecsv

Fixed

  • Get tox working again

Removed

  • Comment out unused org-related code (temporarily)

[1.5.2] – 2018-12-28

Added

  • Deploy to pypi from travis (#26)

Changed

  • Move repo from andylolz to pwyf

[1.5.1] – 2018-12-28

Added

  • Log a warning when dataset XML is invalid
  • Add some documentation in docs/ as well as some docstrings (#16)
  • Be more specific about exception handling (i.e. don’t use except:)
  • Add python 2.7 support
  • Start adding tests; Setup travis and coveralls

Changed

  • Make dataset.root less strict
  • metadata should return empty dict if file not found
  • Simplify some set operations

[1.5.0] – 2018-12-27

Added

  • Add a publisher metadata property (#17)
  • Add filter as a synonym for where
  • Add show() function to Publisher, Dataset and Activity models
  • Add an xpath activity filter

Changed

  • find and first raise errors when no data found. get returns a default value
  • xml property moved to etree; raw_xml property moved to xml

Removed

  • get() shouldn’t be a synonym for all()

[1.4.0] – 2018-12-21

Added

  • Codelist support for all versions (i.e. v1.01+)
  • Add start and end properties to activities
  • Error when there’s no data (#12)
  • Error when there are no codelists
  • Improved sector support (#22)

Changed

  • Change internal representation of codelists (#20)
  • Return python date objects for dates
  • Always strip IATI identifiers

[1.3.1] – 2018-12-17

Fixed

  • README render

[1.3.0] – 2018-12-17

Added

  • Support set slicing
  • Add a getter to all Sets. A set-getter!
  • Ensure only known filters are used
  • Allow for codelist filtering by code
  • Add codelist data and metadata properties

Changed

  • Tidy up Activity model repr
  • Reword examples in README slightly, for clarity
  • Refactor activity property fetching, to be less clever
  • Rename all exceptions, according to PEP8
  • Major improvements to codelist code
  • Always cast codelist codes to string

Fixed

  • Get date type exists filter working
  • DateTypes should also be filterable by generic filters

Removed

  • Remove some cruft from the README about schema download

[1.2.0] – 2018-12-10

### Added

  • Add activity date filters

Changed

  • Add path to codelist functions

Removed

  • Remove schema download code
  • Remove unused xmlschema package

[1.1.1] – 2018-12-10

Fixed

  • Fix exception when there’s no existing data (bug in stale data warning)

[1.1.0] – 2018-12-10

Added

  • raw_xml function added to dataset and activity
  • Expand key-value extras in dataset metadata
  • Add a filetype dataset filter
  • Add a warning about stale data (#8)
  • Add location, sector, title and description schema fields

Changed

  • Make it possible to download data to a custom path
  • is_valid should just check for valid XML
  • Determine filetype from metadata (instead of XML)
  • Make activity.dataset public
  • Use custom exceptions throughout

Fixed

  • Fix broken @version property
  • Fix typo so metadata works again

[1.0.0] – 2018-12-06

Added

  • Add functions to fetch all datasets and all activities

Changed

  • Move examples to README
  • Standardise the set API with an abstract class
  • Compute set size more efficiently
  • Make API more consistent

[0.2.0] – 2018-12-05

Changed

  • Remove download.metadata() (since metadata is now included in the data dump)
  • Start using python logging
  • Don’t parse XML until necessary
  • Where path is useless – scrap it
  • Scrap custom cache location

Fixed

  • Minor README updates

[0.1.0] - 2018-12-04

Added

  • Initial release