Skip to content

Latest commit

 

History

History
135 lines (111 loc) · 5.09 KB

CHANGELOG.md

File metadata and controls

135 lines (111 loc) · 5.09 KB

Changes

1.4.0

  • Correct way to get number of dataframe rows #266
  • Style updates for data service frontend #262
  • SDMX metadata output #261
  • Metadata input from "SDG Metadata" repositories #259
  • Replace values individually to avoid exceptions #258
  • Vary by language in some method caches #257
  • Allow suffix on imported metadata keys #253
  • Fix empty values by column to avoid datatype issues #252
  • Use TIME_PERIOD instead of TIME_DETAIL in SDMX output #250
  • Allow TIME_PERIOD to exist in SDMX DSD data schema #249
  • Defaults for mandatory attributes in SDMX output #247
  • Fallback to translation keys if names are missing #246
  • Do not sort datapackage schemas that were intentionally specified #245
  • New automated metadata fields for Git dates (only) #244
  • SDMX and file helper functions #242
  • Bugfix - structure-specific default #237
  • Allow column and code mappings to be used #236, #251, #256, #264
  • Improvements of SDMX output #217, #218, #231
  • Metadata input for Word template files #230
  • JSON-Stat and PxWeb data inputs #219

1.3.0

  • Allow numeric disaggregaton #198
  • Use only 2 segments for the auto target id #196
  • Allow missing metadata to be altered #194
  • Performance optimizations and logging option #192
  • Allow the export filename to be configured #189
  • Allow missing values in reporting status extra fields #188
  • Input from the UN SDGs-SDMX API #187
  • Better handling of common SDMX data fields #182
  • Input for SDMX metadata #180
  • Standalone indicators #179
  • SDMX output for SDMX-compliant datasets #177

1.2.0

  • Output documentation baseurl (#178)
  • Yaml meta input (#176)
  • Fix default translation repo (#175)
  • Disaggregation stats in Open SDG output (#174)
  • Use column as a default group when translating data (#173)
  • Output untranslated build along with translation builds (#172)
  • Allow multiple meta file inputs from different folders (#170)
  • Service to copy indicator data files to the build for direct downloads (#167)
  • Handle Obs elements without ObsValue (#165)
  • Fix call for base class for indicator ids (#163)
  • Add a disaggregation report to automated docs (#158)
  • More efficient generation of unique combinations of disaggregations (#156)
  • Automatic output documentation (#153)
  • Don't omit the SERIES ids from disaggregations in SDMX inputs (#150)
  • Special treatment of 'Series' column (#148)
  • Do not translate 'Year', 'Value', or 'Units' in data column names (#146)

1.0.0

  • Allow normal Prose fields to be boolean too (#142)
  • Print warning instead of exception when series has duplicate value (#140)
  • Preserve the order of the fields in the schema input/output (#139)
  • Allow translations to override existing ones (#137)
  • Reporting_status also to default to complete if data_non_statistical is true (#128)
  • Add Excel metadata input and make CSV metadata input consistent (#126)
  • SDMX-ML multiple input (#123)
  • Change the Open SDG config syntax to support non-standard inputs (#122)
  • Better Windows support (#119, #124)
  • Support subfolder translation in translated builds (#114)
  • Fix logic of all-numeric check in stats.py (#113)
  • Logic fix to give special validation to multiselects and selects (#111)
  • Restore the 'git' and 'git_data_dir' functionality (#110)
  • Translate nested metadata (#106)
  • Set indicator_name instead of indicator_name_national (#105)
  • Save json info about the zip file (#103)
  • Allow CSV metadata input (#102)
  • Never use [] or {} as parameter defaults (#100)
  • GeoJSON output (#99)
  • Only zip indicators that are complete and statistical (#98)
  • Default to True for git in Yaml metadata input (#96)
  • Allow SDMX to import translation keys instead of text (#91, #107)
  • Fix for list index bug (#89)
  • Allow translations in CSV format (#87)
  • Ensure numeric values in SDMX inputs (#81, #82)
  • Allow reporting stats to include extra fields (#80)
  • Allow decimals in metadata (#72)
  • Remove references to 'published' field (#76)
  • Default to the object-oriented API (#77)
  • Support configuration via a "config.yml" file for Open SDG usage (#77)
  • Add dataframe checks to object-oriented validation (#79)

0.5.0

  • Add an input for CKAN instances (#51)
  • Add ability to alter data and metadata during input (#51)
  • Add inputs for YAML/SDMX translation data and JSON output (#60)
  • Add optional fully-translated builds (#63)
  • Add zips of data to builds (#64)
  • More helpful validation errors (#66)
  • Automated tests for object-oriented builds (#67)
  • More lenient validation rules (#69, #70, #71)

0.4.0

  • Add support for SDMX-ML and SDMX-JSON input
  • Drop support for Python 3.4

0.3.1

Minor release to ignore columns used in SDMX

0.3.0

Adding the stats endpoint which is where reporting statistics will live. This also raises the importance of the schema.

  • Reporting Status page is computed here. Allowed statuses and their names come from the schema, which is the _prose.yml file in the data repo.

0.2.1

Bug fix for edges json file which was being overwritten by headline data.

0.2.0

Generalising beyond UK

  • Support binary chart type
  • Allow flexible data directories
  • Multi-lingual support

0.1.0

Initial release from UK version at ONSdigital