Changelog

Here described only the breaking and most significant changes. The full changelog and documentation for all released versions could be found in nicely formatted commit history.

v5.15

• Local development has been migrated to using Hatch

v5.14

• Rebased packaging on PEP 621
• Extracted experimental application/server from the codebase

v5.13

• Implemented "Metadata.from_descriptor(allow_invalid=False)" (#1501)

v5.10

• Various architectural and standards-compatibility improvements (minor breaking changes):
  • Added new Console commands:
    • list
    • explore
    • query
    • script
    • convert
    • publish
  • Rebased Console commands on Rich (nice output in the Console)
  • Fixed extract returning the results depends on the source type (now it's always a dictionary indexed by the resource name)
  • Enforced type safety -- many tabular command will be marked as impossible for non-tabular resources if a type checker is used
  • Improved frictionless.Resource(source) guessing abilities; if you just like to open a table resource use frictionless.resources.TableResource(path=path)

v5.8

• Implemented Implemented catalog/dataset/package/resource.deference (#1451)

v5.7

• Various architectural and standards-compatibility improvements (minor breaking changes):
  • Improved type detection mechanism (including remote descriptors)
  • Added resources module including File/Text/Json/TableResource
  • Deprecated resource.type argument -- use the classes above
  • Changed catalog.packages[] to catalog.datasets[].package
  • Made resource.schema optional (resource.has_schema is removed)
  • Made resource.normpath optional (resource.normdata is removed)
  • Standards-compatability improvements: profile, stats
  • Renamed system/plugin.select_Check/etc to system/plugin.select_check_class/etc

v5.6

• Added support for sqlalchemy@2 (#1427)

v5.5

• Implemented program/resource.index preview (#1395)

v5.4

• Support dialect.skip_blank_rows (#1387)

v5.3

• Support steps.resource_update for resource transformations (#1381)

v5.2

• Added support for wkt format in fields.StringField (#1363 by @jze)

v5.1

• Support descriptor argument for actions/program.extract (#1372)

v5.0

• Frictionless Framework (v5) is out of Beta and released on PyPi

v5.0.0b19

• Implemented CKAN Integration (#1185)

v5.0.0b8

• ForeignKeyError has been extended with additional information: fieldNames, fieldCells, referenceName, and referenceFieldNames

v5.0.0b2

• Implemented Github Integration (#1185)

v5.0.0b1

• First beta version of Frictionless Framework (v5)

v4.40

• Added Dialect support to packages (#1137)

v4.39

• Fixed processing of incompatible decimal char in table schema and data (#1089)
• Added support for Time Zone data (#1097)
• Improved validation messages by adding summary and partial validation details (#1106)
• Implemented new feature summary (#1127)
  • schema.to_summary
  • report.to_summary
  • Added CLI command summary
• Fixed file compression package.to_zip (#1104)
• Implemented feature to validate single resource (#1112)
• Improved error message to notify about invalid fields (#1117)
• Fixed type conversion of NaN values for data of type Int64 (#1115)
• Exposed valid/invalid flags in CLI extract command (#1130)
• Implemented feature package.to_er_diagram (#1135)

v4.38

• Implemented checks.ascii_value (#1064)
• Implemented checks.deviated_cell (#1069)
• Implemented detector.field_true/false_values (#1074)

v4.37

• Deprecated high-level legacy actions (use class-based alternatives):
  • describe_*
  • extract_*
  • transform_*
  • validate_*

v4.36

• Implemented pipeline actions:
  • pipeline.validate (will replace validate_pipeline in v5)
  • pipeline.transform (will replace transform_pipeline in v5)
• Implemented inqiury actions:
  • inqiury.validate (will replace validate_inqiury in v5)

v4.35

• Implemented schema actions:
  • Schema.describe (will replace describe_schema in v5)
  • schema.validate (will replace validate_schema in v5)
• Implemented new transform steps:
  • steps.field_merge
  • steps.field_pack

v4.34

• Implemented package actions:
  • Package.describe (will replace describe_package in v5)
  • package.extract (will replace extract_package in v5)
  • package.validate (will replace validate_package in v5)
  • package.transform (will replace transform_package in v5)

v4.33

• Implemented resource actions:
  • Resource.describe (will replace describe_resource in v5)
  • resource.extract (will replace extract_resource in v5)
  • resource.validate (will replace validate_resource in v5)
  • resource.transform (will replace transform_resource in v5)

v4.32

• Added to_markdown() feature to metadata (#1052)

v4.31

• Added a feature that allows to export table schema as excel (#1040)
• Added nontabular note to validation results to indicate nontabular file (#1046)
• Excel stats now shows bytes and hash (#1045)
• Added pprint feature which displays metadata in a readable and pretty way (#1039)
• Improved error message if resource.data is not a string (#1036)

v4.29

• Made Detector's private properties public and writable (#1025)

v4.28

• Improved an order of the metadata in YAML representation

v4.27

• Exposed Dialect options via CLI such as sheet, table, keys, and keyed (#886)

v4.26

• Validate 'schema.fields[].example' (#998)

v4.25

• Allows descriptors that subclass collections.abc.Mapping (#985)

v4.24

• Added support for SqlDialect.basepath (#982) (https://framework.frictionlessdata.io/docs/tutorials/formats/sql-tutorial)

v4.23

• Added table dimensions check (#985)

v4.22

• Added "extract --trusted" flag

v4.21

• Added "--json/yaml" CLI options for transform

v4.20

• Improved layout/schema detection algorithms (#945)

v4.19

• Renamed inlineDialect.keys to inlineDialect.data_keys due to a conflict with dict.keys property

v4.18

• Normalized metadata properties (increased type safety)

v4.17

• Add fields, limit, sort and filter options to CkanDialect (#912)

v4.16

• Implemented system/plugin.create_candidates (#893)

v4.15

• Implemented system.get/use_http_session (#892)

v4.14

• SQL Where Clause (#882)

v4.13

• Implemented descriptor type detection for extract/validate (#881)

v4.12

• Support external profiles for data package (#864)

v4.11

• Added json argument to resource.to_snap

v4.10

• Support resource/field renaming in transform (#843)

v4.9

• Support --path CLI argument (#829)

v4.8

• Added support for Package(innerpath) argument for unzipping a data package's descriptor

v4.7

• Support control/dialect as JSON in CLI (#806)

v4.6

• Implemented describe_dialect and describe(path, type="dialect")
• Support --dialect argument in CLI

v4.5

• Implemented Schema.from_jsonschema (#797)

v4.4

• Use field.constraints.maxLength for SQL's VARCHAR (#795)

v4.3

• Implemented resource.to_view() (#781)

v4.2

• Make fields[].arrayItem errors more granular (#767)

v4.1

• Added support for fields[].arrayItem (#750)

v4.0

• Released frictionless@4 :tada:

v4.0.0a15

• Updated loaders (#658) (BREAKING)
  • Renamed filelike loader to stream loader
  • Migrated from text loader to buffer loader

v4.0.0a14

• Improve transform API (#657) (BREAKING)
  • Swithed to the transform_resource(resource) signature
  • Swithed to the transform_package(package) signature

v4.0.0a13

• Improved resource/package import/export (#655) (BREAKING)
  • Reworked parser.write_row_stream API
  • Reworked resource.from/to API
  • Reworked package.from/to API
  • Reworked Storage API
  • Reworked system.create_storage API
  • Merged PandasStorage into PandasParser
  • Merged SpssStorage into SpssParser

v4.0.0a12

• Improved transformation steps (#650) (BREAKING)
  • Split value/formula/function concepts
  • Renamed a few minor step arguments

v4.0.0a11

• Improved layout and data streams concepts (#648) (BREAKING)
  • Renamed data_stream to list_stream
  • Renamed readData to readLists
  • Renamed sample to fragment (sample now is raw lists)
  • Implemented loader.buffer
  • Implemented parser.sample
  • Added support for function based checks
  • Added support for function based steps

v4.0.0a10

• Reworked Error.tags (BREAKING)
• Reworked Check API and split labels/header (BREAKING)

v4.0.0a9

• Rebased on Detector class (BREAKING)
  • Migrated all infer_*, sync/patch_schema and detect_encoding parameters to Detector
  • Made resource.infer omit empty objects
  • Added resource.read_*(size) argument
  • Added resource.labels property

v4.0.0a8

• Improved checks/steps API (#621) (BREAKING)
  • Updated validate(extra_checks=[...]) to validate(checks=[{"code": 'code', ...}])

v4.0.0a7

• Updated describe/extract/transform/validate APIs (BREAKING)
  • Removed validate_table (use validate_resource)
  • Removed legacy Table and File classes
  • Removed dataflows plugin
  • Replaced nopool by parallel (not parallel by default)
  • Renamed report.tables to report.tasks
  • Rebased on report.tasks[].resource (instead of plain path/scheme/format/etc)
  • Flatten Pipeline steps signature

v4.0.0a6

• Introduced Layout class (BREAKING)
  • Renamed Query class and arguments/properties to Layout
  • Moved header options from Dialect to Layout

v4.0.0a5

• Updated transform API
  • Added transform(type) argument

v4.0.0a4

• Updated describe API (BREAKING)
  • Renamed describe(source_type) argument to type

v4.0.0a3

• Updated extract API (BREAKING)
  • Removed extract_table (use extract_resource with the same API)
  • Renamed extract(source_type) argument to type

v4.0.0a1

• Initial API/codebase improvements for v4 (BREAKING)
  • Allow Package/Resource(source) notation (guess descriptor/path/etc)
  • Renamed schema.infer -> Schema.from_sample
  • Renamed resource.inline -> resource.memory
  • Renamed compression_path -> innerpath
  • Renamed compression: no -> compression: ""
  • Updated Package/Resource.infer not to infer stats (use stats=True)
  • Removed Package/Resource.infer(only_sample) argument
  • Removed Resouce.from/to_zip (use Package.from/to_zip)
  • Removed Resouce.source (use Resource.data or Resource.fullpath)
  • Removed package/resource.infer(source) argument (use constructors)
  • Added some new API (will be covered in the updated docs after the v4 release)

v3.48

• Make Resource independent from Table/File (#607) (BREAKING)
  • Resource can be opened like Table (it's recommended to use Resource instead of Table)
  • Renamed resource.read_sample() to resource.sample
  • Renamed resource.read_header() to resource.header
  • Renamed resource.read_stats() to resource.stats
  • Removed resource.to_table()
  • Removed resource.to_file()

v3.47

• Optimize Row/Header/Table and rename header errors (#601) (BREAKING)
  • Row object is now lazy; it casts data on-demand preserving the same API
  • Method resource/table.read_data(_stream) now includes a header row if present
  • Renamed errors.ExtraHeaderError->ExtraLabelError (extra-label-error)
  • Renamed errors.MissingHeaderError->MissingLabelError (missing-label-error)
  • Renamed errors.BlankHeaderError->BlankLabelError (blank-label-error)
  • Renamed errors.DuplicateHeaderError->DuplicateLabelError (duplicate-label-error)
  • Renamed errors.NonMatchingHeaderError->IncorrectLabelError (incorrect-label-error)
  • Renamed schema.read/write_data->read/write_cells

v3.46

• Renamed aws plugin to s3 (#594) (BREAKING)

$ pip install frictionless[aws] # before
$ pip install frictionless[s3] # after

v3.45

• Drafted support for writing Multipart Data (#583)

v3.44

• Added support for writing to Remote Data (#582)

v3.43

• Add support to writing to Google Sheets (#581)
• Renamed gsheet plugin/format to gsheets (BREAKING: minor)

v3.42

• Added support for writing to S3 (#580)

v3.41

• Update Loader/Parser API to write to different targets (#579) (BREAKING: minor)

v3.40

• Implemented a standalone multipart loader (#573)

v3.39

• Fixed Header not being an original one (#572)
• Fix bad format validation (#571)
• Added default errors limit equals to 1000 (#570)
• Added support for field.float_number (#569)

v3.38

• Improved ckan plugin (#560)

v3.37

• Remove not working elastic plugin draft (#558)

v3.36

• Support custom types (#557)

v3.35

• Added "resolve" option to "resource/package.to_zip" (#556)

v3.34

• Moved frictionless.controls to frictionless.plugins.* (BREAKING)
• Moved frictionless.dialects to frictionless.plugins.* (BREAKING)
• Moved frictionless.exceptions.FrictionlessException to frictionless.FrictionlessException (BREAKING)
• Moved excel dependencies to frictionless[excel] extras (BREAKING)
• Moved json dependencies to frictionless[json] extras (BREAKING)
• Consider json files to be a metadata by default (BREAKING)

Code example:

# Before
# pip install frictionless
from frictionless import dialects, exceptions
excel_dialect = dialects.ExcelDialect()
json_dialect = dialects.JsonDialect()
exception = exceptions.FrictionlessException()

# After
# pip install frictionless[excel,json]
from frictionless import FrictionlessException
from frictionless.plugins.excel import ExcelDialect
from frictionless.plugins.json import JsonDialect
excel_dialect = dialects.ExcelDialect()
json_dialect = dialects.JsonDialect()
exception = FrictionlessException()

v3.33

• Implemented resource.write (#537)

v3.32

• Added url parameter to SQL import/export (#535)

v3.31

• Made tables with header and no data rows valid (#534) (BREAKING: minor)

v3.30

• Various CLI improvements (#532)
  • Added autocompletion
  • Added stdin support
  • Added "extract --csv"
  • Exposed more options

v3.29

• Added experimental CKAN support (#528)

v3.28

• Add a "nopool" argument to validate (#527)

v3.27

• Stop sorting keyed sources as the order is now guaranteed by Python (#512) (BREAKING)

v3.26

• Added "nolookup" argument for validate_package (#515)

v3.25

• Add transform functionality (#505)
• Methods schema.get/remove_field now raise if not found (#505) (BREAKING)
• Methods package.get/remove_resource now raise if not found (#505) (BREAKING)

v3.24

• Lower case resource.scheme/format/hashing/encoding/compression (#499) (BREAKING)

v3.23

• Support "header_case" option for dialects (#488)

v3.22

• Added suppport for DB2 format (#485)

v3.21

• Improved SPSS plugin (#483)
• Improved BigQuery plugin (#470)

v3.20

• Added support for SQL Views (#466)

v3.19

• Rebased AwsLoader on streaming (#460)

v3.18

• Added hashing parameter to describe/describe_package
• Removed table.onerror property (BREAKING)

v3.17

• Added timezone for datetime/time parsing (#457) (BREAKING)

v3.16

• Fixed metadata.to_yaml (#455)
• Removed the expand argument from metadata.to_dict (BREAKING)

v3.15

• Added native schema support to SqlParser (#452)

v3.14

• Make Resource the main internal interface (#446) (BREAKING: for plugin authors)
• Move Resource's stats to resource.stats (BREAKING)
• Rename on_error to onerror (BREAKING)
• Added resource.stats.fields

v3.13

• Add an on_error argument to Table/Resource/Package (#445)

v3.12

• Added streaming to the extract functions (#442)

v3.11

• Added experimental BigQuery support (#424)

v3.10

• Added experimental SPSS support (#421)

v3.9

• Rebased on a goodtables successor versioning

v3.8

• Add support SQL/Pandas import/export (#31)

v3.7

• Add support for custom JSONEncoder classes (#24)

v3.6

• Normalize header terminology

v3.5

• Initial public version