v4.0.0

This release contains a number of breaking changes, thus the version bump to 4.0. See the upgrade guide for the full details and steps for quickly updating your code to work with Biip 4.0.

Breaking changes

• Result objects are now immutable. (PR: #393)
• biip.parse() no longer raises ParseError when all parsers fail. (PR: #376)
• All configuration options must be passed via the new ParseConfig class. (PR: #385)
• Several classes and fields in the low-level API have been moved or renamed. (PR: #375, #379, #380, #381, #382, #391, #392)
• Python 3.9 is no longer supported. (PR: #383)

See the upgrade guide for full details.

Support for GS1 Web URIs

Biip can now work with GS1 Web URIs. (Fixes: #83, PR: #387)

All GS1 Application Identifiers are supported:

>>> import biip
>>> result = biip.parse("https://www.example.com/products/gtin/07032069804988?10=0329&15=210526")
>>> result.gs1_web_uri.as_canonical_uri()
'https://id.gs1.org/01/07032069804988/10/0329?15=210526'

Conversion back and forth between GS1 Web URIs and GS1 Messages are fully supported:

>>> message = result.gs1_web_uri.as_gs1_message()
>>> message.as_hri()
'(01)07032069804988(10)0329(15)210526'
>>> web_uri = message.as_gs1_web_uri()
>>> web_uri.as_uri(domain="example.com", prefix="products")
'https://example.com/products/01/07032069804988/10/0329?15=210526'

See the GS1 Web URIs section in the quickstart guide for an introduction.

Date validation in GS1 element strings can now be disabled

It is now possible to disable validation of the dates in GS1 element strings with the gs1_element_strings_verify_date option. When validation is disabled, invalid date values are still available as text in the GS1ElementString.value field, but the GS1ElementString.date field is None. (Fixes: #247, PR: 386, #389)

Revamped documentation

This release comes with revamped documentation, where all parts of the documentation has been reviewed and polished. Also, the documentation system has been switched from Sphinx to MkDocs.

Support for pretty-printing with Rich

All objects returned from Biip now supports the Rich pretty-printing protocol, making them a lot nicer to work with in the terminal. Biip does not depend on Rich, so you have to install it yourself, and then use Rich's print function to print Biip objects.

>>> from rich import print
>>> import biip
>>> print(biip.parse("7032069804988"))

Fixes

• Correctly attribute GTIN-8 prefixes assigned to GS1 UK and GS1 Poland. (Fixes: #162, PR: #384)
• Update GS1 Prefix database. (PR: #382)
• Update GS1 Company Prefix database. (PR: #382)
• Update GS1 Application Identifier database. (PR: #382)

v3.6.0

This release has no changes for users, only for distribution packagers.

Changes

• Switched build backend from poetry to hatchling.
• Development is now done using uv and tox-uv.
• Replaced darglint docs linter with Ruff's pydoclint rules.
• Bumped some dev tooling versions.

v3.5.2

Fixes

• Fix error in AI 703x patterns. The root cause is an error in the data imported from gs1.org, which we now patch. (Fixes: #345, PR: #346)

If you're using Biip, please add yourself to the list of users!

v3.5.1

Changes

• Use Trusted Publishing when releasing to PyPI.

If you're using Biip, please add yourself to the list of users!

v3.5.0

Breaking changes

• Drop support for Python 3.8 (PR: #326)

Changes

• Use importlib.resources to load bundled data files so that Biip can be installed as a zip file. (PR: #324, #327)
• Update all bundled GS1 datasets. The GS1 Company prefixes is the only dataset with any changes.
• Test on Python 3.13

If you're using Biip, please add yourself to the list of users!

v3.4.0

Changes

• Accept redundant GS1 Message separator chars, both after fixed-length fields and at the end of the GS1 Message. The specification says that the separator characters "should not" be present in these locations, but doesn't forbid it. Multiple separator characters are also ignored. (Fixes: #176, PR: #304)

If you're using Biip, please add yourself to the list of users!

v3.3.0

Changes

• Add support for extracting dates and potentially datetimes from GS1 messages with AI 4324/4325/4326/7003/7006/7007/7011/8008. (PR: #301)

Fixes

• In the new GS1 AI dataset introduced in Biip 3.2.0, there are four AIs with optional pattern groups. When the value matches everything but the optional pattern group, the group's value becomes None. This caused Biip to crash when we tried building a string of the characters matching the pattern. (Fixes: #295, PR: #300)

If you're using Biip, please add yourself to the list of users!

v3.2.0

Changes

• Update all bundled GS1 datasets (PR: #292)
  • GS1 Application Identifiers including many stricter regexp patterns
  • GS1 Member Organization prefixes
  • GS1 Company prefixes

Development environment

• Replace Black with ruff. (PR: #239)

If you're using Biip, please add yourself to the list of users!

v3.1.0

Breaking changes

• Drop support for Python 3.7, as it reached end-of-life this summer. (#207)

Changes

• Add Python 3.12 to the test matrix. (#232)

Development environment

• Type check with pyright in addition to mypy, and fix all typing issues indicated by pyright. (#186)
• Replace flake8 and all flake8 plugins with ruff. (#184)

If you're using Biip, please add yourself to the list of users!

v3.0.0

This release contains a couple of breaking changes, thus the version bump to 3.0.

Breaking changes

• Parsing and validation errors of GLNs, GTINs, and SSCCs nested inside GS1 Element Strings no longer raises ParseError. Instead, the exception message is exposed on the GS1ElementString class as the new fields gln_error, gtin_error, and sscc_error. This is a breaking change, but makes it possible to extract some information from GS1 Element Strings that are not entirely valid. (Fixes: #157, PR: #169)
• Remove RcnRegion.from_iso_3166_1_numeric_code() class method which as been deprecated since Biip 2.2. (Fixes: #161, PR: #166)

GS1 Company Prefix support

• Add GS1CompanyPrefix class that can extract the company prefix from any GLN, GTIN, or SSCC. This is based on a data set from GS1 that is bundled with Biip, just like the GS1 Prefixes and GS1 Application Identifiers. The data set is preprocessed into a trie data structure, making company prefix length lookups fast. (Fixes: #160, PR: #172)
• Add and set company_prefix fields on the Gln, Gtin, and Sssc classes, based on the new GS1CompanyPrefix class. (Fixes: #160, PR: #174)
• Change Sscc.as_hri() to use GS1CompanyPrefix to automatically group the company prefix and the serial when rendering SSCC's for human consumption. The existing company_prefix_length parameter is now only used as an override in case you're rendering an SSCC whose company prefix is not included in the data set yet, or where you disagree with the data set. (Fixes: #160, PR: #173)

Other changes

• Add rcn_verify_variable_measure parameter to all functions and methods taking rcn_region as a parameter. The parameter defaults to True and the old behavior. By setting this parameter to False, it is possible to parse and extract variable measure from RCNs that have separate check digit for the variable measure, but where the check digit is either incorrect or simply not in use as a check digit. This parameter only has any effect on RCN regions with formats including check digits, e.g. Great Britain and Germany. (PR: #171)
• Update the bundled GS1 Application Identifier definitions. (PR: #164)
• Update the bundled GS1 Prefix definitions, and add script to automate future updates. (PR: #165)
• Add Python 3.11 to the test matrix, and updated development tooling. (PR: #163)

If you're using Biip, please add yourself to the list of users!