CHANGELOG.md

Changelog

1.8.1

• fix: more restrictive YYYYMM pattern to prevent ValueError with @b3n4kh (#145)
• maintenance: add pre-commit with checks with @nadasuhailAyesh12 (#142)

1.8.0

• change license to Apache 2.0 (#140)
• compile XPath expressions (#136)
• update docs with @EkaterineSheshelidze (#135)

1.7.0

• fix meta property updated vs. original behavior (#121)
• support for LXML version 5.0+ (#127)
• fix image links in Readme

1.6.1

• fix for MacOS: pin LXML dependency with @adamh-oai

1.6.0

• focus on precision, stricter extraction patterns (#103, #105, #106, #112)
• simplified code base (#108, #109)
• replaced lxml.html.Cleaner (#104)
• extended evaluation

1.5.2

• fix for missing months keys in custom extractor (#100)
• fix for None in try_date_expr() (#101)

1.5.1

• fix regression for fast extraction introduced in e8b3538 (#96)
• fix setup by making backports-datetime-fromisoformat optional (#95)

1.5.0

• slightly higher accuracy with revised heuristics
• simplified code structure for better performance
• setup: support for 3.12, fromisoformat backport if applicable
• HTML parsing fixes: more lenient parsing, pinned LXML version for MacOS

1.4.3

• maintenance release: upgrade urllib3 dependency

1.4.2

• support min_date/max_date as datetimes or datetime strings with @kernc (#73)
• add date attributes to HTML extraction with @kernc (#74)
• fix for extraction of updated and original dates in time elements
• code refactoring and maintenance

1.4.1

• better coverage of relevant HTML attributes
• automatically define upper time bound at each function call (#70)
• reviewed and simplified extraction code
• cache validation for format diverging from %Y-%m-%d
• updated dependencies and removed real-world tests from package

1.4.0

• additional search of free text in whole document (#67)
• optional parameter for subdaily precision with @getorca (#66)
• fix for HTML doctype parsing (#44)
• cleaner code for multilingual month expressions
• extended expressions for extraction in HTML meta fields
• update of dependencies and evaluation

1.3.2

• technical release: explicit support for Python 3.11 and logo

1.3.1

• fix for use of min_date & max_date (#62)
• simplified code & updated setup

1.3.0

• entirely type-checked code base
• new function clear_caches() (#57)
• slightly more efficient code (about 5% faster)

1.2.3

• fix for memory leak (#56)
• docs updated

1.2.2

• slightly higher accuracy & faster extensive extraction
• maintenance: code base simplified, more tests
• bugs addressed: #51, #54
• docs: fix by @MSK1582

1.2.1

• speed and accuracy gains
• better extraction coverage, simpler code
• bug fixed (typo in variable)

1.2.0

• better performance
• remove unnecessary ciso8601 dependency
• temporary fix for scrapinghub/dateparser#1045 bug

1.1.1

• bugfix: input encoding
• improved extraction coverage (#47)

1.1.0

• better handling of file encodings
• slight increase in accuracy, more efficient code

1.0.1

• maintenance release, code base cleaned
• command-line interface: --version added
• file parsing reviewed

1.0.0

• faster and more accurate encoding detection
• simplified code base
• include support for Python 3.10 and dropped support for Python 3.5

0.9.1

• improved generic date parsing (thanks @RadhiFadlillah)
• specific support for French and Indonesian (thanks @RadhiFadlillah)
• additional evaluation for English news sites (kudos to @coreydockser & @rahulbot)
• bugs fixed

0.9.0

• improved exhaustive search
• simplified code
• bug fixes
• removed support for Python 3.4

0.8.1

• bugfixes

0.8.0

• dateparser and regex modules fully integrated
• patterns added for coverage
• smarter HTML doc loading

0.7.3

• dependencies updated and reduced: switch from requests to bare urllib3, make chardet standard and cchardet optional
• fixes: downloads, OverflowError in extraction

0.7.2

• compatibility with Python 3.9
• better speed and accuracy

0.7.1

• technical release: package requirements and docs wording

0.7.0

• code base and performance improved
• minimum date available as option
• support for Turkish patterns and CMS idiosyncrasies (thanks @evolutionoftheuniverse)

0.6.3

• more efficient code
• additional evaluation data

0.6.2

• performance and documentation improved

0.6.1

• code base restructured
• bugs fixed and further tests
• restored retro-compatibility with Python 3.4

0.6.0

• reduced number of packages dependencies
• introduced and tested optional dependencies
• more detailed documentation on readthedocs

0.5.6

• tests on Windows
• compataibility and code linting

0.5.5

• tests on Linux & MacOS
• bugs removed

0.5.4

• manually set maximum date
• better precision
• temporarily dropped support for Python 3.4

0.5.3

• coverage extension

0.5.2

• small bugs and coverage issues removed
• streamlined utils
• documentation added

0.5.1

• bugs corrected and cleaner code
• more errors caught and better test coverage

0.5.0

• significant speed-up after code profiling
• better support of free text detection (DE/EN)

0.4.1

• fixed lxml dependency
• reordered XPath-expressions

0.4.0

• refined and combined XPath-expressions
• better extraction of dates in free text
• better coverage and consistency issues solved

0.3.1

• improved consistency and further tests

0.3.0

• improvements in markup analysis along with more tests
• higher resolution for free text detection (e.g. DD/MM/YY)
• download mode (serial on command-line)

0.2.2

• better code consistency
• tested for Python2 and 3 with tox and coverage stats

0.2.1

• refined date comparisons
• debug and logging options
• more tests and test files
• extensive search can be disabled

0.2.0

• refined targeting of HTML structure
• better extraction logic for plain text cases
• further tests

0.1.2

• better extraction
• logging
• further tests
• settings

0.1.1

• tests functions (tox and pytest)
• retro-compatibility (python2)
• minor improvements

0.1.0

• minimum viable package