rubygem-addressable-2.7.0-bp152.1.10

- New upstream release 2.7.0
  * added `:compacted` flag to `normalized_query`
  * `heuristic_parse` handles `mailto:` more intuitively
  * refactored validation to use a prepended module
  * dropped explicit support for JRuby 9.0.5.0
  * compatibility w/ public_suffix 4.x
  * performance improvements

- Update public_suffix gem dependency to < 5.0

- updated to version 2.6.0
  * added `tld=` method to allow assignment to the public suffix
  * most `heuristic_parse` patterns are now case-insensitive
  * `heuristic_parse` handles more `file://` URI variations
  * fixes bug in `heuristic_parse` when uri starts with digit
  * fixes bug in `request_uri=` with query strings
  * fixes template issues with `nil` and `?` operator
  * `frozen_string_literal` pragmas added
  * minor performance improvements in regexps
  * fixes to eliminate warnings

- updated to version 2.5.2
  see installed CHANGELOG.md
  [#] Addressable 2.5.2
  - better support for frozen string literals
  - fixed bug w/ uppercase characters in scheme
  - IDNA errors w/ emoji URLs
  - compatibility w/ public_suffix 3.x

- updated to version 2.5.1
  see installed CHANGELOG.md
  [#] Addressable 2.5.1
  - allow unicode normalization to be disabled for URI Template expansion
  - removed duplicate test

- updated to version 2.5.0
  see installed CHANGELOG.md
  [#] Addressable 2.5.0
  - dropping support for Ruby 1.9
  - adding support for Ruby 2.4 preview
  - add support for public suffixes and tld; first runtime dependency
  - hostname escaping should match RFC; underscores in hostnames no longer escaped
  - paths beginning with // and missing an authority are now considered invalid
  - validation now also takes place after setting a path
  - handle backslashes in authority more like a browser for `heuristic_parse`
  - unescaped backslashes in host now raise an `InvalidURIError`
  - `merge!`, `join!`, `omit!` and `normalize!` don't disable deferred validation
  - `heuristic_parse` now trims whitespace before parsing
  - host parts longer than 63 bytes will be ignored and not passed to libidn
  - normalized values always encoded as UTF-8

- updated to version 2.4.0
  see installed CHANGELOG.md
  [#] Addressable 2.4.0
  - support for 1.8.x dropped
  - double quotes in a host now raises an error
  - newlines in host will no longer get unescaped during normalization
  - stricter handling of bogus scheme values
  - stricter handling of encoded port values
  - calling `require 'addressable'` will now load both the URI and Template files
  - assigning to the `hostname` component with an `IPAddr` object is now supported
  - assigning to the `origin` component is now supported
  - fixed minor bug where an exception would be thrown for a missing ACE suffix
  - better partial expansion of URI templates

- updated to version 2.3.8
  - fix warnings
  - update dependency gems
  - support for 1.8.x officially deprecated

- updated to version 2.3.7
  - fix scenario in which invalid URIs don't get an exception until inspected
  - handle hostnames with two adjacent periods correctly
  - upgrade of RSpec

- adapt to new rubygem packaging

- adapt to new rubygem packaging style

- updated to version 2.3.6
  - normalization drops empty query string
  - better handling in template extract for missing values
  - template modifier for `'?'` now treated as optional
  - fixed issue where character class parameters were modified
  - templates can now be tested for equality
  - added `:sorted` option to normalization of query strings
  - fixed issue with normalization of hosts given in `'example.com.'` form

- updated to version 2.3.5
  - added Addressable::URI#empty? method
  - Addressable::URI#hostname methods now strip square brackets from IPv6 hosts
  - compatibility with Net::HTTP in Ruby 2.0.0
  - Addressable::URI#route_from should always give relative URIs

- updated to version 2.3.4
  - fixed issue with encoding altering its inputs
  - query string normalization now leaves ';' characters alone
  - FakeFS is detected before attempting to load unicode tables
  - additional testing to ensure frozen objects don't cause problems

- updated to version 2.3.3
  - fixed issue with converting common primitives during template expansion
  - fixed port encoding issue
  - removed a few warnings
  - normalize should now ignore %2B in query strings
  - the IDNA logic should now be handled by libidn in Ruby 1.9
  - no template match should now result in nil instead of an empty MatchData
  - added license information to gemspec

- update to 2.3.2
  - added Addressable::URI#default_port method
  - fixed issue with Marshalling Unicode data on Windows
  - improved heuristic parsing to better handle IPv4 addresses

- update to 2.3.1
  - fixed missing unicode data file
  - updated Addressable::Template to use RFC 6570, level 4
  - fixed compatibility problems with some versions of Ruby
  - moved unicode tables into a data file for performance reasons
  - removing support for multiple query value notations

- license update: Apache-2.0
  Look at the LICENSE.txt file

- update to 2.2.8
  - fixed issues with dot segment removal code
  - form encoding can now handle multiple values per key
  - updated development environment

- Update to 2.2.6

- fix license - MIT

- update to 2.2.4 (for dm-core 1.1.0)
- split off -doc subpackage

- update to version 2.2.2
  * added site methods
  * improved documentation
  * added support for application/x-www-form-urlencoded.
  * fixed issue with percent escaping of '+' character in query strings

- update to version 2.1.2
  * added HTTP request URI methods
  * better handling of Windows file paths
  * validation_deferred boolean replaced with defer_validation block
  * normalization of percent-encoded paths should now be correct
  * fixed issue with constructing URIs with relative paths
  * fixed warnings
- drop development requirements
- use rubygems_requires macro

- update to 2.1.1

- downgrade to 2.0.2 until dm-core 0.10.0 gets released