librariesio-url-parser 1.0.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: 4095a2003d55e594f33b981cb4fa104b37ab2ffba7f6f3988b905a901afb9785
+  data.tar.gz: 8dfbab7448f15692cd1064eb119cd55101d5fceab25dd806930780c48f3a65cd
+SHA512:
+  metadata.gz: 5aa7f74fcbaa9d1dc7774d61eedcbf349028a4e3bef26dc4a9e3b07e3856cdc2559feff970a36becf6e2c4e10eb4be77b68969470f814cd422a47e4f800e940e
+  data.tar.gz: f3dfc1047f3141e6a4f9ea39f7ac7514bbb607bd5f2577ac0809903614b3e008d7bb48916a01cb537886f5be0bad4fb793b9130e43d212d85fc0061bf1fb8161

data/.circleci/config.yml

@@ -0,0 +1,22 @@
+version: 2.1
+orbs:
+  ruby: circleci/ruby@1.8.0
+
+jobs:
+  test:
+    docker:
+      - image: cimg/ruby:2.6.5
+    executor: ruby/default
+    steps:
+    - checkout
+    - run:
+        name: Which bundler?
+        command: bundle -v
+    - ruby/install-deps
+    - ruby/rspec-test
+
+workflows:
+  version: 2.1
+  test:
+    jobs:
+      - test

data/.ruby-version

@@ -0,0 +1 @@
+2.6.5

data/Gemfile

@@ -0,0 +1,5 @@
+source "https://rubygems.org"
+ruby "2.6.5"
+
+# Specify your gem's dependencies in librariesio-url-parser.gemspec
+gemspec

data/Gemfile.lock

@@ -0,0 +1,40 @@
+PATH
+  remote: .
+  specs:
+    librariesio-url-parser (1.0.0)
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    diff-lcs (1.5.0)
+    rake (12.3.3)
+    rspec (3.11.0)
+      rspec-core (~> 3.11.0)
+      rspec-expectations (~> 3.11.0)
+      rspec-mocks (~> 3.11.0)
+    rspec-core (3.11.0)
+      rspec-support (~> 3.11.0)
+    rspec-expectations (3.11.0)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.11.0)
+    rspec-mocks (3.11.1)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.11.0)
+    rspec-support (3.11.0)
+    rspec_junit_formatter (0.5.1)
+      rspec-core (>= 2, < 4, != 2.12.0)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  librariesio-url-parser!
+  rake (~> 12.0)
+  rspec (~> 3.0)
+  rspec_junit_formatter (~> 0.5)
+
+RUBY VERSION
+   ruby 2.6.5p114
+
+BUNDLED WITH
+   2.1.4

data/README.md

@@ -0,0 +1,56 @@
+# Libraries.io URL Parser
+
+Repository URL parsing library for https://libraries.io
+
+## Usage
+
+Parse a org/repo string from an input string:
+
+```ruby
+GithubURLParser.parse("https://github.com/rails/rails/") #=> rails/rails
+GithubURLParser.parse("git@github.com:rails/rails.git") #=> rails/rails
+GithubURLParser.parse("https://github.com/rails/rails.git") #=> rails/rails
+GithubURLParser.parse("https://github.com") #=> nil
+```
+
+Parse a full url from an input string:
+
+```ruby
+GithubURLParser.parse_to_full_url("https://github.com/rails/rails/") #=> https://github.com/rails/rails
+GithubURLParser.parse_to_full_url("git@github.com:rails/rails.git") #=> https://github.com/rails/rails
+GithubURLParser.parse_to_full_url("https://github.com/rails/rails.git") #=> https://github.com/rails/rails
+GithubURLParser.parse_to_full_url("https://github.com") #=> nil
+```
+
+Parse an org/user url from an input string:
+
+```ruby
+GithubURLParser.parse_to_full_user_url("https://github.com/rails/rails/") #=> nil
+GithubURLParser.parse_to_full_user_url("git@github.com:rails/rails.git") #=> nil
+GithubURLParser.parse_to_full_user_url("https://github.com/rails/rails.git") #=> nil
+GithubURLParser.parse_to_full_user_url("https://github.com") #=> nil
+GithubURLParser.parse_to_full_user_url("https://github.com/rails") #=> https://github.com/rails
+```
+
+Parse a full url from an input string where the repository is unknown
+
+```ruby
+URLParser.try_all("git@github.com:rails/rails.git") #=> https://github.com/rails/rails
+URLParser.try_all("git@gitlab.com:inkscape/inkscape.git") #=> "https://gitlab.com/inkscape/inkscape"
+URLParser.try_all("git@bitbucket.org:tildeslash/monit.git") #=> "https://bitbucket.org/tildeslash/monit"
+```
+
+## Supported repositories
+
+- GitHub
+- GitLab
+- Bitbucket
+
+## Development
+
+After checking out the repo, run `bundle install` to install dependencies. Then, run `rake spec` to run the tests.
+
+## Contributing
+
+Bug reports and pull requests are welcome on GitHub at https://github.com/librariesio/librariesio-url-parser. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [Contributor Covenant](http://contributor-covenant.org) code of conduct.
+

data/Rakefile

@@ -0,0 +1,6 @@
+require "bundler/gem_tasks"
+require "rspec/core/rake_task"
+
+RSpec::Core::RakeTask.new(:spec)
+
+task :default => :spec

data/lib/bitbucket_url_parser.rb

@@ -0,0 +1,20 @@
+# frozen_string_literal: true
+class BitbucketURLParser < URLParser
+  private
+
+  def full_domain
+    'https://bitbucket.org'
+  end
+
+  def tlds
+    %w(com org)
+  end
+
+  def domain
+    'bitbucket'
+  end
+
+  def remove_domain
+    url.gsub!(/(bitbucket.com|bitbucket.org)+?(:|\/)?/i, '')
+  end
+end

data/lib/github_url_parser.rb

@@ -0,0 +1,20 @@
+# frozen_string_literal: true
+class GithubURLParser < URLParser
+  private
+
+  def full_domain
+    'https://github.com'
+  end
+
+  def tlds
+    %w(io com org)
+  end
+
+  def domain
+    'github'
+  end
+
+  def remove_domain
+    url.gsub!(/(github.io|github.com|github.org|raw.githubusercontent.com)+?(:|\/)?/i, '')
+  end
+end

data/lib/gitlab_url_parser.rb

@@ -0,0 +1,20 @@
+# frozen_string_literal: true
+class GitlabURLParser < URLParser
+  private
+
+  def full_domain
+    'https://gitlab.com'
+  end
+
+  def tlds
+    %w(com)
+  end
+
+  def domain
+    'gitlab'
+  end
+
+  def remove_domain
+    url.gsub!(/(gitlab.com)+?(:|\/)?/i, '')
+  end
+end

data/lib/librariesio-url-parser.rb

@@ -0,0 +1,10 @@
+# frozen_string_literal: true
+
+require_relative "url_parser"
+require_relative "bitbucket_url_parser"
+require_relative "github_url_parser"
+require_relative "gitlab_url_parser"
+
+module LibrariesioURLParser
+  VERSION = "1.0.0"
+end

data/lib/url_parser.rb

@@ -0,0 +1,164 @@
+# frozen_string_literal: true
+class URLParser
+  def self.parse(url)
+    new(url).parse
+  end
+
+  def initialize(url)
+    @url = url.to_s.dup
+  end
+
+  def parse
+    return nil unless parseable?
+
+    if url = extractable_early?
+      url
+    else
+      clean_url
+      format_url
+    end
+  end
+
+  def self.parse_to_full_url(url)
+    new(url).parse_to_full_url
+  end
+
+  def self.parse_to_full_user_url(url)
+    new(url).parse_to_full_user_url
+  end
+
+  def self.try_all(url)
+    GithubURLParser.parse_to_full_url(url) ||
+    GitlabURLParser.parse_to_full_url(url) ||
+    BitbucketURLParser.parse_to_full_url(url)
+  end
+
+  def parse_to_full_url
+    path = parse
+    return nil unless path.present?
+    [full_domain, path].join('/')
+  end
+
+  def parse_to_full_user_url
+    return nil unless parseable?
+    path = clean_url
+    return nil unless path.length == 1
+    [full_domain, path].join('/')
+  end
+
+  private
+
+  attr_accessor :url
+
+  def clean_url
+    remove_whitespace
+    remove_brackets
+    remove_anchors
+    remove_querystring
+    remove_auth_user
+    remove_equals_sign
+    remove_scheme
+    return nil unless includes_domain?
+    remove_subdomain
+    remove_domain
+    remove_git_extension
+    remove_git_scheme
+    remove_extra_segments
+  end
+
+  def format_url
+    return nil unless url.length == 2
+    url.join('/')
+  end
+
+  def parseable?
+    !url.nil? && url.include?(domain)
+  end
+
+  def tlds
+    raise NotImplementedError
+  end
+
+  def domain
+    raise NotImplementedError
+  end
+
+  def includes_domain?
+    raise NotImplementedError
+  end
+
+  def extractable_early?
+    raise NotImplementedError
+  end
+
+  def domain_regex
+    "#{domain}\.(#{tlds.join('|')})"
+  end
+
+  def website_url?
+    url.match(/www\.#{domain_regex}/i)
+  end
+
+  def includes_domain?
+    url.match(/#{domain_regex}/i)
+  end
+
+  def extractable_early?
+    return false if website_url?
+
+    match = url.match(/([\w\.@\:\-_~]+)\.#{domain_regex}\/([\w\.@\:\-\_\~]+)/i)
+    if match && match.length == 4
+      return "#{match[1]}/#{match[3]}"
+    end
+
+    nil
+  end
+
+  def remove_anchors
+    url.gsub!(/(#\S*)$/i, '')
+  end
+
+  def remove_auth_user
+    self.url = url.split('@')[-1]
+  end
+
+  def remove_domain
+    raise NotImplementedError
+  end
+
+  def remove_brackets
+    url.gsub!(/>|<|\(|\)|\[|\]/, '')
+  end
+
+  def remove_equals_sign
+    self.url = url.split('=')[-1]
+  end
+
+  def remove_extra_segments
+    self.url = url.split('/').reject{ |s| s.strip.empty? }[0..1]
+  end
+
+  def remove_git_extension
+    url.gsub!(/(\.git|\/)$/i, '')
+  end
+
+  def remove_git_scheme
+    url.gsub!(/git\/\//i, '')
+  end
+
+  def remove_querystring
+    url.gsub!(/(\?\S*)$/i, '')
+  end
+
+  def remove_scheme
+    url.gsub!(/(((git\+https|git|ssh|hg|svn|scm|http|https)+?:)+?)/i, '')
+  end
+
+  def remove_subdomain
+    url.gsub!(/(www|ssh|raw|git|wiki)+?\./i, '')
+  end
+
+  def remove_whitespace
+    url.gsub!(/\s/, '')
+  end
+end

data/librariesio-url-parser.gemspec

@@ -0,0 +1,23 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require "librariesio-url-parser"
+
+Gem::Specification.new do |spec|
+  spec.name          = "librariesio-url-parser"
+  spec.version       = LibrariesioURLParser::VERSION
+  spec.authors       = ["Matt Pace"]
+  spec.email         = ["matt.pace@tidelift.com"]
+
+  spec.summary       = "Parse the URL for various repositories tracked by libraries.io"
+  spec.homepage      = "https://github.com/librariesio/librariesio-url-parser"
+  spec.license       = "AGPL-3.0"
+  spec.metadata = { "rubygems_mfa_required" => "true" }
+
+  spec.files         = `git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|spec|features)/}) }
+  spec.require_paths = ["lib"]
+
+  spec.add_development_dependency "rake", "~> 12.0"
+  spec.add_development_dependency "rspec", "~> 3.0"
+  spec.add_development_dependency "rspec_junit_formatter", "~> 0.5"
+end

metadata

@@ -0,0 +1,98 @@
+--- !ruby/object:Gem::Specification
+name: librariesio-url-parser
+version: !ruby/object:Gem::Version
+  version: 1.0.0
+platform: ruby
+authors:
+- Matt Pace
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2022-06-13 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '12.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '12.0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+- !ruby/object:Gem::Dependency
+  name: rspec_junit_formatter
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.5'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.5'
+description:
+email:
+- matt.pace@tidelift.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".circleci/config.yml"
+- ".ruby-version"
+- Gemfile
+- Gemfile.lock
+- README.md
+- Rakefile
+- lib/bitbucket_url_parser.rb
+- lib/github_url_parser.rb
+- lib/gitlab_url_parser.rb
+- lib/librariesio-url-parser.rb
+- lib/url_parser.rb
+- librariesio-url-parser.gemspec
+homepage: https://github.com/librariesio/librariesio-url-parser
+licenses:
+- AGPL-3.0
+metadata:
+  rubygems_mfa_required: 'true'
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.0.9
+signing_key:
+specification_version: 4
+summary: Parse the URL for various repositories tracked by libraries.io
+test_files: []