aboutsummaryrefslogtreecommitdiffstats
path: root/Library/Homebrew/rubocops/urls_cop.rb
blob: 4ef801689d102387eda86906ca43ed88eef91136 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
require_relative "./extend/formula_cop"

module RuboCop
  module Cop
    module FormulaAudit
      # This cop audits urls and mirrors in Formulae
      class Urls < FormulaCop
        def audit_formula(_node, _class_node, _parent_class_node, body_node)
          urls = find_every_method_call_by_name(body_node, :url)
          mirrors = find_every_method_call_by_name(body_node, :mirror)

          # GNU urls; doesn't apply to mirrors
          gnu_pattern = %r{^(?:https?|ftp)://ftpmirror.gnu.org/(.*)}
          audit_urls(urls, gnu_pattern) do |match, url|
            problem "Please use \"https://ftp.gnu.org/gnu/#{match[1]}\" instead of #{url}."
          end

          # Fossies upstream requests they aren't used as primary URLs
          # https://github.com/Homebrew/homebrew-core/issues/14486#issuecomment-307753234
          fossies_pattern = %r{^https?://fossies\.org/}
          audit_urls(urls, fossies_pattern) do
            problem "Please don't use fossies.org in the url (using as a mirror is fine)"
          end

          audit_urls(mirrors, /.*/) do |_, mirror|
            urls.each do |url|
              url_string = string_content(parameters(url).first)
              next unless url_string.eql?(mirror)
              problem "URL should not be duplicated as a mirror: #{url_string}"
            end
          end

          urls += mirrors

          # Check a variety of SSL/TLS URLs that don't consistently auto-redirect
          # or are overly common errors that need to be reduced & fixed over time.
          http_to_https_patterns = Regexp.union([%r{^http://ftp\.gnu\.org/},
                                                 %r{^http://ftpmirror\.gnu\.org/},
                                                 %r{^http://download\.savannah\.gnu\.org/},
                                                 %r{^http://download-mirror\.savannah\.gnu\.org/},
                                                 %r{^http://[^/]*\.apache\.org/},
                                                 %r{^http://code\.google\.com/},
                                                 %r{^http://fossies\.org/},
                                                 %r{^http://mirrors\.kernel\.org/},
                                                 %r{^http://(?:[^/]*\.)?bintray\.com/},
                                                 %r{^http://tools\.ietf\.org/},
                                                 %r{^http://launchpad\.net/},
                                                 %r{^http://github\.com/},
                                                 %r{^http://bitbucket\.org/},
                                                 %r{^http://anonscm\.debian\.org/},
                                                 %r{^http://cpan\.metacpan\.org/},
                                                 %r{^http://hackage\.haskell\.org/},
                                                 %r{^http://(?:[^/]*\.)?archive\.org},
                                                 %r{^http://(?:[^/]*\.)?freedesktop\.org},
                                                 %r{^http://(?:[^/]*\.)?mirrorservice\.org/}])
          audit_urls(urls, http_to_https_patterns) do |_, url|
            problem "Please use https:// for #{url}"
          end

          cpan_pattern = %r{^http://search\.mcpan\.org/CPAN/(.*)}i
          audit_urls(urls, cpan_pattern) do |match, url|
            problem "#{url} should be `https://cpan.metacpan.org/#{match[1]}`"
          end

          gnome_pattern = %r{^(http|ftp)://ftp\.gnome\.org/pub/gnome/(.*)}i
          audit_urls(urls, gnome_pattern) do |match, url|
            problem "#{url} should be `https://download.gnome.org/#{match[2]}`"
          end

          debian_pattern = %r{^git://anonscm\.debian\.org/users/(.*)}i
          audit_urls(urls, debian_pattern) do |match, url|
            problem "#{url} should be `https://anonscm.debian.org/git/users/#{match[1]}`"
          end

          # Prefer HTTP/S when possible over FTP protocol due to possible firewalls.
          mirror_service_pattern = %r{^ftp://ftp\.mirrorservice\.org}
          audit_urls(urls, mirror_service_pattern) do |_, url|
            problem "Please use https:// for #{url}"
          end

          cpan_ftp_pattern = %r{^ftp://ftp\.cpan\.org/pub/CPAN(.*)}i
          audit_urls(urls, cpan_ftp_pattern) do |match_obj, url|
            problem "#{url} should be `http://search.cpan.org/CPAN#{match_obj[1]}`"
          end

          # SourceForge url patterns
          sourceforge_patterns = %r{^https?://.*\b(sourceforge|sf)\.(com|net)}
          audit_urls(urls, sourceforge_patterns) do |_, url|
            # Skip if the URL looks like a SVN repo
            next if url.include? "/svnroot/"
            next if url.include? "svn.sourceforge"
            next if url.include? "/p/"

            if url =~ /(\?|&)use_mirror=/
              problem "Don't use #{Regexp.last_match(1)}use_mirror in SourceForge urls (url is #{url})."
            end

            if url.end_with?("/download")
              problem "Don't use /download in SourceForge urls (url is #{url})."
            end

            if url =~ %r{^https?://sourceforge\.}
              problem "Use https://downloads.sourceforge.net to get geolocation (url is #{url})."
            end

            if url =~ %r{^https?://prdownloads\.}
              problem <<~EOS.chomp
                Don't use prdownloads in SourceForge urls (url is #{url}).
                        See: http://librelist.com/browser/homebrew/2011/1/12/prdownloads-is-bad/
              EOS
            end

            if url =~ %r{^http://\w+\.dl\.}
              problem "Don't use specific dl mirrors in SourceForge urls (url is #{url})."
            end

            problem "Please use https:// for #{url}" if url.start_with? "http://downloads"
          end

          # Debian has an abundance of secure mirrors. Let's not pluck the insecure
          # one out of the grab bag.
          unsecure_deb_pattern = %r{^http://http\.debian\.net/debian/(.*)}i
          audit_urls(urls, unsecure_deb_pattern) do |match, _|
            problem <<~EOS
              Please use a secure mirror for Debian URLs.
              We recommend:
                https://mirrors.ocf.berkeley.edu/debian/#{match[1]}
            EOS
          end

          # Check for new-url Google Code download urls, https:// is preferred
          google_code_pattern = Regexp.union([%r{^http://.*\.googlecode\.com/files.*},
                                              %r{^http://code\.google\.com/}])
          audit_urls(urls, google_code_pattern) do |_, url|
            problem "Please use https:// for #{url}"
          end

          # Check for git:// GitHub repo urls, https:// is preferred.
          git_gh_pattern = %r{^git://[^/]*github\.com/}
          audit_urls(urls, git_gh_pattern) do |_, url|
            problem "Please use https:// for #{url}"
          end

          # Check for git:// Gitorious repo urls, https:// is preferred.
          git_gitorious_pattern = %r{^git://[^/]*gitorious\.org/}
          audit_urls(urls, git_gitorious_pattern) do |_, url|
            problem "Please use https:// for #{url}"
          end

          # Check for http:// GitHub repo urls, https:// is preferred.
          gh_pattern = %r{^http://github\.com/.*\.git$}
          audit_urls(urls, gh_pattern) do |_, url|
            problem "Please use https:// for #{url}"
          end

          # Check for master branch GitHub archives.
          tarball_gh_pattern = %r{^https://github\.com/.*archive/master\.(tar\.gz|zip)$}
          audit_urls(urls, tarball_gh_pattern) do
            problem "Use versioned rather than branch tarballs for stable checksums."
          end

          # Use new-style archive downloads
          archive_gh_pattern = %r{https://.*github.*/(?:tar|zip)ball/}
          audit_urls(urls, archive_gh_pattern) do |_, url|
            next unless url !~ /\.git$/
            problem "Use /archive/ URLs for GitHub tarballs (url is #{url})."
          end

          # Don't use GitHub .zip files
          zip_gh_pattern = %r{https://.*github.*/(archive|releases)/.*\.zip$}
          audit_urls(urls, zip_gh_pattern) do |_, url|
            next unless url !~ %r{releases/download}
            problem "Use GitHub tarballs rather than zipballs (url is #{url})."
          end

          # Don't use GitHub codeload URLs
          codeload_gh_pattern = %r{https?://codeload\.github\.com/(.+)/(.+)/(?:tar\.gz|zip)/(.+)}
          audit_urls(urls, codeload_gh_pattern) do |match, url|
            problem <<~EOS
              Use GitHub archive URLs:
                https://github.com/#{match[1]}/#{match[2]}/archive/#{match[3]}.tar.gz
              Rather than codeload:
                #{url}
            EOS
          end

          # Check for Maven Central urls, prefer HTTPS redirector over specific host
          maven_pattern = %r{https?://(?:central|repo\d+)\.maven\.org/maven2/(.+)$}
          audit_urls(urls, maven_pattern) do |match, url|
            problem "#{url} should be `https://search.maven.org/remotecontent?filepath=#{match[1]}`"
          end
        end
      end
    end
    module FormulaAuditStrict
      class PyPiUrls < FormulaCop
        def audit_formula(_node, _class_node, _parent_class_node, body_node)
          urls = find_every_method_call_by_name(body_node, :url)
          mirrors = find_every_method_call_by_name(body_node, :mirror)
          urls += mirrors

          # Check pypi urls
          @pypi_pattern = %r{^https?://pypi.python.org/(.*)}
          audit_urls(urls, @pypi_pattern) do |match, url|
            problem "#{url} should be `https://files.pythonhosted.org/#{match[1]}`"
          end
        end

        def autocorrect(node)
          lambda do |corrector|
            url_string_node = parameters(node).first
            url = string_content(url_string_node)
            match = regex_match_group(url_string_node, @pypi_pattern)
            correction = node.source.sub(url, "https://files.pythonhosted.org/#{match[1]}")
            corrector.insert_before(node.source_range, correction)
            corrector.remove(node.source_range)
          end
        end
      end
    end
  end
end