-
Notifications
You must be signed in to change notification settings - Fork 202
/
nu_get.rb
242 lines (201 loc) · 8.29 KB
/
nu_get.rb
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
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
# frozen_string_literal: true
require "zip"
module PackageManager
class NuGet < Base
HAS_VERSIONS = true
HAS_DEPENDENCIES = true
BIBLIOTHECARY_SUPPORT = true
URL = "https://www.nuget.org"
COLOR = "#178600"
ENTIRE_PACKAGE_CAN_BE_DEPRECATED = true
def self.package_link(db_project, version = nil)
"https://www.nuget.org/packages/#{db_project.name}/#{version}"
end
def self.download_url_by_name(name, version = nil)
"https://www.nuget.org/api/v2/package/#{escaped_name(name)}/#{version}"
end
def self.download_url(db_project, version = nil)
download_url_by_name(db_project.name, version)
end
def self.install_instructions(db_project, version = nil)
"Install-Package #{db_project.name}" + (version ? " -Version #{version}" : "")
end
def self.deprecation_info(db_project)
releases = raw_versions(db_project.name)
deprecation = releases.last&.deprecation
if deprecation.present?
{
is_deprecated: true,
message: deprecation.message,
alternate_package: deprecation.alternate_package,
}
else
{
is_deprecated: false,
message: "",
alternate_package: nil,
}
end
end
def self.load_names(limit = nil)
endpoints = name_endpoints
segment_count = limit || (endpoints.length - 1)
endpoints.reverse[0..segment_count].each do |endpoint|
package_ids = get_names(endpoint)
package_ids.each { |id| REDIS.sadd "nuget-names", id }
end
end
def self.recent_names
name_endpoints.reverse[0..2].map { |url| get_names(url) }.flatten.uniq
end
def self.name_endpoints
get("https://api.nuget.org/v3/catalog0/index.json")["items"].map { |i| i["@id"] }
end
def self.get_names(endpoint)
get(endpoint)["items"].map { |i| i["nuget:id"] }
end
def self.project_names
REDIS.smembers "nuget-names"
end
def self.project(name)
canonical_name = fetch_canonical_nuget_name(name)
if name != canonical_name
StructuredLog.capture("CANONICAL_NAME_DIFFERS", { platform: "nuget", name: name, canonical_name: canonical_name })
name = canonical_name if canonical_name
end
h = {
name: name,
}
h[:raw_versions] = raw_versions(name)
h[:versions] = versions(h, name)
return {} unless h[:versions].any?
h
end
def self.package_file(name, version)
url = download_url_by_name(name, version)
StringIO.new(get_raw(url))
end
def self.nuspec(name, version)
nuspec = nil
Zip::File.open_buffer(package_file(name, version)) do |zip|
zip.each do |file|
# the correct nuspec file will be at the root of the tree with the package name in the filename
# package identifiers are case-insensitive: https://learn.microsoft.com/en-us/nuget/reference/nuspec#id
if file.name.tap { |name| name.force_encoding("UTF-8") }.downcase["#{name}.nuspec".downcase]
file.get_input_stream { |io| nuspec = Ox.parse(io.read) }
end
end
end
unless nuspec
Rails.logger.error(
"Malformed NuGet package file: NuGet/#{name}: No matching nuspec file found"
)
end
nuspec
rescue Zip::Error, Ox::ParseError => e
Rails.logger.error(
"Unable to process NuGet package file: NuGet/#{name}: #{e.message}"
)
nil
end
# These are the raw version data we get from the upstream API
def self.raw_versions(name)
SemverRegistrationApiProjectReleasesBuilder.build(project_name: escaped_name(name)).releases
rescue StandardError => e
Rails.logger.error("Unable to retrieve releases for NuGet project #{name}: #{e.message}")
[]
end
def self.mapping(raw_project)
latest_raw_version = raw_project[:raw_versions].last
raw_nuspec = nuspec(raw_project[:name], latest_raw_version.version_number)
nuspec_repo = raw_nuspec&.locate("package/metadata/repository")&.first
nuspec_repo = nuspec_repo["url"] if nuspec_repo
MappingBuilder.build_hash(
name: raw_project[:name],
description: latest_raw_version.description,
homepage: latest_raw_version.project_url,
keywords_array: latest_raw_version.tags,
repository_url: repo_fallback(nuspec_repo, latest_raw_version.project_url),
licenses: latest_raw_version.licenses,
versions: versions(raw_project, raw_project[:name])
)
end
def self.versions(raw_project, _name)
raw_project[:raw_versions].map do |raw_version|
version_data = {
number: raw_version.version_number,
published_at: raw_version.published_at,
original_license: raw_version.original_license,
# This is a one-off VersionBuilder field for NuGet -- see VersionBuilder for more details.
dependencies: raw_version.dependencies,
}
# NuGet releases can be:
#
# * deprecated (with a "deprecation" field), e.g. "[email protected]"
# * unlisted (with published set to "1900-01-01 00:00:00"), e.g. "[email protected]"
# * or both (we're not sure why they wouldn't always both be set)
#
# We treat all three cases as deprecated on our side.
if raw_version.deprecation.present? || raw_version.published_at.year < 1901
# When "unlisted", we want to maintain the original "published" date instead of overwriting it with "1900-01-01 00:00:00".
version_data.delete(:published_at) if raw_version.published_at.year < 1901
version_data[:status] = "Deprecated"
else
# If the version is listed on NuGet and not deprecated, then we'd currently
# consider it "Active", or status: nil.
version_data[:status] = nil
end
VersionBuilder.build_hash(**version_data)
end
end
def self.dependencies(_name, version, mapped_project)
mapped_version = mapped_project[:versions].find { |v| v[:number] == version }
mapped_version[:dependencies]
.map do |dep|
{
project_name: dep.name,
requirements: dep.requirements.presence || "*",
kind: "runtime",
optional: false,
platform: name.demodulize,
}
end
end
class ParseCanonicalNameFailedError < StandardError; end
# Escape name because NuGet packages can contain non-ASCII characters, e.g. Felsökning
def self.escaped_name(name)
URI::Parser.new.escape(name)
end
# Unescape name when we want to store the actual bytes instead of URI escaped bytes, e.g. Fels%C3%B6kning
def self.unescaped_name(name)
URI::Parser.new.unescape(name)
end
# Returns canonical casing for case-insensitive NuGet package names
# @param name [String] A given project name to check
# @return [String] If successfully found, the canonical form of the given name
# @return [nil] The scrape request was unsuccessful
# @return [false] The scrape succeeded, but we didn't detect a name
def self.fetch_canonical_nuget_name(name)
base_url = "https://nuget.org/packages/"
page = get_html("#{base_url}#{escaped_name(name)}")
og_url_element = page.css("meta[property='og:url']").first
if page.text.empty? # Request failed, likely temporarily
StructuredLog.capture("FETCH_CANONICAL_NAME_FAILED", { platform: "nuget", name: name })
return nil
elsif og_url_element.nil?
# If we got a response, but don't find this element, it most likely means
# the project was removed upstream.
StructuredLog.capture("CANONICAL_NAME_ELEMENT_MISSING", { platform: "nuget", name: name })
return false
end
og_url = og_url_element.attributes["content"]&.text || ""
canonical_name = og_url.sub(base_url, "").sub(/\/$/, "")
# If we got as far as to grab values but they don't meet our assumptions, this should be
# exceptional. It likely means we need to update this method.
if !og_url.start_with?(base_url) || canonical_name.blank?
raise ParseCanonicalNameFailedError, "Could not parse a canonical name for `#{name}`. Did upstream change their markup structure?"
end
unescaped_name(canonical_name)
end
end
end