Class: Archsight::Import::Handlers::Repository

Inherits:
Archsight::Import::Handler show all
Defined in:
lib/archsight/import/handlers/repository.rb

Overview

Repository handler - clones/syncs and analyzes a git repository, generates a TechnologyArtifact

Configuration:

import/config/path - Path where the git repository should be cloned
import/config/gitUrl - Git URL to clone from (if not already cloned)
import/config/archived - Optional "true" if repository is archived
import/config/visibility - Optional visibility (internal, public, open-source)
import/config/sccPath - Optional path to scc binary (default: scc)
import/config/fallbackTeam - Optional team name when no contributor match found
import/config/botTeam - Optional team name for bot-only repositories
import/config/corporateAffixes - Optional comma-separated corporate username affixes for team matching (e.g., "ionos,1and1")

Instance Attribute Summary

Attributes inherited from Archsight::Import::Handler

#database, #import_resource, #progress, #resources_dir, #shared_writer

Instance Method Summary collapse

Methods inherited from Archsight::Import::Handler

#compute_config_hash, #config, #config_all, #import_yaml, #initialize, #resource_yaml, #resources_to_yaml, #self_marker, #write_generates_meta, #write_yaml

Constructor Details

This class inherits a constructor from Archsight::Import::Handler

Instance Method Details

#access_denied_error?(message) ⇒ Boolean

Check if error message indicates access denied

Returns:

  • (Boolean) 


163
164
165
166
167
168
169
170
171
172
173
174
175
 
# File 'lib/archsight/import/handlers/repository.rb', line 163

def access_denied_error?(message)
  return false if message.nil?

  patterns = [
    /could not read from remote repository/i,
    /permission denied/i,
    /access denied/i,
    /authentication failed/i,
    /repository not found/i,
    /fatal: '.*' does not appear to be a git repository/i
  ]
  patterns.any? { |p| message.match?(p) }
end

#clone_repository ⇒ Object 



112
113
114
115
 
# File 'lib/archsight/import/handlers/repository.rb', line 112

def clone_repository
  FileUtils.mkdir_p(File.dirname(@path))
  run_git(%w[git clone --quiet] + [@git_url, @path], Dir.pwd)
end

#empty_repository? ⇒ Boolean

Returns:

  • (Boolean) 


133
134
135
136
137
 
# File 'lib/archsight/import/handlers/repository.rb', line 133

def empty_repository?
  # Check if HEAD exists (empty repos have no commits)
  _, _, status = Open3.capture3("git", "rev-parse", "HEAD", chdir: @path)
  !status.success?
end

#execute ⇒ Object 



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
 
# File 'lib/archsight/import/handlers/repository.rb', line 24

def execute
  @path = config("path")
  @git_url = config("gitUrl")
  raise "Missing required config: path" unless @path

  # Clone or update the repository if gitUrl is provided
  if @git_url
    begin
      sync_repository
      if @skip_analysis
        write_generates_meta
        return
      end
    rescue StandardError => e
      # Access denied or other git errors - create minimal artifact
      if access_denied_error?(e.message)
        progress.update("Access denied - creating minimal artifact")
        write_minimal_artifact(
          status: "inaccessible",
          reason: "Repository not accessible",
          error: e.message,
          visibility: "private"
        )
        write_generates_meta
        return
      end
      raise
    end
  end

  raise "Directory not found: #{@path}" unless File.directory?(@path)
  raise "Not a git repository: #{@path}" unless File.directory?(File.join(@path, ".git"))

  # Check if empty repository (no code)
  progress.update("Analyzing code")
  scc_data = run_scc(@path)
  estimated_cost = scc_data["estimatedCost"]
  if !estimated_cost.nil? && estimated_cost.to_f.zero?
    progress.update("No analyzable code - creating minimal artifact")
    write_minimal_artifact(
      status: "no-code",
      reason: "No analyzable source code found"
    )
    write_generates_meta
    return
  end

  # Run native git analytics
  progress.update("Analyzing git history")
  git_data = run_git_analytics(@path)

  # Analyze licenses and match teams in parallel (independent of each other)
  progress.update("Analyzing licenses & matching teams")
  license_data, team_result = run_license_and_team_analysis(@path, scc_data, git_data)

  # Build resource
  progress.update("Generating resource")
  resource = build_technology_artifact(@path, scc_data, git_data, team_result, license_data)

  # Write output with self-marker for caching
  yaml_content = YAML.dump(resource) + YAML.dump(self_marker)
  write_yaml(yaml_content)

  write_generates_meta
end

#run_git(command, dir) ⇒ String

Run a git command safely using array form to prevent shell injection

Parameters:

  • command (Array<String>) —

    Command and arguments as array

  • dir (String) —

    Working directory

Returns:

  • (String) —

    Command output



143
144
145
146
147
148
 
# File 'lib/archsight/import/handlers/repository.rb', line 143

def run_git(command, dir)
  out, err, status = Open3.capture3(*command, chdir: dir)
  raise "Git command failed: #{sanitize_error(err)}" unless status.success?

  out
end

#sanitize_error(message) ⇒ Object

Sanitize error message to prevent breaking TTY progress display



151
152
153
154
155
156
157
158
159
160
 
# File 'lib/archsight/import/handlers/repository.rb', line 151

def sanitize_error(message)
  return "" if message.nil? || message.empty?

  # Take first meaningful line, strip ANSI codes and remote prefixes
  lines = message.lines.map(&:strip).reject { |l| l.empty? || l.start_with?("remote:") }
  first_line = lines.first || message.lines.first&.strip || ""

  # Truncate if too long
  first_line.length > 100 ? "#{first_line[0, 97]}..." : first_line
end

#sync_repository ⇒ Object 



90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
 
# File 'lib/archsight/import/handlers/repository.rb', line 90

def sync_repository
  if File.directory?(File.join(@path, ".git"))
    # Update existing repository
    progress.update("Updating repository")
    update_repository
  else
    # Clone new repository
    progress.update("Cloning repository")
    clone_repository
  end

  # Check if repository is empty (no commits)
  return unless empty_repository?

  progress.update("Empty repository - creating minimal artifact")
  write_minimal_artifact(
    status: "empty",
    reason: "Repository has no commits"
  )
  @skip_analysis = true
end

#update_repository ⇒ Object 



117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
 
# File 'lib/archsight/import/handlers/repository.rb', line 117

def update_repository
  run_git(%w[git fetch --quiet], @path)
  return if empty_repository? # Skip merge for empty repos

  # Check if update is needed
  current_head = run_git(%w[git rev-parse HEAD], @path).strip
  fetch_head = run_git(%w[git rev-parse FETCH_HEAD], @path).strip
  return if current_head == fetch_head # Already up-to-date

  run_git(%w[git merge --ff-only FETCH_HEAD], @path)
rescue StandardError => e
  # If merge fails (diverged history), reset to remote state
  progress.warn("Merge failed: #{e.message}, resetting to remote")
  run_git(%w[git reset --hard FETCH_HEAD], @path)
end

#write_minimal_artifact(status:, reason:, error: nil, visibility: nil) ⇒ Object

Write a minimal TechnologyArtifact for repositories that can’t be fully analyzed

Parameters:

  • status (String) —

    Activity status (inaccessible, empty, no-code)

  • reason (String) —

    Human-readable reason

  • error (String, nil) (defaults to: nil) —

    Optional error message

  • visibility (String) (defaults to: nil) —

    Repository visibility (default: from config or “internal”)



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
 
# File 'lib/archsight/import/handlers/repository.rb', line 182

def write_minimal_artifact(status:, reason:, error: nil, visibility: nil)
  git_url = @git_url
  vis = visibility || config("visibility", default: "internal")

  annotations = {
    "artifact/type" => "repo",
    "repository/git" => git_url,
    "repository/visibility" => vis,
    "activity/status" => status,
    "activity/reason" => reason,
    "generated/script" => import_resource.name,
    "generated/at" => Time.now.utc.iso8601
  }

  annotations["repository/accessible"] = "false" if status == "inaccessible"
  annotations["repository/error"] = sanitize_error(error) if error

  resource = resource_yaml(
    kind: "TechnologyArtifact",
    name: repository_name(git_url),
    annotations: annotations,
    spec: {}
  )

  # Write output with self-marker for caching
  yaml_content = YAML.dump(resource) + YAML.dump(self_marker)
  write_yaml(yaml_content)
end