Module: Bibliothecary::Analyser::Analysis
- Defined in:
- lib/bibliothecary/analyser/analysis.rb
Instance Method Summary collapse
-
#analyse(folder_path, file_list, options: {}) ⇒ Object
(also: #analyze)
Convenience method to create FileInfo objects from folder path and file list.
- #analyse_contents(filename, contents, options: {}) ⇒ Object (also: #analyze_contents)
-
#analyse_contents_from_info(info, options: {}) ⇒ Object
(also: #analyze_contents_from_info)
This is the place a parsing operation will eventually end up.
-
#analyse_file_info(file_info_list, options: {}) ⇒ Object
(also: #analyze_file_info)
Analyze a set of FileInfo objects and extract manifests from them all.
-
#parse_file(filename, contents, options: {}) ⇒ Object
Call the matching parse class method for this file with these contents.
Instance Method Details
#analyse(folder_path, file_list, options: {}) ⇒ Object Also known as: analyze
Convenience method to create FileInfo objects from folder path and file list.
12 13 14 |
# File 'lib/bibliothecary/analyser/analysis.rb', line 12 def analyse(folder_path, file_list, options: {}) analyse_file_info(file_list.map { |full_path| FileInfo.new(folder_path, full_path) }, options: ) end |
#analyse_contents(filename, contents, options: {}) ⇒ Object Also known as: analyze_contents
31 32 33 |
# File 'lib/bibliothecary/analyser/analysis.rb', line 31 def analyse_contents(filename, contents, options: {}) analyse_contents_from_info(FileInfo.new(nil, filename, contents), options: ) end |
#analyse_contents_from_info(info, options: {}) ⇒ Object Also known as: analyze_contents_from_info
This is the place a parsing operation will eventually end up.
39 40 41 42 43 44 45 46 47 48 49 |
# File 'lib/bibliothecary/analyser/analysis.rb', line 39 def analyse_contents_from_info(info, options: {}) # If your Parser needs to return multiple responses for one file, please override this method # For example see conda.rb kind = determine_kind_from_info(info) parser_result = parse_file(info.relative_path, info.contents, options: ) parser_result = ParserResult.new(dependencies: []) if parser_result.nil? # work around any legacy parsers that return nil Bibliothecary::Analyser.create_analysis(platform_name, info.relative_path, kind, parser_result) rescue Bibliothecary::FileParsingError => e Bibliothecary::Analyser.create_error_analysis(platform_name, info.relative_path, kind, e., e.location) end |
#analyse_file_info(file_info_list, options: {}) ⇒ Object Also known as: analyze_file_info
Analyze a set of FileInfo objects and extract manifests from them all.
20 21 22 23 24 25 26 27 28 |
# File 'lib/bibliothecary/analyser/analysis.rb', line 20 def analyse_file_info(file_info_list, options: {}) matching_info = file_info_list .select(&method(:match_info?)) matching_info.flat_map do |info| analyse_contents_from_info(info, options: ) .merge(related_paths: (info, matching_info)) end end |
#parse_file(filename, contents, options: {}) ⇒ Object
Call the matching parse class method for this file with these contents
54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 |
# File 'lib/bibliothecary/analyser/analysis.rb', line 54 def parse_file(filename, contents, options: {}) details = first_matching_mapping_details(FileInfo.new(nil, filename, contents)) # this can be raised if we don't check match?/match_info?, # OR don't have the file contents when we check them, so # it turns out for example that a .xml file isn't a # manifest after all. raise Bibliothecary::FileParsingError.new("No parser for this file type", filename) unless details[:parser] # The `parser` method should raise an exception if the file is malformed, # should return empty [] if the file is fine but simply doesn't contain # any dependencies, and should never return nil. At the time of writing # this comment, some of the parsers return [] or nil to mean an error # which is confusing to users. send(details[:parser], contents, options: .merge(filename: filename)) rescue Exception => e # default is StandardError but C bindings throw Exceptions # rubocop:disable Lint/RescueException # the C xml parser also puts a newline at the end of the message location = e.backtrace_locations[0] .to_s .then { |l| l =~ /bibliothecary\// ? l.split("bibliothecary/").last : l.split("gems/").last } raise Bibliothecary::FileParsingError.new(e..strip, filename, location) end |