Module: Legion::Extensions::Knowledge::Helpers::Parser

Defined in:
lib/legion/extensions/knowledge/helpers/parser.rb

Class Method Summary collapse

Class Method Details

.extract_via_data(file_path:) ⇒ Object 



51
52
53
54
55
56
57
58
59
60
61
 
# File 'lib/legion/extensions/knowledge/helpers/parser.rb', line 51

def extract_via_data(file_path:)
  return [{ error: 'unsupported format', source_file: file_path }] unless defined?(::Legion::Data::Extract)

  result = ::Legion::Data::Extract.extract(file_path, type: :auto)
  return [{ error: 'extraction_failed', source_file: file_path, detail: result }] unless result.is_a?(Hash) && result[:text]

  heading = ::File.basename(file_path, '.*')
  [{ heading: heading, section_path: [], content: result[:text].strip, source_file: file_path }]
rescue StandardError => e
  [{ error: 'extraction_failed', source_file: file_path, detail: e.message }]
end

.parse(file_path:) ⇒ Object 



10
11
12
13
14
15
16
17
18
19
20
21
22
23
 
# File 'lib/legion/extensions/knowledge/helpers/parser.rb', line 10

def parse(file_path:)
  ext = ::File.extname(file_path).downcase

  case ext
  when '.md'
    parse_markdown(file_path: file_path)
  when '.txt'
    parse_text(file_path: file_path)
  when '.pdf', '.docx'
    extract_via_data(file_path: file_path)
  else
    [{ error: 'unsupported format', source_file: file_path }]
  end
end

.parse_markdown(file_path:) ⇒ Object 



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
 
# File 'lib/legion/extensions/knowledge/helpers/parser.rb', line 25

def parse_markdown(file_path:)
  content = ::File.read(file_path, encoding: 'utf-8')
  sections        = []
  current_heading = ::File.basename(file_path, '.*')
  current_lines   = []
  heading_stack   = {}

  content.each_line do |line|
    level = heading_level(line)
    if level
      flush_section(sections, current_heading, build_section_path(heading_stack), current_lines, file_path)
      title = line.sub(/^#+\s*/, '').chomp
      heading_stack.delete_if { |d, _| d >= level }
      heading_stack[level] = title
      current_heading = title
      current_lines   = []
    else
      current_lines << line
    end
  end

  flush_section(sections, current_heading, build_section_path(heading_stack), current_lines, file_path)

  sections.empty? ? [{ heading: ::File.basename(file_path, '.*'), section_path: [], content: content.strip, source_file: file_path }] : sections
end

.parse_text(file_path:) ⇒ Object 



63
64
65
66
67
68
 
# File 'lib/legion/extensions/knowledge/helpers/parser.rb', line 63

def parse_text(file_path:)
  content = ::File.read(file_path, encoding: 'utf-8')
  heading = ::File.basename(file_path, '.*')

  [{ heading: heading, section_path: [], content: content.strip, source_file: file_path }]
end