Class: RCrewAI::Tools::PdfProcessor

Inherits:
Base
  • Object
show all
Defined in:
lib/rcrewai/tools/pdf_processor.rb

Constant Summary

Constants included from RCrewAI::ToolSchema

RCrewAI::ToolSchema::TYPE_MAP

Instance Method Summary collapse

Methods inherited from Base

available_tools, create_tool, #description, #execute_with_validation, #json_schema, list_available_tools, #name, #validate_params!

Methods included from RCrewAI::ToolSchema

#description, extended, #json_schema, #param, #params, #tool_name

Constructor Details

#initialize(**options) ⇒ PdfProcessor

Returns a new instance of PdfProcessor.



22
23
24
25
26
27
28
 
# File 'lib/rcrewai/tools/pdf_processor.rb', line 22

def initialize(**options)
  super()
  @max_file_size = options.fetch(:max_file_size, 50_000_000) # 50MB
  @max_pages = options.fetch(:max_pages, 100)
  @extract_metadata = options.fetch(:extract_metadata, true)
  @working_directory = options[:working_directory] || Dir.pwd
end

Instance Method Details

#execute(**params) ⇒ Object 



30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
 
# File 'lib/rcrewai/tools/pdf_processor.rb', line 30

def execute(**params)
  validate_params!(
    params,
    required: [:file_path],
    optional: %i[pages extract_text extract_metadata output_format]
  )

  file_path = params[:file_path]
  pages = params[:pages] # Can be array [1,2,3] or range "1-5" or "all"
  extract_text = params.fetch(:extract_text, true)
  extract_metadata = params.fetch(:extract_metadata, @extract_metadata)
  output_format = params.fetch(:output_format, 'text') # 'text', 'json', 'markdown'

  begin
    validate_pdf_file!(file_path)
    result = process_pdf(file_path, pages, extract_text, extract_metadata)
    format_pdf_result(result, output_format)
  rescue StandardError => e
    "PDF processing failed: #{e.message}"
  end
end