Class: Ocrb::Extractors::OpenAi

Inherits:
Object
  • Object
show all
Defined in:
lib/ocrb.rb

Instance Method Summary collapse

Constructor Details

#initialize(url: 'http://127.0.0.1:1234/v1', model: 'zai-org/glm-4.6v-flash', api_key: 'asdf', json: false) ⇒ OpenAi

Returns a new instance of OpenAi.



30
31
32
33
34
35
36
37
38
39
40
41
42
# File 'lib/ocrb.rb', line 30

def initialize(
  # url: 'http://localhost:11434/v1',
  # model: 'glm-ocr:bf16',
  url: 'http://127.0.0.1:1234/v1',
  model: 'zai-org/glm-4.6v-flash',
  api_key: 'asdf',
  json: false
)
  @uri = URI(File.join(url, 'chat/completions'))
  @api_key = api_key
  @model = model
  @json = json
end

Instance Method Details

#extract(image_path, prompt) ⇒ Object



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
# File 'lib/ocrb.rb', line 44

def extract(image_path, prompt)
  req_body = {
    model: @model,
    messages: [
      {
        role: 'user',
        content: [
          { type: 'text', text: prompt },
          { type: 'image_url', image_url: { url: image_data_url(image_path) } }
        ]
      }
    ]
  }
  req_body.merge!(response_format)

  request = Net::HTTP::Post.new(@uri)
  request['Authorization'] = "Bearer #{@api_key}"
  request['Content-Type'] = 'application/json'
  request.body = JSON.generate(req_body)

  response = Net::HTTP.start(@uri.host, @uri.port, use_ssl: false) do |http|
    http.request(request)
  end
  payload = JSON.parse(response.body)

  unless response.is_a?(Net::HTTPSuccess)
    message = response.body
    raise "OpenAI OCR failed: #{message}"
  end

  payload.dig('choices', 0, 'message', 'content').to_s
end