Class: Ultrasafeai::Parser::Client

Inherits:
Object
  • Object
show all
Defined in:
lib/ultrasafeai/parser/client.rb

Instance Method Summary collapse

Constructor Details

#initialize(client:, base_url: nil, environment: nil) ⇒ void

Parameters:



11
12
13
14
15
# File 'lib/ultrasafeai/parser/client.rb', line 11

def initialize(client:, base_url: nil, environment: nil)
  @client = client
  @base_url = base_url
  @environment = environment
end

Instance Method Details

#classify(request_options: {}, **params) ⇒ Ultrasafeai::Parser::Types::ClassifyParserResponse

Categorize a document against custom rules. Each rule specifies a type and a natural-language description.

Parameters:

  • request_options (Hash) (defaults to: {})
  • params (void)

Options Hash (request_options:):

  • :base_url (String)
  • :additional_headers (Hash{String => Object})
  • :additional_query_parameters (Hash{String => Object})
  • :additional_body_parameters (Hash{String => Object})
  • :timeout_in_seconds (Integer)

Returns:



140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
# File 'lib/ultrasafeai/parser/client.rb', line 140

def classify(request_options: {}, **params)
  params = Ultrasafeai::Internal::Types::Utils.normalize_keys(params)
  body = Internal::Multipart::FormData.new

  body.add_part(params[:file].to_form_data_part(name: "file")) if params[:file]
  if params[:rules]
    body.add(
      name: "rules",
      value: params[:rules]
    )
  end

  request = Ultrasafeai::Internal::Multipart::Request.new(
    base_url: request_options[:base_url] || @base_url || @environment&.dig(:base),
    method: "POST",
    path: "parser/classify",
    body: body,
    request_options: request_options
  )
  begin
    response = @client.send(request)
  rescue Net::HTTPRequestTimeout
    raise Ultrasafeai::Errors::TimeoutError
  end
  code = response.code.to_i
  if code.between?(200, 299)
    Ultrasafeai::Parser::Types::ClassifyParserResponse.load(response.body)
  else
    error_class = Ultrasafeai::Errors::ResponseError.subclass_for_code(code)
    raise error_class.new(response.body, code: code)
  end
end

#generate_instruction(request_options: {}, **params) ⇒ Ultrasafeai::Parser::Types::GenerateInstructionParserResponse

Upload a sample document to auto-generate a custom parsing instruction. This endpoint is free and does not consume credits.

Parameters:

  • request_options (Hash) (defaults to: {})
  • params (void)

Options Hash (request_options:):

  • :base_url (String)
  • :additional_headers (Hash{String => Object})
  • :additional_query_parameters (Hash{String => Object})
  • :additional_body_parameters (Hash{String => Object})
  • :timeout_in_seconds (Integer)

Returns:



304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
# File 'lib/ultrasafeai/parser/client.rb', line 304

def generate_instruction(request_options: {}, **params)
  params = Ultrasafeai::Internal::Types::Utils.normalize_keys(params)
  body = Internal::Multipart::FormData.new

  body.add_part(params[:file].to_form_data_part(name: "file")) if params[:file]
  if params[:parse_target]
    body.add(
      name: "parse_target",
      value: params[:parse_target]
    )
  end

  request = Ultrasafeai::Internal::Multipart::Request.new(
    base_url: request_options[:base_url] || @base_url || @environment&.dig(:base),
    method: "POST",
    path: "parser/generate-instruction",
    body: body,
    request_options: request_options
  )
  begin
    response = @client.send(request)
  rescue Net::HTTPRequestTimeout
    raise Ultrasafeai::Errors::TimeoutError
  end
  code = response.code.to_i
  if code.between?(200, 299)
    Ultrasafeai::Parser::Types::GenerateInstructionParserResponse.load(response.body)
  else
    error_class = Ultrasafeai::Errors::ResponseError.subclass_for_code(code)
    raise error_class.new(response.body, code: code)
  end
end

#parse(request_options: {}, **params) ⇒ Ultrasafeai::Types::UltraparserParseResponse

Extract structured content (text, markdown, JSON, images, or other formats) from a document. Submits a parse job and returns when complete. For progress feedback on large documents use POST /parser/stream.

Parameters:

  • request_options (Hash) (defaults to: {})
  • params (void)

Options Hash (request_options:):

  • :base_url (String)
  • :additional_headers (Hash{String => Object})
  • :additional_query_parameters (Hash{String => Object})
  • :additional_body_parameters (Hash{String => Object})
  • :timeout_in_seconds (Integer)

Returns:



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
# File 'lib/ultrasafeai/parser/client.rb', line 29

def parse(request_options: {}, **params)
  params = Ultrasafeai::Internal::Types::Utils.normalize_keys(params)
  body = Internal::Multipart::FormData.new

  body.add_part(params[:file].to_form_data_part(name: "file")) if params[:file]
  if params[:parse]
    body.add(
      name: "parse",
      value: params[:parse]
    )
  end
  if params[:extraction_target]
    body.add(
      name: "extraction_target",
      value: params[:extraction_target]
    )
  end
  if params[:instruction]
    body.add(
      name: "instruction",
      value: params[:instruction]
    )
  end

  request = Ultrasafeai::Internal::Multipart::Request.new(
    base_url: request_options[:base_url] || @base_url || @environment&.dig(:base),
    method: "POST",
    path: "parser",
    body: body,
    request_options: request_options
  )
  begin
    response = @client.send(request)
  rescue Net::HTTPRequestTimeout
    raise Ultrasafeai::Errors::TimeoutError
  end
  code = response.code.to_i
  if code.between?(200, 299)
    Ultrasafeai::Types::UltraparserParseResponse.load(response.body)
  else
    error_class = Ultrasafeai::Errors::ResponseError.subclass_for_code(code)
    raise error_class.new(response.body, code: code)
  end
end

#sheets(request_options: {}, **params) ⇒ Ultrasafeai::Parser::Types::SheetsParserResponse

Extract data from spreadsheet files (XLSX, CSV). Supports filtering by sheet name and cell range.

Parameters:

  • request_options (Hash) (defaults to: {})
  • params (void)

Options Hash (request_options:):

  • :base_url (String)
  • :additional_headers (Hash{String => Object})
  • :additional_query_parameters (Hash{String => Object})
  • :additional_body_parameters (Hash{String => Object})
  • :timeout_in_seconds (Integer)

Returns:



241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
# File 'lib/ultrasafeai/parser/client.rb', line 241

def sheets(request_options: {}, **params)
  params = Ultrasafeai::Internal::Types::Utils.normalize_keys(params)
  body = Internal::Multipart::FormData.new

  body.add_part(params[:file].to_form_data_part(name: "file")) if params[:file]
  if params[:sheet_names]
    body.add(
      name: "sheet_names",
      value: params[:sheet_names]
    )
  end
  if params[:extraction_range]
    body.add(
      name: "extraction_range",
      value: params[:extraction_range]
    )
  end
  if params[:include_hidden_cells]
    body.add(
      name: "include_hidden_cells",
      value: params[:include_hidden_cells]
    )
  end
  if params[:flatten_tables]
    body.add(
      name: "flatten_tables",
      value: params[:flatten_tables]
    )
  end

  request = Ultrasafeai::Internal::Multipart::Request.new(
    base_url: request_options[:base_url] || @base_url || @environment&.dig(:base),
    method: "POST",
    path: "parser/sheets",
    body: body,
    request_options: request_options
  )
  begin
    response = @client.send(request)
  rescue Net::HTTPRequestTimeout
    raise Ultrasafeai::Errors::TimeoutError
  end
  code = response.code.to_i
  if code.between?(200, 299)
    Ultrasafeai::Parser::Types::SheetsParserResponse.load(response.body)
  else
    error_class = Ultrasafeai::Errors::ResponseError.subclass_for_code(code)
    raise error_class.new(response.body, code: code)
  end
end

#split(request_options: {}, **params) ⇒ Ultrasafeai::Parser::Types::SplitParserResponse

Segment a document into sections based on custom categories. Optionally allow uncategorized pages and group sequential pages in the same category together.

Parameters:

  • request_options (Hash) (defaults to: {})
  • params (void)

Options Hash (request_options:):

  • :base_url (String)
  • :additional_headers (Hash{String => Object})
  • :additional_query_parameters (Hash{String => Object})
  • :additional_body_parameters (Hash{String => Object})
  • :timeout_in_seconds (Integer)

Returns:



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
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
# File 'lib/ultrasafeai/parser/client.rb', line 185

def split(request_options: {}, **params)
  params = Ultrasafeai::Internal::Types::Utils.normalize_keys(params)
  body = Internal::Multipart::FormData.new

  body.add_part(params[:file].to_form_data_part(name: "file")) if params[:file]
  if params[:categories]
    body.add(
      name: "categories",
      value: params[:categories]
    )
  end
  if params[:allow_uncategorized]
    body.add(
      name: "allow_uncategorized",
      value: params[:allow_uncategorized]
    )
  end
  if params[:group_sequential]
    body.add(
      name: "group_sequential",
      value: params[:group_sequential]
    )
  end

  request = Ultrasafeai::Internal::Multipart::Request.new(
    base_url: request_options[:base_url] || @base_url || @environment&.dig(:base),
    method: "POST",
    path: "parser/split",
    body: body,
    request_options: request_options
  )
  begin
    response = @client.send(request)
  rescue Net::HTTPRequestTimeout
    raise Ultrasafeai::Errors::TimeoutError
  end
  code = response.code.to_i
  if code.between?(200, 299)
    Ultrasafeai::Parser::Types::SplitParserResponse.load(response.body)
  else
    error_class = Ultrasafeai::Errors::ResponseError.subclass_for_code(code)
    raise error_class.new(response.body, code: code)
  end
end

#stream(request_options: {}, **params) ⇒ untyped

Same as POST /parser but returns an SSE stream of progress events instead of blocking until completion. Set Accept: text/event-stream.

Parameters:

  • request_options (Hash) (defaults to: {})
  • params (void)

Options Hash (request_options:):

  • :base_url (String)
  • :additional_headers (Hash{String => Object})
  • :additional_query_parameters (Hash{String => Object})
  • :additional_body_parameters (Hash{String => Object})
  • :timeout_in_seconds (Integer)

Returns:

  • (untyped)

Raises:

  • (error_class)


86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
# File 'lib/ultrasafeai/parser/client.rb', line 86

def stream(request_options: {}, **params)
  params = Ultrasafeai::Internal::Types::Utils.normalize_keys(params)
  body = Internal::Multipart::FormData.new

  body.add_part(params[:file].to_form_data_part(name: "file")) if params[:file]
  if params[:parse]
    body.add(
      name: "parse",
      value: params[:parse]
    )
  end
  if params[:extraction_target]
    body.add(
      name: "extraction_target",
      value: params[:extraction_target]
    )
  end
  if params[:instruction]
    body.add(
      name: "instruction",
      value: params[:instruction]
    )
  end

  request = Ultrasafeai::Internal::Multipart::Request.new(
    base_url: request_options[:base_url] || @base_url || @environment&.dig(:base),
    method: "POST",
    path: "parser/stream",
    body: body,
    request_options: request_options
  )
  begin
    response = @client.send(request)
  rescue Net::HTTPRequestTimeout
    raise Ultrasafeai::Errors::TimeoutError
  end
  code = response.code.to_i
  return if code.between?(200, 299)

  error_class = Ultrasafeai::Errors::ResponseError.subclass_for_code(code)
  raise error_class.new(response.body, code: code)
end