sapiopycommons 2025.8.29a719__py3-none-any.whl → 2025.9.4a723__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.


This version of sapiopycommons might be problematic. Click here for more details.

Files changed (51) hide show
  1. sapiopycommons/ai/tool_of_tools.py +917 -0
  2. sapiopycommons/callbacks/callback_util.py +26 -16
  3. sapiopycommons/eln/experiment_handler.py +12 -5
  4. sapiopycommons/files/assay_plate_reader.py +93 -0
  5. sapiopycommons/files/file_text_converter.py +207 -0
  6. sapiopycommons/flowcyto/flow_cyto.py +2 -24
  7. sapiopycommons/general/accession_service.py +2 -28
  8. sapiopycommons/multimodal/multimodal.py +2 -24
  9. sapiopycommons/recordmodel/record_handler.py +4 -0
  10. sapiopycommons/rules/eln_rule_handler.py +3 -0
  11. sapiopycommons/rules/on_save_rule_handler.py +3 -0
  12. sapiopycommons/webhook/webservice_handlers.py +1 -1
  13. {sapiopycommons-2025.8.29a719.dist-info → sapiopycommons-2025.9.4a723.dist-info}/METADATA +2 -2
  14. {sapiopycommons-2025.8.29a719.dist-info → sapiopycommons-2025.9.4a723.dist-info}/RECORD +16 -48
  15. sapiopycommons/ai/converter_service_base.py +0 -131
  16. sapiopycommons/ai/protoapi/fielddefinitions/fields_pb2.py +0 -43
  17. sapiopycommons/ai/protoapi/fielddefinitions/fields_pb2.pyi +0 -31
  18. sapiopycommons/ai/protoapi/fielddefinitions/fields_pb2_grpc.py +0 -24
  19. sapiopycommons/ai/protoapi/fielddefinitions/velox_field_def_pb2.py +0 -123
  20. sapiopycommons/ai/protoapi/fielddefinitions/velox_field_def_pb2.pyi +0 -598
  21. sapiopycommons/ai/protoapi/fielddefinitions/velox_field_def_pb2_grpc.py +0 -24
  22. sapiopycommons/ai/protoapi/plan/converter/converter_pb2.py +0 -51
  23. sapiopycommons/ai/protoapi/plan/converter/converter_pb2.pyi +0 -63
  24. sapiopycommons/ai/protoapi/plan/converter/converter_pb2_grpc.py +0 -149
  25. sapiopycommons/ai/protoapi/plan/item/item_container_pb2.py +0 -55
  26. sapiopycommons/ai/protoapi/plan/item/item_container_pb2.pyi +0 -90
  27. sapiopycommons/ai/protoapi/plan/item/item_container_pb2_grpc.py +0 -24
  28. sapiopycommons/ai/protoapi/plan/script/script_pb2.py +0 -59
  29. sapiopycommons/ai/protoapi/plan/script/script_pb2.pyi +0 -102
  30. sapiopycommons/ai/protoapi/plan/script/script_pb2_grpc.py +0 -153
  31. sapiopycommons/ai/protoapi/plan/step_output_pb2.py +0 -45
  32. sapiopycommons/ai/protoapi/plan/step_output_pb2.pyi +0 -42
  33. sapiopycommons/ai/protoapi/plan/step_output_pb2_grpc.py +0 -24
  34. sapiopycommons/ai/protoapi/plan/step_pb2.py +0 -43
  35. sapiopycommons/ai/protoapi/plan/step_pb2.pyi +0 -43
  36. sapiopycommons/ai/protoapi/plan/step_pb2_grpc.py +0 -24
  37. sapiopycommons/ai/protoapi/plan/tool/entry_pb2.py +0 -41
  38. sapiopycommons/ai/protoapi/plan/tool/entry_pb2.pyi +0 -35
  39. sapiopycommons/ai/protoapi/plan/tool/entry_pb2_grpc.py +0 -24
  40. sapiopycommons/ai/protoapi/plan/tool/tool_pb2.py +0 -75
  41. sapiopycommons/ai/protoapi/plan/tool/tool_pb2.pyi +0 -237
  42. sapiopycommons/ai/protoapi/plan/tool/tool_pb2_grpc.py +0 -154
  43. sapiopycommons/ai/protoapi/session/sapio_conn_info_pb2.py +0 -39
  44. sapiopycommons/ai/protoapi/session/sapio_conn_info_pb2.pyi +0 -32
  45. sapiopycommons/ai/protoapi/session/sapio_conn_info_pb2_grpc.py +0 -24
  46. sapiopycommons/ai/protobuf_utils.py +0 -504
  47. sapiopycommons/ai/server.py +0 -104
  48. sapiopycommons/ai/test_client.py +0 -356
  49. sapiopycommons/ai/tool_service_base.py +0 -951
  50. {sapiopycommons-2025.8.29a719.dist-info → sapiopycommons-2025.9.4a723.dist-info}/WHEEL +0 -0
  51. {sapiopycommons-2025.8.29a719.dist-info → sapiopycommons-2025.9.4a723.dist-info}/licenses/LICENSE +0 -0
@@ -1,951 +0,0 @@
1
- from __future__ import annotations
2
-
3
- import base64
4
- import io
5
- import json
6
- import logging
7
- import re
8
- import traceback
9
- from abc import abstractmethod, ABC
10
- from logging import Logger
11
- from typing import Any, Iterable, Sequence, Mapping
12
-
13
- from grpc import ServicerContext
14
- from sapiopylib.rest.User import SapioUser, ensure_logger_initialized
15
- from sapiopylib.rest.pojo.datatype.FieldDefinition import AbstractVeloxFieldDefinition
16
-
17
- from sapiopycommons.ai.protoapi.fielddefinitions.fields_pb2 import FieldValueMapPbo, FieldValuePbo
18
- from sapiopycommons.ai.protoapi.fielddefinitions.velox_field_def_pb2 import VeloxFieldDefPbo, FieldTypePbo, \
19
- SelectionPropertiesPbo, IntegerPropertiesPbo, DoublePropertiesPbo, BooleanPropertiesPbo, StringPropertiesPbo
20
- from sapiopycommons.ai.protoapi.plan.item.item_container_pb2 import ContentTypePbo
21
- from sapiopycommons.ai.protoapi.plan.tool.entry_pb2 import StepOutputBatchPbo, StepItemContainerPbo, \
22
- StepBinaryContainerPbo, StepCsvContainerPbo, StepCsvHeaderRowPbo, StepCsvRowPbo, StepJsonContainerPbo, \
23
- StepTextContainerPbo, StepInputBatchPbo
24
- from sapiopycommons.ai.protoapi.plan.tool.tool_pb2 import ToolDetailsRequestPbo, ToolDetailsResponsePbo, \
25
- ToolDetailsPbo, ProcessStepRequestPbo, ProcessStepResponsePbo, ToolOutputDetailsPbo, ToolIoConfigBasePbo, \
26
- ToolInputDetailsPbo, ExampleContainerPbo, ProcessStepResponseStatusPbo
27
- from sapiopycommons.ai.protoapi.plan.tool.tool_pb2_grpc import ToolServiceServicer
28
- from sapiopycommons.ai.protoapi.session.sapio_conn_info_pb2 import SapioUserSecretTypePbo, SapioConnectionInfoPbo
29
- from sapiopycommons.ai.protobuf_utils import ProtobufUtils
30
- from sapiopycommons.ai.test_client import ContainerType
31
- from sapiopycommons.files.file_util import FileUtil
32
- from sapiopycommons.general.aliases import FieldMap, FieldValue
33
-
34
-
35
- # FR-47422: Created classes.
36
- class SapioToolResult(ABC):
37
- """
38
- A class representing a result from a Sapio tool. Instantiate one of the subclasses to create a result object.
39
- """
40
-
41
- @abstractmethod
42
- def to_proto(self) -> StepOutputBatchPbo | list[FieldValueMapPbo]:
43
- """
44
- Convert this SapioToolResult object to a StepOutputBatchPbo or list of FieldValueMapPbo proto objects.
45
- """
46
- pass
47
-
48
-
49
- class BinaryResult(SapioToolResult):
50
- """
51
- A class representing binary results from a Sapio tool.
52
- """
53
- binary_data: list[bytes]
54
- content_type: str
55
- file_extensions: list[str]
56
- name: str
57
-
58
- def __init__(self, binary_data: list[bytes], content_type: str = "binary", file_extensions: list[str] = None,
59
- name: str | None = None):
60
- """
61
- :param binary_data: The binary data as a list of bytes.
62
- :param content_type: The content type of the data.
63
- :param file_extensions: A list of file extensions that this binary data can be saved as.
64
- :param name: An optional identifying name for this result that will be accessible to the next tool.
65
- """
66
- self.binary_data = binary_data
67
- self.content_type = content_type
68
- self.file_extensions = file_extensions if file_extensions else []
69
- self.name = name
70
-
71
- def to_proto(self) -> StepOutputBatchPbo | list[FieldValueMapPbo]:
72
- return StepOutputBatchPbo(
73
- item_container=StepItemContainerPbo(
74
- content_type=ContentTypePbo(name=self.content_type, extensions=self.file_extensions),
75
- container_name=self.name,
76
- binary_container=StepBinaryContainerPbo(items=self.binary_data)
77
- )
78
- )
79
-
80
-
81
- class CsvResult(SapioToolResult):
82
- """
83
- A class representing CSV results from a Sapio tool.
84
- """
85
- csv_data: list[dict[str, Any]]
86
- content_type: str
87
- file_extensions: list[str]
88
- name: str
89
-
90
- def __init__(self, csv_data: list[dict[str, Any]], content_type: str = "csv", file_extensions: list[str] = None,
91
- name: str | None = None):
92
- """
93
- :param csv_data: The list of CSV data results, provided as a list of dictionaries of column name to value.
94
- :param content_type: The content type of the data.
95
- :param file_extensions: A list of file extensions that this binary data can be saved as.
96
- :param name: An optional identifying name for this result that will be accessible to the next tool.
97
- """
98
- self.csv_data = csv_data
99
- self.content_type = content_type
100
- self.file_extensions = file_extensions if file_extensions else ["csv"]
101
- self.name = name
102
-
103
- def to_proto(self) -> StepOutputBatchPbo | list[FieldValueMapPbo]:
104
- return StepOutputBatchPbo(
105
- item_container=StepItemContainerPbo(
106
- content_type=ContentTypePbo(name=self.content_type, extensions=self.file_extensions),
107
- container_name=self.name,
108
- csv_container=StepCsvContainerPbo(
109
- header=StepCsvHeaderRowPbo(cells=self.csv_data[0].keys()),
110
- items=[StepCsvRowPbo(cells=[str(x) for x in row.values()]) for row in self.csv_data]
111
- )
112
- ) if self.csv_data else None
113
- )
114
-
115
-
116
- class FieldMapResult(SapioToolResult):
117
- """
118
- A class representing field map results from a Sapio tool.
119
- """
120
- field_maps: list[FieldMap]
121
-
122
- def __init__(self, field_maps: list[FieldMap]):
123
- """
124
- :param field_maps: A list of field maps, where each map is a dictionary of field names to values. Each entry
125
- will create a new data record in the system, so long as the tool definition specifies an output data type
126
- name.
127
- """
128
- self.field_maps = field_maps
129
-
130
- def to_proto(self) -> StepOutputBatchPbo | list[FieldValueMapPbo]:
131
- new_records: list[FieldValueMapPbo] = []
132
- for field_map in self.field_maps:
133
- fields: dict[str, FieldValuePbo] = {}
134
- for field, value in field_map.items():
135
- field_value = FieldValuePbo()
136
- if isinstance(value, str):
137
- field_value.string_value = value
138
- elif isinstance(value, int):
139
- field_value.int_value = value
140
- elif isinstance(value, float):
141
- field_value.double_value = value
142
- elif isinstance(value, bool):
143
- field_value.bool_value = value
144
- fields[field] = field_value
145
- new_records.append(FieldValueMapPbo(fields=fields))
146
- return new_records
147
-
148
-
149
- class JsonResult(SapioToolResult):
150
- """
151
- A class representing JSON results from a Sapio tool.
152
- """
153
- json_data: list[Any]
154
- content_type: str
155
- file_extensions: list[str]
156
- name: str
157
-
158
- def __init__(self, json_data: list[Any], content_type: str = "json", file_extensions: list[str] = None,
159
- name: str | None = None):
160
- """
161
- :param json_data: The list of JSON data results. Each entry in the list represents a separate JSON object.
162
- These entries must be able to be serialized to JSON using json.dumps().
163
- :param content_type: The content type of the data.
164
- :param file_extensions: A list of file extensions that this binary data can be saved as.
165
- :param name: An optional identifying name for this result that will be accessible to the next tool.
166
- """
167
- self.json_data = json_data
168
- self.content_type = content_type
169
- self.file_extensions = file_extensions if file_extensions else ["json"]
170
- self.name = name
171
-
172
- def to_proto(self) -> StepOutputBatchPbo | list[FieldValueMapPbo]:
173
- return StepOutputBatchPbo(
174
- item_container=StepItemContainerPbo(
175
- content_type=ContentTypePbo(name=self.content_type, extensions=self.file_extensions),
176
- container_name=self.name,
177
- json_container=StepJsonContainerPbo(items=[json.dumps(x) for x in self.json_data])
178
- )
179
- )
180
-
181
-
182
- class TextResult(SapioToolResult):
183
- """
184
- A class representing text results from a Sapio tool.
185
- """
186
- text_data: list[str]
187
- content_type: str
188
- file_extensions: list[str]
189
- name: str
190
-
191
- def __init__(self, text_data: list[str], content_type: str = "text", file_extensions: list[str] = None,
192
- name: str | None = None):
193
- """
194
- :param text_data: The text data as a list of strings.
195
- :param content_type: The content type of the data.
196
- :param file_extensions: A list of file extensions that this binary data can be saved as.
197
- :param name: An optional identifying name for this result that will be accessible to the next tool.
198
- """
199
- self.text_data = text_data
200
- self.content_type = content_type
201
- self.file_extensions = file_extensions if file_extensions else ["txt"]
202
- self.name = name
203
-
204
- def to_proto(self) -> StepOutputBatchPbo | list[FieldValueMapPbo]:
205
- return StepOutputBatchPbo(
206
- item_container=StepItemContainerPbo(
207
- content_type=ContentTypePbo(name=self.content_type, extensions=self.file_extensions),
208
- container_name=self.name,
209
- text_container=StepTextContainerPbo(items=self.text_data)
210
- )
211
- )
212
-
213
-
214
- class ToolServiceBase(ToolServiceServicer, ABC):
215
- """
216
- A base class for implementing a tool service. Subclasses should implement the register_tools method to register
217
- their tools with the service.
218
- """
219
- def GetToolDetails(self, request: ToolDetailsRequestPbo, context: ServicerContext) -> ToolDetailsResponsePbo:
220
- try:
221
- # Get the tool details from the registered tools.
222
- details: list[ToolDetailsPbo] = []
223
- for tool in self.register_tools():
224
- details.append(tool().to_pbo())
225
- if not details:
226
- raise Exception("No tools registered with this service.")
227
- return ToolDetailsResponsePbo(tool_framework_version=self.tool_version(), tool_details=details)
228
- except Exception as e:
229
- # Woe to you if you somehow cause an exception to be raised when just initializing your tools.
230
- # There's no way to log this.
231
- print(f"CRITICAL ERROR: {e}")
232
- print(traceback.format_exc())
233
- return ToolDetailsResponsePbo()
234
-
235
- def ProcessData(self, request: ProcessStepRequestPbo, context: ServicerContext) -> ProcessStepResponsePbo:
236
- try:
237
- # Convert the SapioConnectionInfo proto object to a SapioUser object.
238
- user = self._create_user(request.sapio_user)
239
- # Get the tool results from the registered tool matching the request.
240
- success, msg, results, logs = self.run(user, request, context)
241
- # Convert the results to protobuf objects.
242
- output_data: list[StepOutputBatchPbo] = []
243
- new_records: list[FieldValueMapPbo] = []
244
- for result in results:
245
- data: StepOutputBatchPbo | list[FieldValueMapPbo] = result.to_proto()
246
- if isinstance(data, StepOutputBatchPbo):
247
- output_data.append(data)
248
- else:
249
- new_records.extend(data)
250
- # Return a ProcessStepResponse proto object containing the results to the caller.
251
- status = ProcessStepResponseStatusPbo.SUCCESS if success else ProcessStepResponseStatusPbo.FAILURE
252
- return ProcessStepResponsePbo(status=status, status_message=msg, output=output_data, log=logs,
253
- new_records=new_records)
254
- except Exception as e:
255
- # This try/except should never be needed, as the tool should handle its own exceptions, but better safe
256
- # than sorry.
257
- print(f"CRITICAL ERROR: {e}")
258
- print(traceback.format_exc())
259
- return ProcessStepResponsePbo(status=ProcessStepResponseStatusPbo.FAILURE,
260
- status_message=f"CRITICAL ERROR: {e}",
261
- log=[traceback.format_exc()])
262
-
263
- @staticmethod
264
- def _create_user(info: SapioConnectionInfoPbo, timeout_seconds: int = 60) -> SapioUser:
265
- """
266
- Create a SapioUser object from the given SapioConnectionInfo proto object.
267
-
268
- :param info: The SapioConnectionInfo proto object.
269
- :param timeout_seconds: The request timeout for calls made from this user object.
270
- """
271
- user = SapioUser(info.webservice_url, True, timeout_seconds, guid=info.app_guid)
272
- match info.secret_type:
273
- case SapioUserSecretTypePbo.SESSION_TOKEN:
274
- user.api_token = info.secret
275
- case SapioUserSecretTypePbo.PASSWORD:
276
- secret: str = info.secret
277
- if secret.startswith("Basic "):
278
- secret = secret[6:]
279
- credentials: list[str] = base64.b64decode(secret).decode().split(":", 1)
280
- user.username = credentials[0]
281
- user.password = credentials[1]
282
- case _:
283
- raise Exception(f"Unexpected secret type: {info.secret_type}")
284
- return user
285
-
286
- @staticmethod
287
- def tool_version() -> int:
288
- """
289
- :return: The version of this set of tools.
290
- """
291
- return 1
292
-
293
- @abstractmethod
294
- def register_tools(self) -> list[type[ToolBase]]:
295
- """
296
- Register tool types with this service. Provided tools should implement the ToolBase class.
297
-
298
- :return: A list of tools to register to this service.
299
- """
300
- pass
301
-
302
- def run(self, user: SapioUser, request: ProcessStepRequestPbo, context: ServicerContext) \
303
- -> tuple[bool, str, list[SapioToolResult], list[str]]:
304
- """
305
- Execute a tool from this service.
306
-
307
- :param user: A user object that can be used to initialize manager classes using DataMgmtServer to query the
308
- system.
309
- :param request: The request object containing the input data.
310
- :param context: The gRPC context.
311
- :return: Whether or not the tool succeeded, the status message, the results of the tool, and any logs
312
- generated by the tool.
313
- """
314
- # Locate the tool named in the request.
315
- find_tool: str = request.tool_name
316
- registered_tools: dict[str, type[ToolBase]] = {t.name(): t for t in self.register_tools()}
317
- if find_tool not in registered_tools:
318
- # If the tool is not found, list all of the registered tools for this service so that the LLM can correct
319
- # the tool it is requesting.
320
- all_tool_names: str = "\n".join(registered_tools.keys())
321
- msg: str = (f"Tool \"{find_tool}\" not found in the registered tools for this service. The registered tools "
322
- f"for this service are: \n{all_tool_names}")
323
- return False, msg, [], []
324
-
325
- # Instantiate the tool class.
326
- tool: ToolBase = registered_tools[find_tool]()
327
- try:
328
- # Setup the tool with details from the request.
329
- tool.setup(user, request, context)
330
- # Validate that the provided inputs match the tool's expected inputs.
331
- if len(request.input) != len(tool.input_configs):
332
- msg: str = f"Expected {len(tool.input_configs)} inputs for this tool, but got {len(request.input)} instead."
333
- else:
334
- msg: str = tool.validate_input()
335
- # If there is no error message, then the inputs are valid.
336
- success: bool = not bool(msg)
337
- # If this is a dry run, then provide the fixed dry run output.
338
- # Otherwise, if the inputs were successfully validated, then the tool is executed normally.
339
- results: list[SapioToolResult] = []
340
- if request.dry_run:
341
- results = tool.dry_run_output()
342
- elif success:
343
- results = tool.run(user)
344
- # Update the status message to reflect the successful execution of the tool.
345
- msg = f"{tool.name()} successfully completed."
346
- return success, msg, results, tool.logs
347
- except Exception as e:
348
- tool.log_exception("Exception occurred during tool execution.", e)
349
- return False, str(e), [], tool.logs
350
-
351
-
352
- class ToolBase(ABC):
353
- """
354
- A base class for implementing a tool.
355
- """
356
- _name: str
357
- _description: str
358
- _data_type_name: str | None
359
- input_configs: list[ToolInputDetailsPbo]
360
- _input_container_types: list[ContainerType]
361
- output_configs: list[ToolOutputDetailsPbo]
362
- _output_container_types: list[ContainerType]
363
- config_fields: list[VeloxFieldDefPbo]
364
-
365
- logs: list[str]
366
- logger: Logger
367
- verbose_logging: bool
368
-
369
- user: SapioUser
370
- request: ProcessStepRequestPbo
371
- context: ServicerContext
372
-
373
- @staticmethod
374
- @abstractmethod
375
- def name() -> str:
376
- """
377
- :return: The name of the tool. This should be unique across all tools in the service.
378
- """
379
- pass
380
-
381
- @staticmethod
382
- @abstractmethod
383
- def description() -> str:
384
- """
385
- :return: A description of the tool.
386
- """
387
- pass
388
-
389
- @staticmethod
390
- def data_type_name() -> str | None:
391
- """
392
- :return: The name of the output data type of this tool, if applicable. When this tool returns
393
- FieldMapResult objects in its run method, this name will be used to set the data type of the output data.
394
- """
395
- return None
396
-
397
- def __init__(self):
398
- self._name = self.name()
399
- self._description = self.description()
400
- self._data_type_name = self.data_type_name()
401
- self.input_configs = []
402
- self._input_container_types = []
403
- self.output_configs = []
404
- self._output_container_types = []
405
- self.config_fields = []
406
- self.logs = []
407
- self.logger = logging.getLogger(f"ToolBase.{self._name}")
408
- ensure_logger_initialized(self.logger)
409
-
410
- def setup(self, user: SapioUser, request: ProcessStepRequestPbo, context: ServicerContext) -> None:
411
- """
412
- Setup the tool with the user, request, and context. This method can be overridden by subclasses to perform
413
- additional setup.
414
-
415
- :param user: A user object that can be used to initialize manager classes using DataMgmtServer to query the
416
- system.
417
- :param request: The request object containing the input data.
418
- :param context: The gRPC context.
419
- """
420
- self.user = user
421
- self.request = request
422
- self.context = context
423
- self.verbose_logging = request.verbose_logging
424
-
425
- def add_input(self, container_type: ContainerType, content_type: str, display_name: str, description: str,
426
- structure_example: str | bytes | None = None, validation: str | None = None,
427
- input_count: tuple[int, int] | None = None, is_paged: bool = False,
428
- page_size: tuple[int, int] | None = None, max_request_bytes: int | None = None) -> None:
429
- """
430
- Add an input configuration to the tool. This determines how many inputs this tool will accept in the plan
431
- manager, as well as what those inputs are. The IO number of the input will be set to the current number of
432
- inputs. That is, the first time this is called, the IO number will be 0, the second time it is called, the IO
433
- number will be 1, and so on.
434
-
435
- :param container_type: The container type of the input.
436
- :param content_type: The content type of the input.
437
- :param display_name: The display name of the input.
438
- :param description: The description of the input.
439
- :param structure_example: An optional example of the structure of the input, such as how the structure of a
440
- JSON output may look. This does not need to be an entirely valid example, and should often be truncated for
441
- brevity.
442
- :param validation: An optional validation string for the input.
443
- :param input_count: A tuple of the minimum and maximum number of inputs allowed for this tool.
444
- :param is_paged: If true, this input will be paged. If false, this input will not be paged.
445
- :param page_size: A tuple of the minimum and maximum page size for this tool. The input must be paged in order
446
- for this to have an effect.
447
- :param max_request_bytes: The maximum request size in bytes for this tool.
448
- """
449
- structure: ExampleContainerPbo | None = None
450
- if isinstance(structure_example, str):
451
- structure = ExampleContainerPbo(text_example=structure_example)
452
- elif isinstance(structure_example, bytes):
453
- structure = ExampleContainerPbo(binary_example=structure_example)
454
- self.input_configs.append(ToolInputDetailsPbo(
455
- base_config=ToolIoConfigBasePbo(
456
- io_number=len(self.input_configs),
457
- content_type=content_type,
458
- display_name=display_name,
459
- description=description,
460
- structure_example=structure
461
- ),
462
- validation=validation,
463
- min_input_count=input_count[0] if input_count else None,
464
- max_input_count=input_count[1] if input_count else None,
465
- paged=is_paged,
466
- min_page_size=page_size[0] if page_size else None,
467
- max_page_size=page_size[1] if page_size else None,
468
- max_request_bytes=max_request_bytes,
469
- ))
470
- self._input_container_types.append(container_type)
471
-
472
- def add_output(self, container_type: ContainerType, content_type: str, display_name: str, description: str,
473
- testing_example: str | bytes, structure_example: str | bytes | None = None) -> None:
474
- """
475
- Add an output configuration to the tool. This determines how many inputs this tool will accept in the plan
476
- manager, as well as what those inputs are. The IO number of the output will be set to the current number of
477
- outputs. That is, the first time this is called, the IO number will be 0, the second time it is called, the IO
478
- number will be 1, and so on.
479
-
480
- :param container_type: The container type of the output.
481
- :param content_type: The content type of the output.
482
- :param display_name: The display name of the output.
483
- :param description: The description of the output.
484
- :param testing_example: An example of the input to be used when testing this tool in the system. This must be
485
- an entirely valid example of what an output of this tool could look like so that it can be properly used
486
- to run tests with. The provided example may be a string, such as for representing JSON or CSV outputs,
487
- or bytes, such as for representing binary outputs like images or files.
488
- :param structure_example: An optional example of the structure of the input, such as how the structure of a
489
- JSON output may look. This does not need to be an entirely valid example, and should often be truncated for
490
- brevity.
491
- """
492
- testing: ExampleContainerPbo | None = None
493
- if isinstance(testing_example, str):
494
- testing = ExampleContainerPbo(text_example=testing_example)
495
- elif isinstance(testing_example, bytes):
496
- testing = ExampleContainerPbo(binary_example=testing_example)
497
-
498
- structure: ExampleContainerPbo | None = None
499
- if isinstance(structure_example, str):
500
- structure = ExampleContainerPbo(text_example=structure_example)
501
- elif isinstance(structure_example, bytes):
502
- structure = ExampleContainerPbo(binary_example=structure_example)
503
-
504
- self.output_configs.append(ToolOutputDetailsPbo(
505
- base_config=ToolIoConfigBasePbo(
506
- io_number=len(self.output_configs),
507
- content_type=content_type,
508
- display_name=display_name,
509
- description=description,
510
- structure_example=structure,
511
- testing_example=testing
512
- )))
513
- self._output_container_types.append(container_type)
514
-
515
- def add_config_field(self, field: VeloxFieldDefPbo) -> None:
516
- """
517
- Add a configuration field to the tool. This field will be used to configure the tool in the plan manager.
518
-
519
- :param field: The configuration field details.
520
- """
521
- self.config_fields.append(field)
522
-
523
- def add_config_field_def(self, field: AbstractVeloxFieldDefinition) -> None:
524
- """
525
- Add a configuration field to the tool. This field will be used to configure the tool in the plan manager.
526
-
527
- :param field: The configuration field details.
528
- """
529
- self.config_fields.append(ProtobufUtils.field_def_to_pbo(field))
530
-
531
- def add_boolean_config_field(self, field_name: str, display_name: str, description: str, default_value: bool,
532
- optional: bool = False) -> None:
533
- """
534
- Add a boolean configuration field to the tool. This field will be used to configure the tool in the plan
535
- manager.
536
-
537
- :param field_name: The name of the field.
538
- :param display_name: The display name of the field.
539
- :param description: The description of the field.
540
- :param default_value: The default value of the field.
541
- :param optional: If true, this field is optional. If false, this field is required.
542
- """
543
- self.config_fields.append(VeloxFieldDefPbo(
544
- data_field_type=FieldTypePbo.BOOLEAN,
545
- data_field_name=field_name,
546
- display_name=display_name,
547
- description=description,
548
- required=not optional,
549
- editable=True,
550
- boolean_properties=BooleanPropertiesPbo(
551
- default_value=default_value
552
- )
553
- ))
554
-
555
- def add_double_config_field(self, field_name: str, display_name: str, description: str, default_value: float,
556
- min_value: float = -10.**120, max_value: float = 10.**120, precision: int = 2,
557
- optional: bool = False) -> None:
558
- """
559
- Add a double configuration field to the tool. This field will be used to configure the tool in the plan
560
- manager.
561
-
562
- :param field_name: The name of the field.
563
- :param display_name: The display name of the field.
564
- :param description: The description of the field.
565
- :param default_value: The default value of the field.
566
- :param min_value: The minimum value of the field.
567
- :param max_value: The maximum value of the field.
568
- :param precision: The precision of the field.
569
- :param optional: If true, this field is optional. If false, this field is required.
570
- """
571
- self.config_fields.append(VeloxFieldDefPbo(
572
- data_field_type=FieldTypePbo.DOUBLE,
573
- data_field_name=field_name,
574
- display_name=display_name,
575
- description=description,
576
- required=not optional,
577
- editable=True,
578
- double_properties=DoublePropertiesPbo(
579
- default_value=default_value,
580
- min_value=min_value,
581
- max_value=max_value,
582
- precision=precision
583
- )
584
- ))
585
-
586
- def add_integer_config_field(self, field_name: str, display_name: str, description: str,
587
- default_value: int, min_value: int = -2**31, max_value: int = 2**31-1,
588
- optional: bool = False) -> None:
589
- """
590
- Add an integer configuration field to the tool. This field will be used to configure the tool in the plan
591
- manager.
592
-
593
- :param field_name: The name of the field.
594
- :param display_name: The display name of the field.
595
- :param description: The description of the field.
596
- :param default_value: The default value of the field.
597
- :param min_value: The minimum value of the field.
598
- :param max_value: The maximum value of the field.
599
- :param optional: If true, this field is optional. If false, this field is required.
600
- """
601
- self.config_fields.append(VeloxFieldDefPbo(
602
- data_field_type=FieldTypePbo.INTEGER,
603
- data_field_name=field_name,
604
- display_name=display_name,
605
- description=description,
606
- required=not optional,
607
- editable=True,
608
- integer_properties=IntegerPropertiesPbo(
609
- default_value=default_value,
610
- min_value=min_value,
611
- max_value=max_value
612
- )
613
- ))
614
-
615
- def add_string_config_field(self, field_name: str, display_name: str, description: str,
616
- default_value: str, max_length: int = 1000, optional: bool = False) -> None:
617
- """
618
- Add a string configuration field to the tool. This field will be used to configure the tool in the plan
619
- manager.
620
-
621
- :param field_name: The name of the field.
622
- :param display_name: The display name of the field.
623
- :param description: The description of the field.
624
- :param default_value: The default value of the field.
625
- :param max_length: The maximum length of the field.
626
- :param optional: If true, this field is optional. If false, this field is required.
627
- """
628
- self.config_fields.append(VeloxFieldDefPbo(
629
- data_field_type=FieldTypePbo.STRING,
630
- data_field_name=field_name,
631
- display_name=display_name,
632
- description=description,
633
- required=not optional,
634
- editable=True,
635
- string_properties=StringPropertiesPbo(
636
- default_value=default_value,
637
- max_length=max_length
638
- )
639
- ))
640
-
641
- def add_list_config_field(self, field_name: str, display_name: str, description: str, default_value: str,
642
- allowed_values: list[str], direct_edit: bool = False, optional: bool = False) -> None:
643
- """
644
- Add a list configuration field to the tool. This field will be used to configure the tool in the plan
645
- manager.
646
-
647
- :param field_name: The name of the field.
648
- :param display_name: The display name of the field.
649
- :param description: The description of the field.
650
- :param default_value: The default value of the field.
651
- :param allowed_values: The list of allowed values for the field.
652
- :param direct_edit: If true, the user can enter a value that is not in the list of allowed values. If false,
653
- the user can only select from the list of allowed values.
654
- :param optional: If true, this field is optional. If false, this field is required.
655
- """
656
- self.config_fields.append(VeloxFieldDefPbo(
657
- data_field_type=FieldTypePbo.SELECTION,
658
- data_field_name=field_name,
659
- display_name=display_name,
660
- description=description,
661
- required=not optional,
662
- editable=True,
663
- selection_properties=SelectionPropertiesPbo(
664
- default_value=default_value,
665
- static_list_values=allowed_values,
666
- direct_edit=direct_edit,
667
- )
668
- ))
669
-
670
- def add_multi_list_config_field(self, field_name: str, display_name: str, description: str,
671
- default_value: list[str], allowed_values: list[str], direct_edit: bool = False,
672
- optional: bool = False) -> None:
673
- """
674
- Add a multi-select list configuration field to the tool. This field will be used to configure the tool in the
675
- plan manager.
676
-
677
- :param field_name: The name of the field.
678
- :param display_name: The display name of the field.
679
- :param description: The description of the field.
680
- :param default_value: The default value of the field.
681
- :param allowed_values: The list of allowed values for the field.
682
- :param direct_edit: If true, the user can enter a value that is not in the list of allowed values. If false,
683
- the user can only select from the list of allowed values.
684
- :param optional: If true, this field is optional. If false, this field is required.
685
- """
686
- self.config_fields.append(VeloxFieldDefPbo(
687
- data_field_type=FieldTypePbo.SELECTION,
688
- data_field_name=field_name,
689
- display_name=display_name,
690
- description=description,
691
- required=not optional,
692
- editable=True,
693
- selection_properties=SelectionPropertiesPbo(
694
- default_value=",".join(default_value),
695
- static_list_values=allowed_values,
696
- multi_select=True,
697
- direct_edit=direct_edit,
698
- )
699
- ))
700
-
701
- def to_pbo(self) -> ToolDetailsPbo:
702
- """
703
- :return: The ToolDetailsPbo proto object representing this tool.
704
- """
705
- return ToolDetailsPbo(
706
- name=self._name,
707
- description=self._description,
708
- input_configs=self.input_configs,
709
- output_configs=self.output_configs,
710
- output_data_type_name=self._data_type_name,
711
- config_fields=self.config_fields
712
- )
713
-
714
- @abstractmethod
715
- def validate_input(self) -> str | None:
716
- """
717
- Validate the request given to this tool. If the request is validly formatted, this method should return None.
718
- If the request is not valid, this method should return an error message indicating what is wrong with the
719
- request.
720
-
721
- This method should not perform any actual processing of the request. It should only validate the inputs and
722
- configurations provided in the request.
723
-
724
- The request inputs can be accessed using the self.get_input_*() methods.
725
- The request settings can be accessed using the self.get_config_fields() method.
726
- The request itself can be accessed using self.request.
727
-
728
- :return: A tuple containing a boolean indicating whether the request is valid and a message describing the
729
- result of the validation.
730
- """
731
- pass
732
-
733
- def dry_run_output(self) -> list[SapioToolResult]:
734
- """
735
- Provide fixed results for a dry run of this tool. This method should not perform any actual processing of the
736
- request. It should only return example outputs that can be used to test the next tool in the plan.
737
-
738
- The default implementation of this method looks at the testing_example field of each output configuration
739
- and returns a SapioToolResult object based on the content type of the output.
740
-
741
- :return: A list of SapioToolResult objects containing example outputs for this tool. Each result in the list
742
- corresponds to a separate output from the tool.
743
- """
744
- results: list[SapioToolResult] = []
745
- for output, container_type in zip(self.output_configs, self._output_container_types):
746
- config: ToolIoConfigBasePbo = output.base_config
747
- example: ExampleContainerPbo = config.testing_example
748
- content_type: str = config.content_type
749
- match container_type:
750
- case ContainerType.BINARY:
751
- example: bytes = example.binary_example
752
- results.append(BinaryResult(binary_data=[example], content_type=content_type))
753
- case ContainerType.CSV:
754
- example: str = example.text_example
755
- results.append(CsvResult(FileUtil.tokenize_csv(example.encode())[0], content_type=content_type))
756
- case ContainerType.JSON:
757
- # The example may be in the JSONL format instead of plain JSON, so we need to use Pandas to parse
758
- # the example into plain JSON.
759
- example: str = example.text_example
760
- # Format the JSONL in a way that Pandas likes. Collapse everything into a single line, and then
761
- # split it back into multiple lines where each line is a single JSON list or dictionary.
762
- example: str = re.sub("([]}])\s*([\[{])", r"\1\n\2", example.replace("\n", "")).strip()
763
- # Read the JSONL into a Pandas DataFrame and convert it back to plain JSON.
764
- import pandas as pd
765
- with io.StringIO(example) as stream:
766
- example: str = pd.read_json(path_or_buf=stream, lines=True).to_json()
767
- results.append(JsonResult(json_data=[json.loads(example)], content_type=content_type))
768
- case ContainerType.TEXT:
769
- example: str = example.text_example
770
- results.append(TextResult(text_data=[example], content_type=content_type))
771
- return results
772
-
773
- @abstractmethod
774
- def run(self, user: SapioUser) -> list[SapioToolResult]:
775
- """
776
- Execute this tool.
777
-
778
- The request inputs can be accessed using the self.get_input_*() methods.
779
- The request settings can be accessed using the self.get_config_fields() method.
780
- The request itself can be accessed using self.request.
781
-
782
- :param user: A user object that can be used to initialize manager classes using DataMgmtServer to query the
783
- system.
784
- :return: A list of SapioToolResult objects containing the response data. Each result in the list corresponds to
785
- a separate output from the tool. Field map results do not appear as tool output in the plan manager, instead
786
- appearing as records related to the plan step during the run.
787
- """
788
- pass
789
-
790
- def log_info(self, message: str) -> None:
791
- """
792
- Log an info message for this tool. If verbose logging is enabled, this message will be included in the logs
793
- returned to the caller. Empty/None inputs will not be logged.
794
-
795
- :param message: The message to log.
796
- """
797
- if not message:
798
- return
799
- if self.verbose_logging:
800
- self.logs.append(f"INFO: {self._name}: {message}")
801
- self.logger.info(message)
802
-
803
- def log_warning(self, message: str) -> None:
804
- """
805
- Log a warning message for this tool. This message will be included in the logs returned to the caller.
806
- Empty/None inputs will not be logged.
807
-
808
- :param message: The message to log.
809
- """
810
- if not message:
811
- return
812
- self.logs.append(f"WARNING: {self._name}: {message}")
813
- self.logger.warning(message)
814
-
815
- def log_error(self, message: str) -> None:
816
- """
817
- Log an error message for this tool. This message will be included in the logs returned to the caller.
818
- Empty/None inputs will not be logged.
819
-
820
- :param message: The message to log.
821
- """
822
- if not message:
823
- return
824
- self.logs.append(f"ERROR: {self._name}: {message}")
825
- self.logger.error(message)
826
-
827
- def log_exception(self, message: str, e: Exception) -> None:
828
- """
829
- Log an exception for this tool. This message will be included in the logs returned to the caller.
830
- Empty/None inputs will not be logged.
831
-
832
- :param message: The message to log.
833
- :param e: The exception to log.
834
- """
835
- if not message and not e:
836
- return
837
- self.logs.append(f"EXCEPTION: {self._name}: {message} - {e}")
838
- self.logger.error(f"{message}\n{traceback.format_exc()}")
839
-
840
- def get_input_name(self, index: int = 0) -> str | None:
841
- """
842
- Get the name of the input from the request object.
843
-
844
- :param index: The index of the input to parse. Defaults to 0. Used for tools that accept multiple inputs.
845
- :return: The name of the input from the request object, or None if no name is set.
846
- """
847
- return self.request.input[index].item_container.container_name
848
-
849
- def get_input_binary(self, index: int = 0) -> list[bytes]:
850
- """
851
- Get the binary data from the request object.
852
-
853
- :param index: The index of the input to parse. Defaults to 0. Used for tools that accept multiple inputs.
854
- :return: The binary data from the request object.
855
- """
856
- return list(self.request.input[index].item_container.binary_container.items)
857
-
858
- def get_input_csv(self, index: int = 0) -> tuple[list[str], list[dict[str, str]]]:
859
- """
860
- Parse the CSV data from the request object.
861
-
862
- :param index: The index of the input to parse. Defaults to 0. Used for tools that accept multiple inputs.
863
- :return: A tuple containing the header row and the data rows. The header row is a list of strings representing
864
- the column names, and the data rows are a list of dictionaries where each dictionary represents a row in the
865
- CSV with the column names as keys and the corresponding values as strings.
866
- """
867
- input_data: Sequence[StepInputBatchPbo] = self.request.input
868
- ret_val: list[dict[str, str]] = []
869
- headers: Iterable[str] = input_data[index].item_container.csv_container.header.cells
870
- for row in input_data[index].item_container.csv_container.items:
871
- row_dict: dict[str, str] = {}
872
- for header, value in zip(headers, row.cells):
873
- row_dict[header] = value
874
- ret_val.append(row_dict)
875
- return list(headers), ret_val
876
-
877
- def get_input_json(self, index: int = 0) -> list[list[Any]] | list[dict[str, Any]]:
878
- """
879
- Parse the JSON data from the request object.
880
-
881
- :param index: The index of the input to parse. Defaults to 0. Used for tools that accept multiple inputs.
882
- :return: A list of parsed JSON objects. Each entry in the list represents a separate JSON entry from the input.
883
- Depending on this tool, this may be a list of dictionaries or a list of lists.
884
- """
885
- return [json.loads(x) for x in self.request.input[index].item_container.json_container.items]
886
-
887
- def get_input_text(self, index: int = 0) -> list[str]:
888
- """
889
- Parse the text data from the request object.
890
-
891
- :param index: The index of the input to parse. Defaults to 0. Used for tools that accept multiple inputs.
892
- :return: A list of text data as strings.
893
- """
894
- return list(self.request.input[index].item_container.text_container.items)
895
-
896
- def get_config_defs(self) -> dict[str, VeloxFieldDefPbo]:
897
- """
898
- Get the config field definitions for this tool.
899
-
900
- :return: A dictionary of field definitions, where the keys are the field names and the values are the
901
- VeloxFieldDefPbo objects representing the field definitions.
902
- """
903
- field_defs: dict[str, VeloxFieldDefPbo] = {}
904
- for field_def in self.to_pbo().config_fields:
905
- field_defs[field_def.data_field_name] = field_def
906
- return field_defs
907
-
908
- def get_config_fields(self) -> dict[str, FieldValue | list[str]]:
909
- """
910
- Get the configuration field values from the request object. If a field is not present in the request,
911
- the default value from the config definition will be returned.
912
-
913
- :return: A dictionary of configuration field names and their values. For multi-select selection list fields,
914
- a list of strings will be returned. For all other field types, the value will match the field type
915
- (bool for boolean fields, float for double fields, int for short, integer, long, and enum fields, and
916
- string for everything else).
917
- """
918
- config_fields: dict[str, Any] = {}
919
- raw_configs: Mapping[str, FieldValuePbo] = self.request.config_field_values
920
- for field_name, field_def in self.get_config_defs().items():
921
- field_value: FieldValue = None
922
- # If the field is present in the request, convert the protobuf value to a Python value.
923
- if field_name in raw_configs:
924
- field_value = ProtobufUtils.field_pbo_to_value(raw_configs[field_name])
925
- # If the field isn't present or is None, use the default value from the field definition.
926
- if field_value is None:
927
- field_value = ProtobufUtils.field_def_pbo_to_default_value(field_def)
928
- # If the field is a multi-select selection list, split the value by commas and strip whitespace.
929
- if field_def.data_field_type == FieldTypePbo.SELECTION and field_def.selection_properties.multi_select:
930
- field_value: list[str] = [x.strip() for x in field_value.split(',') if x.strip()]
931
- config_fields[field_name] = field_value
932
- return config_fields
933
-
934
- @staticmethod
935
- def read_from_json(json_data: list[dict[str, Any]], key: str) -> list[Any]:
936
- """
937
- From a list of dictionaries, return a list of values for the given key from each dictionary. Skips null values.
938
-
939
- :param json_data: The JSON data to read from.
940
- :param key: The key to read the values from.
941
- :return: A list of values corresponding to the given key in the JSON data.
942
- """
943
- ret_val: list[Any] = []
944
- for entry in json_data:
945
- if key in entry:
946
- value = entry[key]
947
- if isinstance(value, list):
948
- ret_val.extend(value)
949
- elif value is not None:
950
- ret_val.append(value)
951
- return ret_val