24
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
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
76
77
78
79
80
81
82
83
84
85
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
128
129
130
131
132
133
134
135
136
137
138
139
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
172
173
174
175
176
177
178
179
180
181 | class BatchOperation(BaseModel):
class Config:
validate_assignment = True
@classmethod
def from_file(
cls,
path: str,
kiara: Union["Kiara", None] = None,
):
data = get_data_from_file(path)
pipeline_id = data.get("pipeline_name", None)
if pipeline_id is None:
name = os.path.basename(path)
if name.endswith(".json"):
name = name[0:-5]
elif name.endswith(".yaml"):
name = name[0:-5]
data["pipeline_name"] = name
alias = os.path.basename(path)
return cls.from_config(alias=alias, data=data, kiara=kiara)
@classmethod
def from_config(
cls,
alias: str,
data: Mapping[str, Any],
kiara: Union["Kiara", None],
):
data = dict(data)
inputs = data.pop("inputs", {})
save = data.pop("save", False)
pipeline_id = data.pop("pipeline_name", None)
if pipeline_id is None:
pipeline_id = str(uuid.uuid4())
if kiara is None:
kiara = Kiara.instance()
pipeline_config = PipelineConfig.from_config(
pipeline_name=pipeline_id, data=data, kiara=kiara
)
result = cls(
alias=alias,
pipeline_config=pipeline_config,
inputs=inputs,
save_defaults=save,
)
result._kiara = kiara
return result
alias: str = Field(description="The batch name/alias.")
pipeline_config: PipelineConfig = Field(
description="The configuration of the underlying pipeline."
)
inputs: Dict[str, Any] = Field(
description="The (base) inputs to use. Can be augmented before running the operation."
)
save_defaults: Dict[str, List[str]] = Field(
description="Configuration which values to save, under which alias(es).",
default_factory=dict,
)
_kiara: Kiara = PrivateAttr(default=None)
@root_validator(pre=True)
def add_alias(cls, values):
if not values.get("alias", None):
pc = values.get("pipeline_config", None)
if not pc:
raise ValueError("No pipeline config provided.")
if isinstance(pc, PipelineConfig):
alias = pc.pipeline_name
else:
alias = pc.get("pipeline_name", None)
values["alias"] = alias
return values
@validator("save_defaults", always=True, pre=True)
def validate_save(cls, save, values):
alias = values["alias"]
pipeline_config = values["pipeline_config"]
return cls.create_save_aliases(
save=save, alias=alias, pipeline_config=pipeline_config
)
@classmethod
def create_save_aliases(
cls,
save: Union[bool, None, str, Mapping],
alias: str,
pipeline_config: PipelineConfig,
) -> Mapping[str, Any]:
assert isinstance(pipeline_config, PipelineConfig)
if save in [False, None]:
save_new: Dict[str, Any] = {}
elif save is True:
field_names = pipeline_config.structure.pipeline_outputs_schema.keys()
save_new = create_save_config(field_names=field_names, aliases=alias)
elif isinstance(save, str):
field_names = pipeline_config.structure.pipeline_outputs_schema.keys()
save_new = create_save_config(field_names=field_names, aliases=save)
elif isinstance(save, Mapping):
save_new = dict(save)
else:
raise ValueError(
f"Invalid type '{type(save)}' for 'save' attribute: must be None, bool, string or Mapping."
)
return save_new
def run(
self,
inputs: Union[Mapping[str, Any], None] = None,
save: Union[None, bool, str, Mapping[str, Any]] = None,
) -> ValueMap:
pipeline = Pipeline(
structure=self.pipeline_config.structure,
kiara=self._kiara,
)
pipeline_controller = SinglePipelineBatchController(
pipeline=pipeline, job_registry=self._kiara.job_registry
)
run_inputs = dict(self.inputs)
if inputs:
run_inputs.update(inputs)
pipeline.set_pipeline_inputs(inputs=run_inputs)
pipeline_controller.process_pipeline()
result = self._kiara.data_registry.load_values(
pipeline.get_current_pipeline_outputs()
)
if save is not None:
if save is True:
save = self.save_defaults
else:
save = self.__class__.create_save_aliases(
save=save, alias=self.alias, pipeline_config=self.pipeline_config
)
self._kiara.save_values(values=result, alias_map=save)
return result
|