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 | class ArrayType(AnyType[KiaraArray, DataTypeConfig]):
"""An array, in most cases used as a column within a table.
Internally, this type uses the [KiaraArray][kiara_plugin.tabular.models.array.KiaraArray] wrapper class to manage array data. This wrapper class, in turn, uses an [Apache Arrow](https://arrow.apache.org) [Array](https://arrow.apache.org/docs/python/generated/pyarrow.Array.html#pyarrow.Array) to store the data in memory (and on disk).
"""
_data_type_name = "array"
@classmethod
def python_class(cls) -> Type:
return KiaraArray
def parse_python_obj(self, data: Any) -> KiaraArray:
return KiaraArray.create_array(data)
def _validate(cls, value: Any) -> None:
if not isinstance(value, (KiaraArray)):
raise Exception(
f"Invalid type '{type(value).__name__}', must be an instance of the 'KiaraArray' class."
)
def serialize(self, data: KiaraArray) -> SerializedData:
import pyarrow as pa
# TODO: make sure temp dir is in the same partition as file store
temp_f = tempfile.mkdtemp()
def cleanup():
shutil.rmtree(temp_f, ignore_errors=True)
atexit.register(cleanup)
column: pa.Array = data.arrow_array
file_name = os.path.join(temp_f, "array.arrow")
store_array(array_obj=column, file_name=file_name, column_name="array")
chunks = {"array.arrow": {"type": "file", "codec": "raw", "file": file_name}}
serialized_data = {
"data_type": self.data_type_name,
"data_type_config": self.type_config.dict(),
"data": chunks,
"serialization_profile": "feather",
"metadata": {
"environment": {},
"deserialize": {
"python_object": {
"module_type": "load.array",
"module_config": {
"value_type": "array",
"target_profile": "python_object",
"serialization_profile": "feather",
},
}
},
},
}
serialized = SerializationResult(**serialized_data)
return serialized
def pretty_print_as__terminal_renderable(
self, value: Value, render_config: Mapping[str, Any]
) -> Any:
max_rows = render_config.get(
"max_no_rows", DEFAULT_PRETTY_PRINT_CONFIG["max_no_rows"]
)
max_row_height = render_config.get(
"max_row_height", DEFAULT_PRETTY_PRINT_CONFIG["max_row_height"]
)
max_cell_length = render_config.get(
"max_cell_length", DEFAULT_PRETTY_PRINT_CONFIG["max_cell_length"]
)
half_lines: Union[int, None] = None
if max_rows:
half_lines = int(max_rows / 2)
import pyarrow as pa
array: pa.Array = value.data.arrow_array
temp_table = pa.Table.from_arrays(arrays=[array], names=["array"])
atw = ArrowTabularWrap(temp_table)
result = atw.as_terminal_renderable(
rows_head=half_lines,
rows_tail=half_lines,
max_row_height=max_row_height,
max_cell_length=max_cell_length,
show_table_header=False,
)
return result
def pretty_print_as__string(
self, value: Value, render_config: Mapping[str, Any]
) -> Any:
max_rows = render_config.get(
"max_no_rows", DEFAULT_PRETTY_PRINT_CONFIG["max_no_rows"]
)
max_row_height = render_config.get(
"max_row_height", DEFAULT_PRETTY_PRINT_CONFIG["max_row_height"]
)
max_cell_length = render_config.get(
"max_cell_length", DEFAULT_PRETTY_PRINT_CONFIG["max_cell_length"]
)
half_lines: Union[int, None] = None
if max_rows:
half_lines = int(max_rows / 2)
import pyarrow as pa
array: pa.Array = value.data.arrow_array
temp_table = pa.Table.from_arrays(arrays=[array], names=["array"])
atw = ArrowTabularWrap(temp_table)
result = atw.as_string(
rows_head=half_lines,
rows_tail=half_lines,
max_row_height=max_row_height,
max_cell_length=max_cell_length,
)
return result
|