13651977173

Committed 04 Mar 2025 10:48AM UTC coverage: 90.218% (+0.2%) from 90.017%

Build # 13651977173

Build Type

Pull #8906

github

Committed by

web-flow

Commit Message

Merge f9482a15a into 0d65b4caa

Pull Request Pull Request #8906: refactor!: remove `dataframe` field from `Document` and `ExtractedTableAnswer`; make `pandas` optional

Run Details

9601 of 10642 relevant lines covered (90.22%)

0.9 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

96.61

haystack/dataclasses/answer.py

# SPDX-FileCopyrightText: 2022-present deepset GmbH <info@deepset.ai>
#
# SPDX-License-Identifier: Apache-2.0

from dataclasses import asdict, dataclass, field
from typing import Any, Dict, List, Optional, Protocol, runtime_checkable

from haystack.core.serialization import default_from_dict, default_to_dict
from haystack.dataclasses.document import Document


@runtime_checkable
@dataclass
class Answer(Protocol):
    data: Any
    query: str
    meta: Dict[str, Any]

    def to_dict(self) -> Dict[str, Any]:  # noqa: D102
        ...

    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> "Answer":  # noqa: D102
        ...


@dataclass
class ExtractedAnswer:
    query: str
    score: float
    data: Optional[str] = None
    document: Optional[Document] = None
    context: Optional[str] = None
    document_offset: Optional["Span"] = None
    context_offset: Optional["Span"] = None
    meta: Dict[str, Any] = field(default_factory=dict)

    @dataclass
    class Span:
        start: int
        end: int

    def to_dict(self) -> Dict[str, Any]:
        """
        Serialize the object to a dictionary.

        :returns:
            Serialized dictionary representation of the object.
        """
        document = self.document.to_dict(flatten=False) if self.document is not None else None
        document_offset = asdict(self.document_offset) if self.document_offset is not None else None
        context_offset = asdict(self.context_offset) if self.context_offset is not None else None
        return default_to_dict(
            self,
            data=self.data,
            query=self.query,
            document=document,
            context=self.context,
            score=self.score,
            document_offset=document_offset,
            context_offset=context_offset,
            meta=self.meta,
        )

    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> "ExtractedAnswer":
        """
        Deserialize the object from a dictionary.

        :param data:
            Dictionary representation of the object.
        :returns:
            Deserialized object.
        """
        init_params = data.get("init_parameters", {})
        if (doc := init_params.get("document")) is not None:
            data["init_parameters"]["document"] = Document.from_dict(doc)

        if (offset := init_params.get("document_offset")) is not None:
            data["init_parameters"]["document_offset"] = ExtractedAnswer.Span(**offset)

        if (offset := init_params.get("context_offset")) is not None:
            data["init_parameters"]["context_offset"] = ExtractedAnswer.Span(**offset)
        return default_from_dict(cls, data)


@dataclass
class GeneratedAnswer:
    data: str
    query: str
    documents: List[Document]
    meta: Dict[str, Any] = field(default_factory=dict)

    def to_dict(self) -> Dict[str, Any]:
        """
        Serialize the object to a dictionary.

        :returns:
            Serialized dictionary representation of the object.
        """
        documents = [doc.to_dict(flatten=False) for doc in self.documents]
        return default_to_dict(self, data=self.data, query=self.query, documents=documents, meta=self.meta)

    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> "GeneratedAnswer":
        """
        Deserialize the object from a dictionary.

        :param data:
            Dictionary representation of the object.

        :returns:
            Deserialized object.
        """
        init_params = data.get("init_parameters", {})
        if (documents := init_params.get("documents")) is not None:
            data["init_parameters"]["documents"] = [Document.from_dict(d) for d in documents]

        return default_from_dict(cls, data)

1	# SPDX-FileCopyrightText: 2022-present deepset GmbH <info@deepset.ai>
2	#
3	# SPDX-License-Identifier: Apache-2.0
4
5	from dataclasses import asdict, dataclass, field	1✔
6	from typing import Any, Dict, List, Optional, Protocol, runtime_checkable	1✔
7
8	from haystack.core.serialization import default_from_dict, default_to_dict	1✔
9	from haystack.dataclasses.document import Document	1✔
10
11
12	@runtime_checkable	1✔
13	@dataclass	1✔
14	class Answer(Protocol):	1✔
15	data: Any	1✔
16	query: str	1✔
17	meta: Dict[str, Any]	1✔
18
19	def to_dict(self) -> Dict[str, Any]: # noqa: D102	1✔
20	...	×
21
22	@classmethod	1✔
23	def from_dict(cls, data: Dict[str, Any]) -> "Answer": # noqa: D102	1✔
24	...	×
25
26
27	@dataclass	1✔
28	class ExtractedAnswer:	1✔
29	query: str	1✔
30	score: float	1✔
31	data: Optional[str] = None	1✔
32	document: Optional[Document] = None	1✔
33	context: Optional[str] = None	1✔
34	document_offset: Optional["Span"] = None	1✔
35	context_offset: Optional["Span"] = None	1✔
36	meta: Dict[str, Any] = field(default_factory=dict)	1✔
37
38	@dataclass	1✔
39	class Span:	1✔
40	start: int	1✔
41	end: int	1✔
42
43	def to_dict(self) -> Dict[str, Any]:	1✔
44	"""
45	Serialize the object to a dictionary.
46
47	:returns:
48	Serialized dictionary representation of the object.
49	"""
50	document = self.document.to_dict(flatten=False) if self.document is not None else None	1✔
51	document_offset = asdict(self.document_offset) if self.document_offset is not None else None	1✔
52	context_offset = asdict(self.context_offset) if self.context_offset is not None else None	1✔
53	return default_to_dict(	1✔
54	self,
55	data=self.data,
56	query=self.query,
57	document=document,
58	context=self.context,
59	score=self.score,
60	document_offset=document_offset,
61	context_offset=context_offset,
62	meta=self.meta,
63	)
64
65	@classmethod	1✔
66	def from_dict(cls, data: Dict[str, Any]) -> "ExtractedAnswer":	1✔
67	"""
68	Deserialize the object from a dictionary.
69
70	:param data:
71	Dictionary representation of the object.
72	:returns:
73	Deserialized object.
74	"""
75	init_params = data.get("init_parameters", {})	1✔
76	if (doc := init_params.get("document")) is not None:	1✔
77	data["init_parameters"]["document"] = Document.from_dict(doc)	1✔
78
79	if (offset := init_params.get("document_offset")) is not None:	1✔
80	data["init_parameters"]["document_offset"] = ExtractedAnswer.Span(**offset)	1✔
81
82	if (offset := init_params.get("context_offset")) is not None:	1✔
83	data["init_parameters"]["context_offset"] = ExtractedAnswer.Span(**offset)	1✔
84	return default_from_dict(cls, data)	1✔
85
86
87	@dataclass	1✔
88	class GeneratedAnswer:	1✔
89	data: str	1✔
90	query: str	1✔
91	documents: List[Document]	1✔
92	meta: Dict[str, Any] = field(default_factory=dict)	1✔
93
94	def to_dict(self) -> Dict[str, Any]:	1✔
95	"""
96	Serialize the object to a dictionary.
97
98	:returns:
99	Serialized dictionary representation of the object.
100	"""
101	documents = [doc.to_dict(flatten=False) for doc in self.documents]	1✔
102	return default_to_dict(self, data=self.data, query=self.query, documents=documents, meta=self.meta)	1✔
103
104	@classmethod	1✔
105	def from_dict(cls, data: Dict[str, Any]) -> "GeneratedAnswer":	1✔
106	"""
107	Deserialize the object from a dictionary.
108
109	:param data:
110	Dictionary representation of the object.
111
112	:returns:
113	Deserialized object.
114	"""
115	init_params = data.get("init_parameters", {})	1✔
116	if (documents := init_params.get("documents")) is not None:	1✔
117	data["init_parameters"]["documents"] = [Document.from_dict(d) for d in documents]	1✔
118
119	return default_from_dict(cls, data)	1✔

deepset-ai / haystack / 13651977173

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous