Skip to content

page

PageNode

Bases: BaseNode, Generic[PageNodeMetadata]

Represents a single page in a document, with some metadata

Source code in docprompt/schema/pipeline/node/page.py 
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
class PageNode(BaseNode, Generic[PageNodeMetadata]):
    """
    Represents a single page in a document, with some metadata
    """

    document: "DocumentNode" = Field(exclude=True, repr=False)
    page_number: PositiveInt = Field(description="The page number")
    metadata: PageNodeMetadata = Field(
        description="Application-specific metadata for the page",
        default_factory=BaseMetadata,
    )
    extra: Dict[str, Any] = Field(
        description="Extra data that can be stored on the page node",
        default_factory=dict,
    )

    ocr_results: ResultContainer[OcrPageResult] = Field(
        default_factory=_result_container_factory,
        description="The OCR results for the page",
        repr=False,
    )

    _raster_cache: Dict[str, bytes] = PrivateAttr(default_factory=dict)

    def __getstate__(self):
        state = super().__getstate__()

        state["__pydantic_private__"]["_raster_cache"] = {}

        return state

    @property
    def rasterizer(self):
        return PageRasterizer(self._raster_cache, self)

    def search(
        self, query: str, refine_to_words: bool = True, require_exact_match: bool = True
    ):
        return self.document.locator.search(
            query,
            page_number=self.page_number,
            refine_to_word=refine_to_words,
            require_exact_match=require_exact_match,
        )