Skip to content

chore: add DoclingDocument validation rules #349

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 1 commit into
base: main
Choose a base branch
from
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 38 additions & 1 deletion docling_core/types/doc/document.py
Original file line number Diff line number Diff line change
Expand Up @@ -568,7 +568,7 @@
class RefItem(BaseModel):
"""RefItem."""

cref: str = Field(alias="$ref", pattern=_JSON_POINTER_REGEX)
cref: Annotated[str, Field(alias="$ref", pattern=_JSON_POINTER_REGEX)]

# This method makes RefItem compatible with DocItem
def get_ref(self):
Expand Down Expand Up @@ -4538,6 +4538,43 @@
self.form_items = item_lists["form_items"] # type: ignore
self.body = new_body

def _validate_rules(self):
def validate_list_group(doc: DoclingDocument, item: ListGroup):
for ref in item.children:
child = ref.resolve(doc)
if not isinstance(child, ListItem):
raise ValueError(
f"ListGroup {item.self_ref} contains non-ListItem {child.self_ref} ({child.label=})"
)

def validate_list_item(doc: DoclingDocument, item: ListItem):
if item.parent is None:
raise ValueError(f"ListItem {item.self_ref} has no parent")

Check warning on line 4552 in docling_core/types/doc/document.py

View check run for this annotation

Codecov / codecov/patch

docling_core/types/doc/document.py#L4552

Added line #L4552 was not covered by tests
if not isinstance(item.parent.resolve(doc), ListGroup):
raise ValueError(
f"ListItem {item.self_ref} has non-ListGroup parent: {item.parent.cref}"
)

def validate_group(doc: DoclingDocument, item: GroupItem):
if (
item.parent and not item.children
): # tolerate empty body, but not other groups
raise ValueError(f"Group {item.self_ref} has no children")

for item, _ in self.iterate_items(
with_groups=True,
traverse_pictures=True,
included_content_layers={c for c in ContentLayer},
):
if isinstance(item, ListGroup):
validate_list_group(self, item)

elif isinstance(item, GroupItem):
validate_group(self, item)

elif isinstance(item, ListItem):
validate_list_item(self, item)


# deprecated aliases (kept for backwards compatibility):
BasePictureData = BaseAnnotation
Expand Down
91 changes: 90 additions & 1 deletion test/test_docling_doc.py
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
from collections import deque
from copy import deepcopy
from pathlib import Path
from typing import List, Optional
from typing import List, Optional, Union
from unittest.mock import Mock

import pytest
Expand Down Expand Up @@ -1659,3 +1659,92 @@ def test_misplaced_list_items():
else:
exp_doc = DoclingDocument.load_from_yaml(exp_file)
assert doc == exp_doc


def test_list_group_with_list_items():
good_doc = DoclingDocument(name="")
l1 = good_doc.add_list_group()
good_doc.add_list_item(text="ListItem 1", parent=l1)
good_doc.add_list_item(text="ListItem 2", parent=l1)

good_doc._validate_rules()


def test_list_group_with_non_list_items():
bad_doc = DoclingDocument(name="")
l1 = bad_doc.add_list_group()
bad_doc.add_list_item(text="ListItem 1", parent=l1)
bad_doc.add_text(
text="non-ListItem in ListGroup", label=DocItemLabel.TEXT, parent=l1
)

with pytest.raises(ValueError):
bad_doc._validate_rules()


def test_list_item_outside_list_group():
def unsafe_add_list_item(
doc: DoclingDocument,
text: str,
enumerated: bool = False,
marker: Optional[str] = None,
orig: Optional[str] = None,
prov: Optional[ProvenanceItem] = None,
parent: Optional[NodeItem] = None,
content_layer: Optional[ContentLayer] = None,
formatting: Optional[Formatting] = None,
hyperlink: Optional[Union[AnyUrl, Path]] = None,
):
if not parent:
parent = doc.body

if not orig:
orig = text

text_index = len(doc.texts)
cref = f"#/texts/{text_index}"
list_item = ListItem(
text=text,
orig=orig,
self_ref=cref,
parent=parent.get_ref(),
enumerated=enumerated,
marker=marker or "",
formatting=formatting,
hyperlink=hyperlink,
)
if prov:
list_item.prov.append(prov)
if content_layer:
list_item.content_layer = content_layer

doc.texts.append(list_item)
parent.children.append(RefItem(cref=cref))

return list_item

bad_doc = DoclingDocument(name="")
unsafe_add_list_item(doc=bad_doc, text="ListItem outside ListGroup")
with pytest.raises(ValueError):
bad_doc._validate_rules()


def test_list_item_inside_list_group():
doc = DoclingDocument(name="")
l1 = doc.add_list_group()
doc.add_list_item(text="ListItem inside ListGroup", parent=l1)
doc._validate_rules()


def test_group_with_children():
good_doc = DoclingDocument(name="")
grp = good_doc.add_group()
good_doc.add_text(text="Text in group", label=DocItemLabel.TEXT, parent=grp)
good_doc._validate_rules()


def test_group_without_children():
bad_doc = DoclingDocument(name="")
bad_doc.add_group()
with pytest.raises(ValueError):
bad_doc._validate_rules()
Loading