docling-core 1.1.2__tar.gz → 1.1.4__tar.gz

{docling_core-1.1.2 → docling_core-1.1.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling-core
-Version: 1.1.2
+Version: 1.1.4
 Summary: A python library to define and validate data types in Docling.
 Home-page: https://ds4sd.github.io/
 License: MIT
@@ -28,7 +28,6 @@ Classifier: Typing :: Typed
 Requires-Dist: json-schema-for-humans (>=1.0.0,<2.0.0)
 Requires-Dist: jsonref (>=1.1.0,<2.0.0)
 Requires-Dist: jsonschema (>=4.16.0,<5.0.0)
-Requires-Dist: poetry (>=1.8.3,<2.0.0)
 Requires-Dist: pydantic (>=2.6.0,<3.0.0)
 Requires-Dist: pyproject-toml (>=0.0.10,<0.0.11)
 Requires-Dist: tabulate (>=0.9.0,<0.10.0)
@@ -47,7 +46,7 @@ Description-Content-Type: text/markdown
 [![pre-commit](https://img.shields.io/badge/pre--commit-enabled-brightgreen?logo=pre-commit&logoColor=white)](https://github.com/pre-commit/pre-commit)
 [![License MIT](https://img.shields.io/github/license/ds4sd/docling-core)](https://opensource.org/licenses/MIT)
 
-Docling Core is a library that defines the data types in [Docling](https://ds4sd.github.io), leveraging pydantic models.
+Docling Core is a library that defines the data types in [Docling](https://github.com/DS4SD/docling), leveraging pydantic models.
 
 ## Installation
 
@@ -116,13 +115,15 @@ Please read [Contributing to Docling Core](./CONTRIBUTING.md) for details.
 If you use Docling Core in your projects, please consider citing the following:
 
 ```bib
-@software{Docling,
-author = {Deep Search Team},
-month = {7},
-title = {{Docling}},
-url = {https://github.com/DS4SD/docling},
-version = {main},
-year = {2024}
+@techreport{Docling,
+  author = "Deep Search Team",
+  month = 8,
+  title = "Docling Technical Report",
+  url = "https://arxiv.org/abs/2408.09869",
+  eprint = "2408.09869",
+  doi = "10.48550/arXiv.2408.09869",
+  version = "1.0.0",
+  year = 2024
 }
 ```
 

{docling_core-1.1.2 → docling_core-1.1.4}/README.md

@@ -10,7 +10,7 @@
 [![pre-commit](https://img.shields.io/badge/pre--commit-enabled-brightgreen?logo=pre-commit&logoColor=white)](https://github.com/pre-commit/pre-commit)
 [![License MIT](https://img.shields.io/github/license/ds4sd/docling-core)](https://opensource.org/licenses/MIT)
 
-Docling Core is a library that defines the data types in [Docling](https://ds4sd.github.io), leveraging pydantic models.
+Docling Core is a library that defines the data types in [Docling](https://github.com/DS4SD/docling), leveraging pydantic models.
 
 ## Installation
 
@@ -79,13 +79,15 @@ Please read [Contributing to Docling Core](./CONTRIBUTING.md) for details.
 If you use Docling Core in your projects, please consider citing the following:
 
 ```bib
-@software{Docling,
-author = {Deep Search Team},
-month = {7},
-title = {{Docling}},
-url = {https://github.com/DS4SD/docling},
-version = {main},
-year = {2024}
+@techreport{Docling,
+  author = "Deep Search Team",
+  month = 8,
+  title = "Docling Technical Report",
+  url = "https://arxiv.org/abs/2408.09869",
+  eprint = "2408.09869",
+  doi = "10.48550/arXiv.2408.09869",
+  version = "1.0.0",
+  year = 2024
 }
 ```
 

{docling_core-1.1.2 → docling_core-1.1.4}/docling_core/types/doc/base.py

@@ -128,37 +128,28 @@ class GlmTableCell(TableCell):
     )
 
 
-class Table(AliasModel):
-    """Table."""
+class BaseCell(AliasModel):
+    """Base cell."""
 
-    num_cols: int = Field(alias="#-cols")
-    num_rows: int = Field(alias="#-rows")
     bounding_box: Optional[BoundingBoxContainer] = Field(
         default=None, alias="bounding-box", json_schema_extra=es_field(suppress=True)
     )
-    data: Optional[list[list[Union[GlmTableCell, TableCell]]]] = None
-    model: Optional[str] = None
     prov: Optional[list[Prov]] = None
     text: Optional[str] = Field(
         default=None, json_schema_extra=es_field(term_vector="with_positions_offsets")
     )
     obj_type: str = Field(
-        alias="type",
-        json_schema_extra=es_field(type="keyword", ignore_above=8191),
+        alias="type", json_schema_extra=es_field(type="keyword", ignore_above=8191)
     )
 
 
-class BaseCell(AliasModel):
-    """Base cell."""
+class Table(BaseCell):
+    """Table."""
 
-    bounding_box: Optional[BoundingBoxContainer] = Field(
-        default=None, alias="bounding-box", json_schema_extra=es_field(suppress=True)
-    )
-    prov: Optional[list[Prov]] = None
-    text: Optional[str] = None
-    obj_type: str = Field(
-        alias="type", json_schema_extra=es_field(type="keyword", ignore_above=8191)
-    )
+    num_cols: int = Field(alias="#-cols")
+    num_rows: int = Field(alias="#-rows")
+    data: Optional[list[list[Union[GlmTableCell, TableCell]]]] = None
+    model: Optional[str] = None
 
 
 class BaseText(AliasModel):

{docling_core-1.1.2 → docling_core-1.1.4}/docling_core/types/doc/document.py

@@ -311,6 +311,8 @@ class CCSDocument(
     @classmethod
     def from_dict(cls, data):
         """Validates and fixes the input data."""
+        if not isinstance(data, dict):
+            return data
         description_collection = data["description"].get("collection")
         if not description_collection:
             data["description"].setdefault("collection", {})
@@ -386,6 +388,8 @@ class ExportedCCSDocument(
     @classmethod
     def from_dict(cls, data):
         """Fix ref in main-text."""
+        if not isinstance(data, dict):
+            return data
         if data.get("main-text"):
             for item in data["main-text"]:
                 if ref := item.pop("__ref", None):
@@ -393,14 +397,28 @@ class ExportedCCSDocument(
 
         return data
 
-    def _resolve_ref(self, item: Ref) -> Optional[Table]:
-        """Return the resolved reference in case of table reference, otherwise None."""
-        result: Optional[Table] = None
+    def _resolve_ref(self, item: Ref) -> Optional[Union[BaseCell, BaseText]]:
+        """Return the resolved reference.
 
-        # NOTE: currently only resolves table refs & makes assumptions on ref parts
+        Resolved the Ref object within the document.
+        If the object is not found, None is returned.
+        """
+        result: Optional[Union[BaseCell, BaseText]] = None
+
+        # NOTE: currently only resolves refs explicitely, such that we can make
+        # assumptions on ref parts
         if item.obj_type == "table" and self.tables:
             parts = item.ref.split("/")
             result = self.tables[int(parts[2])]
+        elif item.obj_type == "figure" and self.figures:
+            parts = item.ref.split("/")
+            result = self.figures[int(parts[2])]
+        elif item.obj_type == "equation" and self.equations:
+            parts = item.ref.split("/")
+            result = self.equations[int(parts[2])]
+        elif item.obj_type == "footnote" and self.footnotes:
+            parts = item.ref.split("/")
+            result = self.footnotes[int(parts[2])]
 
         return result
 

{docling_core-1.1.2 → docling_core-1.1.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "docling-core"
-version = "1.1.2"
+version = "1.1.4"
 description = "A python library to define and validate data types in Docling."
 license = "MIT"
 authors = [
@@ -51,7 +51,6 @@ jsonschema = "^4.16.0"
 pydantic = "^2.6.0"
 jsonref = "^1.1.0"
 json-schema-for-humans = "^1.0.0"
-poetry = "^1.8.3"
 pyproject-toml = "^0.0.10"
 tabulate = "^0.9.0"