Add dataclasses-json (de)serialization

Author: memowe

README.md

@@ -23,6 +23,8 @@ for line in page.all_text():
     print(line)
 ```
 
+All dataclasses are serializable with `to_dict`/`from_dict` and `to_json`/`from_json` via [dataclasses-json][dcj].
+
 ### Data model
 
 | Class | Import from |
@@ -81,6 +83,7 @@ Released under the [MIT License](LICENSE).
 [workflows]: https://github.com/SCDH/pygexml/actions/workflows/checks_tests_docs.yml
 [workflows-badge]: https://github.com/SCDH/pygexml/actions/workflows/checks_tests_docs.yml/badge.svg
 [hypothesis]: https://hypothesis.readthedocs.io
+[dcj]: https://pypi.org/project/dataclasses-json/
 [api-docs]: https://scdh.github.io/pygexml
 [api-docs-strategies]: https://scdh.github.io/pygexml/pygexml/strategies.html
 [api-docs-badge]: https://img.shields.io/badge/API%20docs-online-blue?logo=gitbook&logoColor=lightgrey

pygexml/page.py

@@ -1,6 +1,7 @@
 from re import Pattern, compile
 from warnings import warn
 from dataclasses import dataclass
+from dataclasses_json import DataClassJsonMixin
 from typing import ClassVar
 from collections.abc import Iterable
 from lxml import etree
@@ -25,7 +26,7 @@ class PageXMLError(Exception):
 
 
 @dataclass
-class Coords:
+class Coords(DataClassJsonMixin):
     polygon: Polygon
 
     # Loose regex that allows for negative values that can be handled by
@@ -79,7 +80,7 @@ def parse(cls, points_str: str) -> "Coords":
 
 
 @dataclass
-class TextLine:
+class TextLine(DataClassJsonMixin):
     id: ID
     coords: Coords
     text: str
@@ -112,7 +113,7 @@ def from_xml(cls, element: Element) -> "TextLine":
 
 
 @dataclass
-class TextRegion:
+class TextRegion(DataClassJsonMixin):
     id: ID
     coords: Coords
     textlines: dict[ID, TextLine]
@@ -149,7 +150,7 @@ def all_words(self) -> Iterable[str]:
 
 
 @dataclass
-class Page:
+class Page(DataClassJsonMixin):
     image_filename: str
     regions: dict[ID, TextRegion]
 

pyproject.toml

@@ -14,6 +14,7 @@ authors = [
 ]
 dependencies = [
     "lxml",
+    "dataclasses-json"
 ]
 
 [project.urls]