feat: support for python 3.14 (#131)

Signed-off-by: Michele Dolfi <[email protected]>

Author: dolfim-ibm

.github/workflows/checks.yml

@@ -19,7 +19,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ['3.9', '3.10', '3.11', '3.12', '3.13']
+        python-version: ['3.9', '3.10', '3.11', '3.12', '3.13', '3.14']
     steps:
       - uses: actions/checkout@v4
       - name: Cache Hugging Face models

pyproject.toml

@@ -27,13 +27,14 @@ classifiers = [
   "Programming Language :: Python :: 3.11",
   "Programming Language :: Python :: 3.12",
   "Programming Language :: Python :: 3.13",
+  "Programming Language :: Python :: 3.14",
 ]
 requires-python = '>=3.9,<4.0'
 dependencies = [
   'torch (>=2.2.2,<3.0.0)',
   'torchvision (>=0,<1)',
-  'jsonlines (>=3.1.0,<4.0.0)',
-  'Pillow (>=10.0.0,<12.0.0)',
+  'jsonlines (>=3.1.0,<5.0.0)',
+  'Pillow (>=10.0.0,<13.0.0)',
   'tqdm (>=4.64.0,<5.0.0)',
   'huggingface_hub (>=0.23,<1)',
   'safetensors[torch] (>=0.4.3,<1)',
@@ -74,7 +75,7 @@ dev = [
     "pytest-dependency~=0.6",
     "pytest-xdist~=3.3",
     "python-semantic-release~=7.32",
-    "datasets~=3.2",
+    'datasets~=3.2; python_version < "3.14"',
 ]
 
 [tool.uv]

tests/test_reading_order.py

@@ -1,24 +1,17 @@
-import os
-import json
-import glob
 import copy
 
 import logging
 
-from pathlib import Path
+import sys
 
-import numpy as np
 import pytest
-from PIL import Image
-
-from datasets import load_dataset
 
 from typing import List, Dict
 import random
 
 from docling_ibm_models.reading_order.reading_order_rb import PageElement, ReadingOrderPredictor
 
-from docling_core.types.doc.document import DoclingDocument, DocItem, RefItem, TextItem, ContentLayer
+from docling_core.types.doc.document import DoclingDocument, DocItem, TextItem, ContentLayer
 
 # Configure logging
 logging.basicConfig(
@@ -60,6 +53,11 @@ def spearman_rank_correlation(arr1, arr2):
     return rho
 
 def test_readingorder():
+    if sys.version_info >= (3, 14):
+        pytest.skip("Pyarrow is not yet available for Python 3.14, hence we cannot load the dataset.")
+    
+    from datasets import load_dataset
+
 
     ro_scores, caption_scores, footnote_scores = [], [], []