docling-ocr-onnxtr 0.2.0__tar.gz → 0.2.1__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/PKG-INFO +1 -1
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr/onnxtr_model.py +1 -1
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr/options.py +1 -1
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr/plugin.py +1 -1
- docling_ocr_onnxtr-0.2.1/docling_ocr_onnxtr/version.py +1 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/PKG-INFO +1 -1
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/setup.py +2 -2
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/tests/test_plugin.py +3 -10
- docling_ocr_onnxtr-0.2.0/docling_ocr_onnxtr/version.py +0 -1
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/LICENSE +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/README.md +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr/__init__.py +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr/py.typed +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/SOURCES.txt +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/dependency_links.txt +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/entry_points.txt +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/requires.txt +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/top_level.txt +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/zip-safe +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/pyproject.toml +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/setup.cfg +0 -0
- {docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/tests/test_pipeline_invalid_cases.py +0 -0
|
@@ -0,0 +1 @@
|
|
|
1
|
+
__version__ = 'v0.2.1'
|
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
# Copyright (C) 2021-
|
|
1
|
+
# Copyright (C) 2021-2026, Felix Dittrich.
|
|
2
2
|
|
|
3
3
|
# This program is licensed under the Apache License 2.0.
|
|
4
4
|
# See LICENSE or go to <https://opensource.org/licenses/Apache-2.0> for full license details.
|
|
@@ -9,7 +9,7 @@ from pathlib import Path
|
|
|
9
9
|
from setuptools import setup
|
|
10
10
|
|
|
11
11
|
PKG_NAME = "docling_ocr_onnxtr"
|
|
12
|
-
VERSION = os.getenv("BUILD_VERSION", "0.2.
|
|
12
|
+
VERSION = os.getenv("BUILD_VERSION", "0.2.1a0")
|
|
13
13
|
|
|
14
14
|
|
|
15
15
|
if __name__ == "__main__":
|
|
@@ -15,9 +15,8 @@ from docling.document_converter import DocumentConverter, PdfFormatOption
|
|
|
15
15
|
from docling_ocr_onnxtr import OnnxtrOcrOptions
|
|
16
16
|
|
|
17
17
|
from .test_data_gen_flag import GEN_TEST_DATA
|
|
18
|
-
from .verify_utils import
|
|
18
|
+
from .verify_utils import verify_conversion_result_v2
|
|
19
19
|
|
|
20
|
-
GENERATE_V1 = GEN_TEST_DATA
|
|
21
20
|
GENERATE_V2 = GEN_TEST_DATA
|
|
22
21
|
|
|
23
22
|
|
|
@@ -73,7 +72,7 @@ def test_e2e_conversions(ocr_options: OcrOptions):
|
|
|
73
72
|
print(f"Converting with ocr_engine: {ocr_options.kind}, language: {ocr_options.lang}")
|
|
74
73
|
converter = get_converter(ocr_options=ocr_options)
|
|
75
74
|
for pdf_path in pdf_paths:
|
|
76
|
-
if not ocr_options.auto_correct_orientation and "rotated" in pdf_path.name:
|
|
75
|
+
if not ocr_options.auto_correct_orientation and ("rotated" in pdf_path.name or "rotation" in pdf_path.name):
|
|
77
76
|
# Skip rotated PDFs if orientation correction is disabled
|
|
78
77
|
print(f"Skipping {pdf_path} due to orientation correction settings.")
|
|
79
78
|
continue
|
|
@@ -82,12 +81,6 @@ def test_e2e_conversions(ocr_options: OcrOptions):
|
|
|
82
81
|
doc_result: ConversionResult = converter.convert(pdf_path)
|
|
83
82
|
|
|
84
83
|
try:
|
|
85
|
-
verify_conversion_result_v1(
|
|
86
|
-
input_path=pdf_path,
|
|
87
|
-
doc_result=doc_result,
|
|
88
|
-
generate=GENERATE_V1,
|
|
89
|
-
fuzzy=True,
|
|
90
|
-
)
|
|
91
84
|
verify_conversion_result_v2(
|
|
92
85
|
input_path=pdf_path,
|
|
93
86
|
doc_result=doc_result,
|
|
@@ -95,7 +88,7 @@ def test_e2e_conversions(ocr_options: OcrOptions):
|
|
|
95
88
|
fuzzy=True,
|
|
96
89
|
)
|
|
97
90
|
except AssertionError as e:
|
|
98
|
-
if "rotated" in pdf_path.name:
|
|
91
|
+
if "rotated" in pdf_path.name or "rotation" in pdf_path.name or ocr_options.auto_correct_orientation:
|
|
99
92
|
pytest.xfail(f"Skipping {pdf_path} due to orientation correction settings: {e}")
|
|
100
93
|
else:
|
|
101
94
|
raise # Unexpected failure — re-raise the error
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
__version__ = 'v0.2.0'
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/SOURCES.txt
RENAMED
|
File without changes
|
|
File without changes
|
{docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/entry_points.txt
RENAMED
|
File without changes
|
{docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/requires.txt
RENAMED
|
File without changes
|
{docling_ocr_onnxtr-0.2.0 → docling_ocr_onnxtr-0.2.1}/docling_ocr_onnxtr.egg-info/top_level.txt
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|