pharmacode-toolkit 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.
Files changed (67) hide show
  1. pharmacode_toolkit-0.2.1/.github/workflows/publish.yml +41 -0
  2. pharmacode_toolkit-0.2.1/.github/workflows/test.yml +64 -0
  3. pharmacode_toolkit-0.2.1/.gitignore +15 -0
  4. pharmacode_toolkit-0.2.1/CHANGELOG.md +35 -0
  5. pharmacode_toolkit-0.2.1/LICENSE +21 -0
  6. pharmacode_toolkit-0.2.1/PKG-INFO +129 -0
  7. pharmacode_toolkit-0.2.1/README.md +99 -0
  8. pharmacode_toolkit-0.2.1/docs/algorithm.md +285 -0
  9. pharmacode_toolkit-0.2.1/docs/benchmark.md +138 -0
  10. pharmacode_toolkit-0.2.1/docs/cli.md +138 -0
  11. pharmacode_toolkit-0.2.1/docs/limitations.md +56 -0
  12. pharmacode_toolkit-0.2.1/docs/provenance.md +34 -0
  13. pharmacode_toolkit-0.2.1/examples/README.md +5 -0
  14. pharmacode_toolkit-0.2.1/examples/annotated/three_codes_scanned.jpg +0 -0
  15. pharmacode_toolkit-0.2.1/examples/annotated/value_1234.jpg +0 -0
  16. pharmacode_toolkit-0.2.1/examples/annotated/value_12345_rotated_90.jpg +0 -0
  17. pharmacode_toolkit-0.2.1/examples/expected/three_codes_scanned.json +117 -0
  18. pharmacode_toolkit-0.2.1/examples/expected/value_1234.json +49 -0
  19. pharmacode_toolkit-0.2.1/examples/expected/value_12345_rotated_90.json +55 -0
  20. pharmacode_toolkit-0.2.1/examples/generated/three_codes_scanned.png +0 -0
  21. pharmacode_toolkit-0.2.1/examples/generated/value_1234.png +0 -0
  22. pharmacode_toolkit-0.2.1/examples/generated/value_12345_rotated_90.png +0 -0
  23. pharmacode_toolkit-0.2.1/pyproject.toml +78 -0
  24. pharmacode_toolkit-0.2.1/src/pharmacode/__init__.py +71 -0
  25. pharmacode_toolkit-0.2.1/src/pharmacode/__main__.py +10 -0
  26. pharmacode_toolkit-0.2.1/src/pharmacode/benchmark.py +248 -0
  27. pharmacode_toolkit-0.2.1/src/pharmacode/cli.py +214 -0
  28. pharmacode_toolkit-0.2.1/src/pharmacode/decoding.py +33 -0
  29. pharmacode_toolkit-0.2.1/src/pharmacode/detection.py +318 -0
  30. pharmacode_toolkit-0.2.1/src/pharmacode/encoding.py +59 -0
  31. pharmacode_toolkit-0.2.1/src/pharmacode/imageops.py +59 -0
  32. pharmacode_toolkit-0.2.1/src/pharmacode/io.py +41 -0
  33. pharmacode_toolkit-0.2.1/src/pharmacode/models.py +280 -0
  34. pharmacode_toolkit-0.2.1/src/pharmacode/pipeline.py +77 -0
  35. pharmacode_toolkit-0.2.1/src/pharmacode/rendering.py +233 -0
  36. pharmacode_toolkit-0.2.1/src/pharmacode/segmentation.py +377 -0
  37. pharmacode_toolkit-0.2.1/src/pharmacode/visualization.py +68 -0
  38. pharmacode_toolkit-0.2.1/tests/__init__.py +0 -0
  39. pharmacode_toolkit-0.2.1/tests/conftest.py +20 -0
  40. pharmacode_toolkit-0.2.1/tests/unit/__init__.py +0 -0
  41. pharmacode_toolkit-0.2.1/tests/unit/test_benchmark.py +71 -0
  42. pharmacode_toolkit-0.2.1/tests/unit/test_cli.py +198 -0
  43. pharmacode_toolkit-0.2.1/tests/unit/test_decoding.py +21 -0
  44. pharmacode_toolkit-0.2.1/tests/unit/test_detection.py +149 -0
  45. pharmacode_toolkit-0.2.1/tests/unit/test_direction.py +32 -0
  46. pharmacode_toolkit-0.2.1/tests/unit/test_encoding.py +85 -0
  47. pharmacode_toolkit-0.2.1/tests/unit/test_examples.py +27 -0
  48. pharmacode_toolkit-0.2.1/tests/unit/test_io.py +37 -0
  49. pharmacode_toolkit-0.2.1/tests/unit/test_models.py +97 -0
  50. pharmacode_toolkit-0.2.1/tests/unit/test_pipeline.py +66 -0
  51. pharmacode_toolkit-0.2.1/tests/unit/test_rendering.py +176 -0
  52. pharmacode_toolkit-0.2.1/tests/unit/test_segmentation.py +225 -0
  53. pharmacode_toolkit-0.2.1/tests/unit/test_validation.py +67 -0
  54. pharmacode_toolkit-0.2.1/tests/unit/test_visualization.py +26 -0
  55. pharmacode_toolkit-0.2.1/tests/vision/__init__.py +0 -0
  56. pharmacode_toolkit-0.2.1/tests/vision/test_blur_and_noise.py +54 -0
  57. pharmacode_toolkit-0.2.1/tests/vision/test_captions.py +93 -0
  58. pharmacode_toolkit-0.2.1/tests/vision/test_framing.py +156 -0
  59. pharmacode_toolkit-0.2.1/tests/vision/test_multiple_codes.py +46 -0
  60. pharmacode_toolkit-0.2.1/tests/vision/test_negative_images.py +47 -0
  61. pharmacode_toolkit-0.2.1/tests/vision/test_perspective.py +21 -0
  62. pharmacode_toolkit-0.2.1/tests/vision/test_quiet_zone.py +67 -0
  63. pharmacode_toolkit-0.2.1/tests/vision/test_rotations.py +45 -0
  64. pharmacode_toolkit-0.2.1/tests/vision/test_roundtrip.py +51 -0
  65. pharmacode_toolkit-0.2.1/tests/vision/test_scale_and_dpi.py +34 -0
  66. pharmacode_toolkit-0.2.1/tools/generate_samples.py +60 -0
  67. pharmacode_toolkit-0.2.1/tools/run_benchmark.py +10 -0
@@ -0,0 +1,41 @@
1
+ name: publish
2
+
3
+ on:
4
+ push:
5
+ tags: ["v*"]
6
+
7
+ jobs:
8
+ build:
9
+ name: build sdist and wheel
10
+ runs-on: ubuntu-latest
11
+ steps:
12
+ - uses: actions/checkout@v7
13
+ - uses: actions/setup-python@v7
14
+ with:
15
+ python-version: "3.12"
16
+ - name: Build
17
+ run: python -m pip install --upgrade pip build twine && python -m build
18
+ - name: Check the archives and the README rendering
19
+ run: python -m twine check dist/*
20
+ - uses: actions/upload-artifact@v7
21
+ with:
22
+ name: dist
23
+ path: dist/
24
+
25
+ publish:
26
+ name: publish to PyPI
27
+ needs: build
28
+ runs-on: ubuntu-latest
29
+ environment:
30
+ name: pypi
31
+ url: https://pypi.org/project/pharmacode-toolkit/
32
+ permissions:
33
+ id-token: write
34
+ contents: read
35
+ steps:
36
+ - uses: actions/download-artifact@v8
37
+ with:
38
+ name: dist
39
+ path: dist/
40
+ - name: Publish (trusted publishing, no token)
41
+ uses: pypa/gh-action-pypi-publish@release/v1
@@ -0,0 +1,64 @@
1
+ name: test
2
+
3
+ on:
4
+ push:
5
+ branches: [main]
6
+ tags: ["v*"]
7
+ pull_request:
8
+
9
+ jobs:
10
+ test:
11
+ name: ${{ matrix.os }} / py${{ matrix.python }}
12
+ runs-on: ${{ matrix.os }}
13
+ strategy:
14
+ fail-fast: false
15
+ matrix:
16
+ os: [ubuntu-latest, windows-latest]
17
+ python: ["3.10", "3.12"]
18
+ steps:
19
+ - uses: actions/checkout@v7
20
+ - uses: actions/setup-python@v7
21
+ with:
22
+ python-version: ${{ matrix.python }}
23
+ - name: Install
24
+ run: python -m pip install --upgrade pip && python -m pip install -e .[dev]
25
+ - name: Lint
26
+ run: ruff check . && ruff format --check .
27
+ - name: Tests
28
+ run: pytest
29
+ - name: Quick benchmark
30
+ run: pharmacode benchmark --seed 20260919 --output benchmark-output --quick
31
+ - name: Upload benchmark failures
32
+ if: always()
33
+ uses: actions/upload-artifact@v7
34
+ with:
35
+ name: benchmark-failures-${{ matrix.os }}-py${{ matrix.python }}
36
+ path: benchmark-output/failures
37
+ if-no-files-found: ignore
38
+
39
+ package:
40
+ name: build and install smoke test
41
+ runs-on: ${{ matrix.os }}
42
+ strategy:
43
+ matrix:
44
+ os: [ubuntu-latest, windows-latest]
45
+ steps:
46
+ - uses: actions/checkout@v7
47
+ - uses: actions/setup-python@v7
48
+ with:
49
+ python-version: "3.12"
50
+ - name: Build wheel
51
+ run: python -m pip install --upgrade pip build && python -m build
52
+ - name: Install wheel into a fresh environment and decode a generated code
53
+ shell: bash
54
+ run: |
55
+ python -m venv smoke
56
+ if [ -x smoke/bin/python ]; then PY=smoke/bin/python; else PY=smoke/Scripts/python.exe; fi
57
+ "$PY" -m pip install --quiet dist/*.whl
58
+ "$PY" -m pharmacode generate --value 12345 --dpi 300 --output smoke.png
59
+ "$PY" -m pharmacode decode smoke.png --dpi 300 --json smoke.json
60
+ "$PY" -c "import json; d = json.load(open('smoke.json')); assert d['detections'][0]['value'] == 12345, d"
61
+ - uses: actions/upload-artifact@v7
62
+ with:
63
+ name: wheel-${{ matrix.os }}
64
+ path: dist/*.whl
@@ -0,0 +1,15 @@
1
+ __pycache__/
2
+ *.py[cod]
3
+ .venv/
4
+ venv/
5
+ dist/
6
+ build/
7
+ *.egg-info/
8
+ .pytest_cache/
9
+ .ruff_cache/
10
+ bench/
11
+ benchmark-output*/
12
+ tests/fixtures/generated/
13
+ *.log
14
+ .DS_Store
15
+ Thumbs.db
@@ -0,0 +1,35 @@
1
+ # Changelog
2
+
3
+ ## 0.2.1 - 2026-09-20
4
+
5
+ - Documentation describes the real-world sample check made before each release.
6
+ - README renders on PyPI: absolute image and documentation links, `pip install pharmacode-toolkit`.
7
+ - Releases are published to PyPI by a tag-triggered workflow using trusted publishing.
8
+ - CI actions updated to their current major versions.
9
+
10
+ ## 0.2.0 - 2026-09-19
11
+
12
+ Fixes and options that came out of checking the decoder on publicly available
13
+ sample images.
14
+
15
+ - Human-readable captions printed next to a code no longer break the bar-height check.
16
+ - Codes that fill the frame keep their wide bars: the background kernel is sized from the strokes.
17
+ - A thin rule crossing the bars is removed by a directional opening when nothing else is found.
18
+ - `decode --allow-cropped-quiet-zone` downgrades a quiet zone cut by the image edge to a warning.
19
+ - `decode --min-confidence` reports weak detections as `LOW_CONFIDENCE` errors.
20
+ - The documentation states how the decoder was checked beyond the synthetic benchmark.
21
+
22
+ ## 0.1.0 - 2026-09-18
23
+
24
+ First release.
25
+
26
+ - Encoding and decoding of one-track Pharmacode values 3..131070 with both reading directions.
27
+ - Synthetic renderer with Laetus standard and miniature dimensions, DPI, rotation, blur, noise,
28
+ JPEG, contrast, illumination, perspective, scaling, multi-code scenes and negative samples.
29
+ - Detector and decoder for PNG, JPEG and TIFF: 0/90/180/270 degrees and small tilts,
30
+ quiet-zone and geometry validation, stable error codes, confidence.
31
+ - `pharmacode generate`, `pharmacode decode` (JSON, annotated image, exit codes) and
32
+ `pharmacode benchmark` (condition matrix with failure artefacts); `python -m pharmacode ...`
33
+ is equivalent to the `pharmacode` console script.
34
+ - Documentation: algorithm and threshold provenance, CLI, benchmark, limitations, provenance.
35
+ - CI on Linux and Windows with a wheel install smoke test.
@@ -0,0 +1,21 @@
1
+ MIT License
2
+
3
+ Copyright (c) 2026 Konstantin Dotsenko
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
@@ -0,0 +1,129 @@
1
+ Metadata-Version: 2.5
2
+ Name: pharmacode-toolkit
3
+ Version: 0.2.1
4
+ Summary: Generate, detect and decode one-track Pharmacode barcodes in images
5
+ Project-URL: Homepage, https://github.com/Dotsen/pharmacode-toolkit
6
+ Project-URL: Repository, https://github.com/Dotsen/pharmacode-toolkit
7
+ Project-URL: Issues, https://github.com/Dotsen/pharmacode-toolkit/issues
8
+ Project-URL: Changelog, https://github.com/Dotsen/pharmacode-toolkit/blob/main/CHANGELOG.md
9
+ Project-URL: Author, https://www.linkedin.com/in/dotsen/
10
+ Author-email: Konstantin Dotsenko <konstantin@dotsenko.pro>
11
+ License-Expression: MIT
12
+ License-File: LICENSE
13
+ Keywords: barcode,computer-vision,laetus,opencv,pharmacode
14
+ Classifier: Development Status :: 3 - Alpha
15
+ Classifier: Intended Audience :: Developers
16
+ Classifier: Programming Language :: Python :: 3
17
+ Classifier: Programming Language :: Python :: 3.10
18
+ Classifier: Programming Language :: Python :: 3.11
19
+ Classifier: Programming Language :: Python :: 3.12
20
+ Classifier: Programming Language :: Python :: 3.13
21
+ Classifier: Topic :: Scientific/Engineering :: Image Recognition
22
+ Requires-Python: >=3.10
23
+ Requires-Dist: numpy>=1.26
24
+ Requires-Dist: opencv-python-headless>=4.8
25
+ Provides-Extra: dev
26
+ Requires-Dist: build>=1.2; extra == 'dev'
27
+ Requires-Dist: pytest>=8; extra == 'dev'
28
+ Requires-Dist: ruff>=0.5; extra == 'dev'
29
+ Description-Content-Type: text/markdown
30
+
31
+ # pharmacode-toolkit
32
+
33
+ Generate, detect and decode one-track [Pharmacode](https://en.wikipedia.org/wiki/Pharmacode)
34
+ barcodes in PNG, JPEG and TIFF images. Pure Python on numpy and OpenCV.
35
+
36
+ ![Three codes found and annotated](https://raw.githubusercontent.com/Dotsen/pharmacode-toolkit/main/examples/annotated/three_codes_scanned.jpg)
37
+
38
+ ```json
39
+ {
40
+ "detections": [
41
+ {"bbox": {"x": 45, "y": 63, "width": 149, "height": 94}, "orientation_deg": 0.0,
42
+ "bars": ["narrow", "wide", "wide", "wide", "narrow", "narrow"],
43
+ "bar_widths_px": [3, 9, 9, 9, 3, 3],
44
+ "value": 91, "mirror_value": 77, "confidence": 1.0, "warnings": []}
45
+ ],
46
+ "errors": []
47
+ }
48
+ ```
49
+
50
+ ## What it does
51
+
52
+ - Encodes any value from 3 to 131070 into the 2..16 narrow/wide bars of the format.
53
+ - Renders synthetic codes with the Laetus physical dimensions at any DPI, with
54
+ optional rotation, blur, noise, JPEG artefacts, contrast loss, uneven lighting,
55
+ perspective and scaling, all seeded and reproducible.
56
+ - Finds one or more codes in an image at 0, 90, 180 and 270 degrees and small tilts.
57
+ - Classifies bars, checks quiet zones and geometry, and reports **both** reading
58
+ directions, because the format has no start or stop pattern.
59
+ - Writes a JSON result with stable error codes and an annotated image.
60
+ - Ships a benchmark that reports a matrix of conditions, not one number.
61
+
62
+ ## Install
63
+
64
+ ```bash
65
+ pip install pharmacode-toolkit
66
+ ```
67
+
68
+ From a clone of the repository: `pip install .` (or `pip install -e .[dev]` for
69
+ development). Requires Python 3.10 or newer. Runtime dependencies: numpy,
70
+ opencv-python-headless.
71
+
72
+ ## Quick start
73
+
74
+ ```bash
75
+ pharmacode generate --value 12345 --dpi 300 --output sample.png
76
+ pharmacode decode sample.png --dpi 300
77
+ pharmacode decode sample.png --dpi 300 --json result.json --annotated result.png
78
+ pharmacode decode sample.png --dpi 300 --min-confidence 0.8
79
+ ```
80
+
81
+ `python -m pharmacode ...` works the same as the `pharmacode` command above,
82
+ for environments where installing a console script is inconvenient.
83
+
84
+ From Python:
85
+
86
+ ```python
87
+ from pharmacode import DecoderConfig, decode_image, load_image
88
+
89
+ result = decode_image(load_image("sample.png"), DecoderConfig(dpi=300))
90
+ for code in result.detections:
91
+ print(code.value, code.mirror_value, code.confidence)
92
+ ```
93
+
94
+ ## Supported input
95
+
96
+ 8-bit grayscale or colour PNG, JPEG and TIFF; dark bars on a light background;
97
+ codes upright, rotated by multiples of 90 degrees, or tilted a few degrees.
98
+ Pass `--dpi` whenever you know the resolution: it enables physical checks and
99
+ reliable classification of codes that use a single bar width.
100
+
101
+ ## Limitations
102
+
103
+ See [docs/limitations.md](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/limitations.md). In short: no inverted codes, no
104
+ two-track or colour Pharmacode, no DPI from file metadata. Thresholds come from the
105
+ specification and a synthetic benchmark; on top of that the decoder is checked
106
+ against real-world sample images (codes as they appear in the wild: with captions,
107
+ tightly cropped, with a rule drawn across the bars, inside a chart of symbologies).
108
+
109
+ ## Documentation
110
+
111
+ - [Algorithm and threshold provenance](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/algorithm.md)
112
+ - [Command line and exit codes](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/cli.md)
113
+ - [Benchmark](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/benchmark.md)
114
+ - [Provenance](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/provenance.md)
115
+ - [Changelog](https://github.com/Dotsen/pharmacode-toolkit/blob/main/CHANGELOG.md)
116
+
117
+ ## Disclaimer
118
+
119
+ This project is an independent implementation from public descriptions of the
120
+ format. It is **not validated** for regulated pharmaceutical packaging control and
121
+ must not be used as the sole check in such a process.
122
+
123
+ ## Author
124
+
125
+ Konstantin Dotsenko ([LinkedIn](https://www.linkedin.com/in/dotsen/)).
126
+
127
+ ## License
128
+
129
+ MIT, see [LICENSE](https://github.com/Dotsen/pharmacode-toolkit/blob/main/LICENSE).
@@ -0,0 +1,99 @@
1
+ # pharmacode-toolkit
2
+
3
+ Generate, detect and decode one-track [Pharmacode](https://en.wikipedia.org/wiki/Pharmacode)
4
+ barcodes in PNG, JPEG and TIFF images. Pure Python on numpy and OpenCV.
5
+
6
+ ![Three codes found and annotated](https://raw.githubusercontent.com/Dotsen/pharmacode-toolkit/main/examples/annotated/three_codes_scanned.jpg)
7
+
8
+ ```json
9
+ {
10
+ "detections": [
11
+ {"bbox": {"x": 45, "y": 63, "width": 149, "height": 94}, "orientation_deg": 0.0,
12
+ "bars": ["narrow", "wide", "wide", "wide", "narrow", "narrow"],
13
+ "bar_widths_px": [3, 9, 9, 9, 3, 3],
14
+ "value": 91, "mirror_value": 77, "confidence": 1.0, "warnings": []}
15
+ ],
16
+ "errors": []
17
+ }
18
+ ```
19
+
20
+ ## What it does
21
+
22
+ - Encodes any value from 3 to 131070 into the 2..16 narrow/wide bars of the format.
23
+ - Renders synthetic codes with the Laetus physical dimensions at any DPI, with
24
+ optional rotation, blur, noise, JPEG artefacts, contrast loss, uneven lighting,
25
+ perspective and scaling, all seeded and reproducible.
26
+ - Finds one or more codes in an image at 0, 90, 180 and 270 degrees and small tilts.
27
+ - Classifies bars, checks quiet zones and geometry, and reports **both** reading
28
+ directions, because the format has no start or stop pattern.
29
+ - Writes a JSON result with stable error codes and an annotated image.
30
+ - Ships a benchmark that reports a matrix of conditions, not one number.
31
+
32
+ ## Install
33
+
34
+ ```bash
35
+ pip install pharmacode-toolkit
36
+ ```
37
+
38
+ From a clone of the repository: `pip install .` (or `pip install -e .[dev]` for
39
+ development). Requires Python 3.10 or newer. Runtime dependencies: numpy,
40
+ opencv-python-headless.
41
+
42
+ ## Quick start
43
+
44
+ ```bash
45
+ pharmacode generate --value 12345 --dpi 300 --output sample.png
46
+ pharmacode decode sample.png --dpi 300
47
+ pharmacode decode sample.png --dpi 300 --json result.json --annotated result.png
48
+ pharmacode decode sample.png --dpi 300 --min-confidence 0.8
49
+ ```
50
+
51
+ `python -m pharmacode ...` works the same as the `pharmacode` command above,
52
+ for environments where installing a console script is inconvenient.
53
+
54
+ From Python:
55
+
56
+ ```python
57
+ from pharmacode import DecoderConfig, decode_image, load_image
58
+
59
+ result = decode_image(load_image("sample.png"), DecoderConfig(dpi=300))
60
+ for code in result.detections:
61
+ print(code.value, code.mirror_value, code.confidence)
62
+ ```
63
+
64
+ ## Supported input
65
+
66
+ 8-bit grayscale or colour PNG, JPEG and TIFF; dark bars on a light background;
67
+ codes upright, rotated by multiples of 90 degrees, or tilted a few degrees.
68
+ Pass `--dpi` whenever you know the resolution: it enables physical checks and
69
+ reliable classification of codes that use a single bar width.
70
+
71
+ ## Limitations
72
+
73
+ See [docs/limitations.md](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/limitations.md). In short: no inverted codes, no
74
+ two-track or colour Pharmacode, no DPI from file metadata. Thresholds come from the
75
+ specification and a synthetic benchmark; on top of that the decoder is checked
76
+ against real-world sample images (codes as they appear in the wild: with captions,
77
+ tightly cropped, with a rule drawn across the bars, inside a chart of symbologies).
78
+
79
+ ## Documentation
80
+
81
+ - [Algorithm and threshold provenance](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/algorithm.md)
82
+ - [Command line and exit codes](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/cli.md)
83
+ - [Benchmark](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/benchmark.md)
84
+ - [Provenance](https://github.com/Dotsen/pharmacode-toolkit/blob/main/docs/provenance.md)
85
+ - [Changelog](https://github.com/Dotsen/pharmacode-toolkit/blob/main/CHANGELOG.md)
86
+
87
+ ## Disclaimer
88
+
89
+ This project is an independent implementation from public descriptions of the
90
+ format. It is **not validated** for regulated pharmaceutical packaging control and
91
+ must not be used as the sole check in such a process.
92
+
93
+ ## Author
94
+
95
+ Konstantin Dotsenko ([LinkedIn](https://www.linkedin.com/in/dotsen/)).
96
+
97
+ ## License
98
+
99
+ MIT, see [LICENSE](https://github.com/Dotsen/pharmacode-toolkit/blob/main/LICENSE).