docling
LOW
maintainer RubenKelevra
0 votes
scanned 2026-08-28 07:34:53.969727
Why flagged
Uploaded within the last 14 days with 2 or fewer community votes — little peer review so far.
Triggered rules
Low
Few votes, recently uploaded
zero_votes_recent
Uploaded within the last 14 days with 2 or fewer community votes — little peer review so far.
PKGBUILD
1
# Maintainer: @RubenKelevra <rubenkelevra@gmail.com>
2
3
pkgname='docling'
4
pkgver='2.122.0'
5
pkgrel=1
6
pkgdesc='Meta-package and command-line interface for local document processing'
7
url="https://github.com/docling-project/${pkgname}"
8
license=('MIT')
9
arch=('any')
10
depends=(
11
"python-${pkgname}=${pkgver}"
12
'python-numpy>=1.24.0'
13
'python-pillow>=10.0.0'
14
'python-rtree>=1.3.0'
15
'python-scipy>=1.6.0'
16
'python-pypdfium2>=4.30.2'
17
'python-docling-parse>=7.12.0'
18
'python-docling-ibm-models>=3.13.0'
19
'python-pytorch>=2.2.2'
20
'python-torchvision'
21
'python-transformers>=5.4.0'
22
'python-lxml'
23
'python-defusedxml>=0.7.1'
24
'python-typer026>=0.12.5'
25
'python-rich>=13.0.0'
26
'python-opencv>=4.6.0'
27
)
28
makedepends=(
29
'python-build'
30
'python-installer'
31
'python-hatchling'
32
'python-wheel'
33
)
34
checkdepends=('python-pytest')
35
optdepends=(
36
'python-polyfactory>=2.22.2: document extraction support'
37
'python-docx>=1.2.0: DOCX input support'
38
'python-pptx>=1.0.2: PowerPoint input support'
39
'python-openpyxl>=3.1.5: Excel input support'
40
'python-beautifulsoup4>=4.12.3: HTML input support'
41
'python-marko>=2.1.2: Markdown input support'
42
'python-pylatexenc>=2.10: LaTeX input support'
43
'python-odfdo>=3.22.0: OpenDocument input support'
44
'python-mail-parser>=4.1.4: email input support'
45
'python-oxmsg>=0.0.2: Outlook MSG input support'
46
'python-rapidocr>=3.9.1: RapidOCR engine'
47
'python-easyocr>=1.7: EasyOCR engine'
48
'python-tesserocr>=2.7.1: Tesseract OCR engine'
49
'python-playwright>=1.58.0: dynamic HTML rendering'
50
'python-semchunk>=2.2.0: semantic chunking support'
51
'python-tree-sitter>=0.25.0: semantic chunking support'
52
'python-tree-sitter-python>=0.23.6: Python-aware semantic chunking'
53
'python-tree-sitter-c>=0.23.4: C-aware semantic chunking'
54
'python-tree-sitter-javascript>=0.23.1: JavaScript-aware semantic chunking'
55
'python-tree-sitter-typescript>=0.23.2: TypeScript-aware semantic chunking'
56
'python-httpx>=0.28: remote Docling service support'
57
'python-websockets>=14.0: remote Docling service streaming'
58
'python-dotenv>=1.0: load CLI configuration from .env files'
59
'python-openai-whisper>=20250625: audio transcription support'
60
'python-numba>=0.63.0: audio transcription support'
61
)
62
source=(
63
"${pkgname}-${pkgver}.tar.gz::${url}/archive/refs/tags/v${pkgver}.tar.gz"
64
'slim_meta_package.patch'
65
)
66
b2sums=(
67
'efb947d366cbf33ace15341879a3bbc97860e025a7acd18d4cc82f204459bc2fffc981033ba1bcf0ebfa97660d7d5e0706ef384edb2d742946e59a66bff50c8a'
68
'534eecb1b7af0d09196c0f431867fbe0344d754445f233438a87e6ce9168a12e0b18021a88a8368d34d37e57f430f50a427c4165f6cdeac973546acdab6a2929'
69
)
70
71
prepare() {
72
cd -- "${pkgname}-${pkgver}" || return 1
73
74
# Arch exposes the standard feature set through optdepends, so keep the
75
# Python meta-package dependency limited to the separately packaged SDK.
76
patch -Np1 -i "${srcdir}/slim_meta_package.patch"
77
}
78
79
build() {
80
cd -- "${pkgname}-${pkgver}/packages/docling" || return 1
81
python -m build --wheel --no-isolation
82
}
83
84
check() {
85
local _python_version
86
local _checkdir="${srcdir}/${pkgname}-check"
87
88
_python_version=$(python --version)
89
_python_version=${_python_version#Python }
90
_python_version=${_python_version%.*}
91
if [[ ! ${_python_version} =~ ^[0-9]+\.[0-9]+$ ]]; then
92
printf 'Unable to determine Python major.minor version: %s\n' "${_python_version}" >&2
93
return 1
94
fi
95
96
rm -rf -- "${_checkdir}"
97
cd -- "${pkgname}-${pkgver}" || return 1
98
python -m installer --destdir="${_checkdir}" packages/docling/dist/*.whl
99
100
PYTHONPATH="${_checkdir}/usr/lib/python${_python_version}/site-packages" \
101
HF_HUB_OFFLINE=1 TRANSFORMERS_OFFLINE=1 python -m pytest -q \
102
tests/test_cli.py::test_cli_help \
103
tests/test_cli.py::test_cli_convert_help \
104
tests/test_cli.py::test_cli_version \
105
tests/test_cli.py::test_split_list_handles_none_and_delimiters \
106
tests/test_cli.py::test_image_export_policy_covers_all_output_formats \
107
tests/test_cli.py::test_cli_invalid_ocr_mode_is_rejected \
108
tests/test_cli.py::test_cli_page_range_accepts_single_page \
109
tests/test_cli.py::test_cli_invalid_page_range_is_rejected \
110
tests/test_cli_tools.py::test_tools_help_lists_models_subcommand \
111
tests/test_cli_tools.py::test_tools_without_arguments_shows_help \
112
tests/test_cli_tools.py::test_models_without_arguments_shows_help
113
114
# Default PDF table structure uses TableFormer V1.
115
python -m pydoc docling_ibm_models.tableformer.data_management.tf_predictor >/dev/null
116
}
117
118
package() {
119
cd -- "${pkgname}-${pkgver}" || return 1
120
python -m installer --destdir="${pkgdir}" packages/docling/dist/*.whl
121
install -D -m644 -- LICENSE "${pkgdir}/usr/share/licenses/${pkgname}/LICENSE"
122
}
123
Scan history
| Scanned at (UTC) | Severity | Rules |
|---|---|---|
| 2026-08-28 07:34:53 | Low | 1 |