2019-03-14 03:13:25 +00:00
|
|
|
{ fetchFromGitHub
|
|
|
|
, ghostscript
|
|
|
|
, img2pdf
|
|
|
|
, jbig2enc
|
|
|
|
, leptonica
|
|
|
|
, pngquant
|
|
|
|
, python3
|
|
|
|
, python3Packages
|
|
|
|
, qpdf
|
|
|
|
, stdenv
|
|
|
|
, tesseract4
|
|
|
|
, unpaper
|
|
|
|
}:
|
|
|
|
|
|
|
|
let
|
|
|
|
inherit (python3Packages) buildPythonApplication;
|
|
|
|
|
|
|
|
runtimeDeps = with python3Packages; [
|
|
|
|
ghostscript
|
|
|
|
jbig2enc
|
|
|
|
leptonica
|
|
|
|
pngquant
|
|
|
|
qpdf
|
|
|
|
tesseract4
|
|
|
|
unpaper
|
|
|
|
pillow
|
|
|
|
];
|
|
|
|
|
|
|
|
in buildPythonApplication rec {
|
|
|
|
pname = "ocrmypdf";
|
2020-01-08 18:55:00 +00:00
|
|
|
version = "9.4.0";
|
2019-03-14 03:13:25 +00:00
|
|
|
disabled = ! python3Packages.isPy3k;
|
|
|
|
|
|
|
|
src = fetchFromGitHub {
|
|
|
|
owner = "jbarlow83";
|
|
|
|
repo = "OCRmyPDF";
|
|
|
|
rev = "v${version}";
|
2020-01-08 18:55:00 +00:00
|
|
|
sha256 = "07356syg7w9m9h4yl5zk2y4s8hsb7kybbm4x9vm7mli4gw9dwv1m";
|
2019-03-14 03:13:25 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
nativeBuildInputs = with python3Packages; [
|
|
|
|
pytestrunner
|
|
|
|
setuptools
|
|
|
|
setuptools-scm-git-archive
|
|
|
|
setuptools_scm
|
|
|
|
];
|
|
|
|
|
|
|
|
propagatedBuildInputs = with python3Packages; [
|
|
|
|
cffi
|
|
|
|
chardet
|
|
|
|
img2pdf
|
|
|
|
pdfminer
|
|
|
|
pikepdf
|
2019-09-13 19:16:51 +01:00
|
|
|
pillow
|
2019-03-14 03:13:25 +00:00
|
|
|
reportlab
|
2019-09-13 19:16:51 +01:00
|
|
|
setuptools
|
|
|
|
tqdm
|
2019-03-14 03:13:25 +00:00
|
|
|
];
|
|
|
|
|
|
|
|
checkInputs = with python3Packages; [
|
|
|
|
pypdf2
|
|
|
|
pytest
|
|
|
|
pytest-helpers-namespace
|
|
|
|
pytest_xdist
|
|
|
|
pytestcov
|
|
|
|
pytestrunner
|
|
|
|
python-xmp-toolkit
|
|
|
|
setuptools
|
|
|
|
] ++ runtimeDeps;
|
|
|
|
|
|
|
|
postPatch = ''
|
|
|
|
substituteInPlace src/ocrmypdf/leptonica.py \
|
2020-01-12 23:24:59 +00:00
|
|
|
--replace '_libpath = find_library(libname)' \
|
|
|
|
'_libpath = "${stdenv.lib.getLib leptonica}/lib/liblept${stdenv.hostPlatform.extensions.sharedLibrary}"'
|
2019-03-14 03:13:25 +00:00
|
|
|
'';
|
|
|
|
|
|
|
|
# The tests take potentially 20+ minutes, depending on machine
|
|
|
|
doCheck = false;
|
|
|
|
|
|
|
|
# These tests fail and it might be upstream problem... or packaging. :)
|
|
|
|
# development is happening on macos and the pinned test versions are
|
|
|
|
# significantly newer than nixpkgs has. Program still works...
|
|
|
|
# (to the extent I've used it) -- Kiwi
|
|
|
|
checkPhase = ''
|
|
|
|
export HOME=$TMPDIR
|
|
|
|
pytest -k 'not test_force_ocr_on_pdf_with_no_images \
|
|
|
|
and not test_tesseract_crash \
|
|
|
|
and not test_tesseract_crash_autorotate \
|
|
|
|
and not test_ghostscript_pdfa_failure \
|
|
|
|
and not test_gs_render_failure \
|
|
|
|
and not test_gs_raster_failure \
|
|
|
|
and not test_bad_utf8 \
|
|
|
|
and not test_old_unpaper'
|
|
|
|
'';
|
|
|
|
|
2019-09-13 19:16:51 +01:00
|
|
|
makeWrapperArgs = [ "--prefix PATH : ${stdenv.lib.makeBinPath [ ghostscript jbig2enc pngquant qpdf tesseract4 unpaper ]}" ];
|
|
|
|
|
2019-03-14 03:13:25 +00:00
|
|
|
meta = with stdenv.lib; {
|
|
|
|
homepage = "https://github.com/jbarlow83/OCRmyPDF";
|
|
|
|
description = "Adds an OCR text layer to scanned PDF files, allowing them to be searched";
|
|
|
|
license = licenses.gpl3;
|
|
|
|
platforms = platforms.linux;
|
|
|
|
maintainers = [ maintainers.kiwi ];
|
|
|
|
};
|
|
|
|
}
|