textproc/py-spacy: Add py-spacy 3.7.4

spaCy is a library for advanced Natural Language Processing in Python and
Cython. It's built on the very latest research, and was designed from day one to
be used in real products.

spaCy comes with pretrained pipelines and currently supports tokenization and
training for 70+ languages. It features state-of-the-art speed and neural
network models for tagging, parsing, named entity recognition, text
classification and more, multi-task learning with pretrained transformers like
BERT, as well as a production-ready training system and easy model packaging,
deployment and workflow management.

4 files changed, 63 insertions, 0 deletions

diff --git a/textproc/Makefile b/textproc/Makefile
index 0f74579122f1..9037cc0cec99 100644
--- a/textproc/Makefile
+++ b/textproc/Makefile
@@ -1557,6 +1557,7 @@
     SUBDIR += py-simplebayes
     SUBDIR += py-smartypants
     SUBDIR += py-snowballstemmer
+    SUBDIR += py-spacy
     SUBDIR += py-sparqlwrapper
     SUBDIR += py-sphinx
     SUBDIR += py-sphinx-argparse
diff --git a/textproc/py-spacy/Makefile b/textproc/py-spacy/Makefile
new file mode 100644
index 000000000000..eae8b11184a2
--- /dev/null
+++ b/textproc/py-spacy/Makefile
@@ -0,0 +1,49 @@
+PORTNAME=	spacy
+PORTVERSION=	3.7.4
+CATEGORIES=	textproc python
+MASTER_SITES=	PYPI
+PKGNAMEPREFIX=	${PYTHON_PKGNAMEPREFIX}
+
+MAINTAINER=	sunpoet@FreeBSD.org
+COMMENT=	Industrial-strength Natural Language Processing (NLP) in Python
+WWW=		https://spacy.io/ \
+		https://github.com/explosion/spaCy
+
+LICENSE=	MIT
+LICENSE_FILE=	${WRKSRC}/LICENSE
+
+BUILD_DEPENDS=	${PYTHON_PKGNAMEPREFIX}cymem>=2.0.2<2.1.0:devel/py-cymem@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}murmurhash>=0.28.0<1.1.0:devel/py-murmurhash@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}numpy>=1.25.0,1:math/py-numpy@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}preshed3>=3.0.2<3.1.0:devel/py-preshed3@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}setuptools>=0:devel/py-setuptools@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}thinc>=8.2.2<8.3.0:devel/py-thinc@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}wheel>=0:devel/py-wheel@${PY_FLAVOR}
+RUN_DEPENDS=	${PYTHON_PKGNAMEPREFIX}catalogue>=2.0.6<2.1.0:devel/py-catalogue@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}cymem>=2.0.2<2.1.0:devel/py-cymem@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}Jinja2>=0:devel/py-Jinja2@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}langcodes>=3.2.0<4.0.0:textproc/py-langcodes@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}murmurhash>=0.28.0<1.1.0:devel/py-murmurhash@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}numpy>=1.19.0,1:math/py-numpy@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}packaging>=20.0:devel/py-packaging@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}preshed3>=3.0.2<3.1.0:devel/py-preshed3@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}pydantic2>=1.7.4<3.0.0:devel/py-pydantic2@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}requests>=2.13.0<3.0.0:www/py-requests@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}setuptools>=0:devel/py-setuptools@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}smart-open>=5.2.1<7.0.0:net/py-smart-open@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}spacy-legacy>=3.0.11<3.1.0:textproc/py-spacy-legacy@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}spacy-loggers>=1.0.0<2.0.0:textproc/py-spacy-loggers@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}srsly>=2.4.3<3.0.0:devel/py-srsly@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}thinc>=8.2.2<8.3.0:devel/py-thinc@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}tqdm>=4.38.0<5.0.0:misc/py-tqdm@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}typer>=0.3.0<0.10.0:devel/py-typer@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}wasabi>=0.9.1<1.2.0:textproc/py-wasabi@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}weasel>=0.1.0<0.4.0:devel/py-weasel@${PY_FLAVOR}
+
+USES=		python
+USE_PYTHON=	autoplist concurrent cython pep517
+
+post-install:
+	${FIND} ${STAGEDIR}${PYTHON_SITELIBDIR} -name '*.so' -exec ${STRIP_CMD} {} +
+
+.include <bsd.port.mk>
diff --git a/textproc/py-spacy/distinfo b/textproc/py-spacy/distinfo
new file mode 100644
index 000000000000..fa963b5f6216
--- /dev/null
+++ b/textproc/py-spacy/distinfo
@@ -0,0 +1,3 @@
+TIMESTAMP = 1708448852
+SHA256 (spacy-3.7.4.tar.gz) = 525f2ced2e40761562c8cace93ef6a1e6e8c483f27bd564bc1b15f608efbe85b
+SIZE (spacy-3.7.4.tar.gz) = 1273006
diff --git a/textproc/py-spacy/pkg-descr b/textproc/py-spacy/pkg-descr
new file mode 100644
index 000000000000..960cadef099b
--- /dev/null
+++ b/textproc/py-spacy/pkg-descr
@@ -0,0 +1,10 @@
+spaCy is a library for advanced Natural Language Processing in Python and
+Cython. It's built on the very latest research, and was designed from day one to
+be used in real products.
+
+spaCy comes with pretrained pipelines and currently supports tokenization and
+training for 70+ languages. It features state-of-the-art speed and neural
+network models for tagging, parsing, named entity recognition, text
+classification and more, multi-task learning with pretrained transformers like
+BERT, as well as a production-ready training system and easy model packaging,
+deployment and workflow management.