summaryrefslogtreecommitdiff
path: root/textproc/py-conllu
diff options
context:
space:
mode:
Diffstat (limited to 'textproc/py-conllu')
-rw-r--r--textproc/py-conllu/Makefile34
-rw-r--r--textproc/py-conllu/distinfo3
-rw-r--r--textproc/py-conllu/pkg-descr11
3 files changed, 48 insertions, 0 deletions
diff --git a/textproc/py-conllu/Makefile b/textproc/py-conllu/Makefile
new file mode 100644
index 000000000000..e093e001db96
--- /dev/null
+++ b/textproc/py-conllu/Makefile
@@ -0,0 +1,34 @@
+PORTNAME= conllu
+DISTVERSION= 6.0.0
+CATEGORIES= textproc python # machine-learning
+#MASTER_SITES= PYPI # incomplete tests
+PKGNAMEPREFIX= ${PYTHON_PKGNAMEPREFIX}
+
+MAINTAINER= yuri@FreeBSD.org
+COMMENT= CoNLL-U format parser for Python
+WWW= https://github.com/EmilStenstrom/conllu/
+
+LICENSE= MIT
+LICENSE_FILE= ${WRKSRC}/LICENSE
+
+BUILD_DEPENDS= ${PY_SETUPTOOLS} \
+ ${PYTHON_PKGNAMEPREFIX}wheel>0:devel/py-wheel@${PY_FLAVOR}
+
+USES= python
+USE_PYTHON= pep517 autoplist pytest
+
+USE_GITHUB= yes
+GH_ACCOUNT= EmilStenstrom
+
+TEST_ENV= ${MAKE_ENV} PYTHONPATH=${STAGEDIR}${PYTHONPREFIX_SITELIBDIR}
+TEST_WRKSRC= ${WRKSRC}/tests
+
+NO_ARCH= yes
+
+# tests as of 6.0.0 are incomplete
+# Results (0.65s):
+# 138 passed
+# 1 failed
+# - tests/test_readme.py:30 test_readme
+
+.include <bsd.port.mk>
diff --git a/textproc/py-conllu/distinfo b/textproc/py-conllu/distinfo
new file mode 100644
index 000000000000..67d02f3de4d9
--- /dev/null
+++ b/textproc/py-conllu/distinfo
@@ -0,0 +1,3 @@
+TIMESTAMP = 1763235299
+SHA256 (EmilStenstrom-conllu-6.0.0_GH0.tar.gz) = ced6404c17b4d6b217c431fe73bcf700ef01615707338969c428810bf4891a3a
+SIZE (EmilStenstrom-conllu-6.0.0_GH0.tar.gz) = 29621
diff --git a/textproc/py-conllu/pkg-descr b/textproc/py-conllu/pkg-descr
new file mode 100644
index 000000000000..69b4d6a8ef1e
--- /dev/null
+++ b/textproc/py-conllu/pkg-descr
@@ -0,0 +1,11 @@
+CoNLL-U format parser for Python.
+
+The conllu library is a lightweight, dependency-free parser for the
+CoNLL-U format. It provides a simple API to parse CoNLL-U formatted
+strings into Python data structures.
+
+CoNLL is a family of plain text file formats to annotate natural language.
+
+CoNLL-U is a plain text file format used by the Universal Dependencies project
+to annotate natural language, with each sentence's words or tokens represented
+by a line of 10 tab-separated fields.