Add gensim package #2545 (#3326)

docs/project/changelog.md

@@ -262,6 +262,7 @@ substitutions:
   pyheif, pillow_heif, libheif, libde265 {pr}`3161`, wordcloud {pr}`3173`,
   gdal, fiona, geopandas {pr}`3213`,
   the standard library \_hashlib module {pr}`3206` , pyinstrument {pr}`3258`,
+  gensim {pr}`3326`, smart_open {pr}`3326`,
 
 - {{ Update }} Upgraded pandas to version 1.5.0.
   {pr}`3134`

packages/gensim/meta.yaml

@@ -0,0 +1,21 @@
+package:
+  name: gensim
+  version: 4.2.0
+  top-level:
+    - gensim
+source:
+  url: https://files.pythonhosted.org/packages/24/97/2197f018ee9f8ce2f071b2d9c6711c76159aead710f8d24a2bf006082a28/gensim-4.2.0.tar.gz
+  sha256: 995ebd2970a31d47c100aaac10212f47e2bf12e2b06536d38883c951ff34eef1
+  patches:
+    - patches/builtins.patch
+requirements:
+  run:
+    - numpy
+    - scipy
+    - six
+    - smart_open
+about:
+  home: http://radimrehurek.com/gensim
+  PyPI: https://pypi.org/project/gensim
+  summary: Python framework for fast Vector Space Modelling
+  license: LGPL-2.1-only

packages/gensim/patches/builtins.patch

@@ -0,0 +1,14 @@
+diff --git a/setup.py b/setup.py
+index e3ee0c3..595e8d6 100644
+--- a/setup.py
++++ b/setup.py
+@@ -102,9 +102,6 @@ class CustomBuildExt(build_ext):
+     #
+     def finalize_options(self):
+         build_ext.finalize_options(self)
+-        # Prevent numpy from thinking it is still in its setup process:
+-        # https://docs.python.org/2/library/__builtin__.html#module-__builtin__
+-        __builtins__.__NUMPY_SETUP__ = False
+ 
+         import numpy
+         self.include_dirs.append(numpy.get_include())

packages/smart_open/meta.yaml

@@ -0,0 +1,15 @@
+package:
+  name: smart_open
+  version: 6.2.0
+  top-level:
+    - smart_open
+source:
+  url: https://files.pythonhosted.org/packages/3e/07/36678c6ff0dfa6cf445d0e00bf4f013de3b86ec1a2e8bfd1e5df69b2d91d/smart_open-6.2.0-py3-none-any.whl
+  sha256: 088bf00f9327c71e549bc2f86567d3320df5d89667f009ce1c16568976068ef7
+about:
+  home: https://github.com/piskvorky/smart_open
+  PyPI: https://pypi.org/project/smart_open
+  summary:
+    Utils for streaming large files (S3, HDFS, GCS, Azure Blob Storage, gzip,
+    bz2...)
+  license: MIT