From 57e877c3b3dcd5dc00ccf26543e5a66d2b09e318 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Radim=20=C5=98eh=C5=AF=C5=99ek?= <radimrehurek@seznam.cz>
Date: Thu, 4 Jul 2019 23:29:47 +0200
Subject: [PATCH 1/4] Fix bug with inadvertent logging setup

- module imports must never issue logging events (logging not set up yet in main app, triggers a default config init)
- warnings better, but in this case, we can more simply re-use the existing functionality from `gensim.utils`
---
 gensim/summarization/textcleaner.py | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/gensim/summarization/textcleaner.py b/gensim/summarization/textcleaner.py
index ba51b7691a..fe5cbf2931 100644
--- a/gensim/summarization/textcleaner.py
+++ b/gensim/summarization/textcleaner.py
@@ -22,21 +22,14 @@
 
 from gensim.summarization.syntactic_unit import SyntacticUnit
 from gensim.parsing.preprocessing import preprocess_documents
-from gensim.utils import tokenize
+from gensim.utils import tokenize, has_pattern
 from six.moves import range
 import re
 import logging
 
 logger = logging.getLogger(__name__)
 
-try:
-    from pattern.en import tag
-    logger.info("'pattern' package found; tag filters are available for English")
-    HAS_PATTERN = True
-except ImportError:
-    logger.info("'pattern' package not found; tag filters are not available for English")
-    HAS_PATTERN = False
-
+HAS_PATTERN = has_pattern()
 
 SEPARATOR = r'@'
 RE_SENTENCE = re.compile(r'(\S.+?[.!?])(?=\s+|$)|(\S.+?)(?=[\n]|$)', re.UNICODE)

From 497c2a244c154d8129c04f65ef36bba85d366eec Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Radim=20=C5=98eh=C5=AF=C5=99ek?= <radimrehurek@seznam.cz>
Date: Sun, 21 Jul 2019 15:51:55 +0200
Subject: [PATCH 2/4] Update textcleaner.py

---
 gensim/summarization/textcleaner.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/gensim/summarization/textcleaner.py b/gensim/summarization/textcleaner.py
index fe5cbf2931..64f7af4bda 100644
--- a/gensim/summarization/textcleaner.py
+++ b/gensim/summarization/textcleaner.py
@@ -30,6 +30,8 @@
 logger = logging.getLogger(__name__)
 
 HAS_PATTERN = has_pattern()
+if HAS_PATTERN:
+    from pattern.en import tag
 
 SEPARATOR = r'@'
 RE_SENTENCE = re.compile(r'(\S.+?[.!?])(?=\s+|$)|(\S.+?)(?=[\n]|$)', re.UNICODE)

From 0bad84489dcfa0c9dbb4e5122af2dabf7dcfcc36 Mon Sep 17 00:00:00 2001
From: Michael Penkov <m@penkov.dev>
Date: Mon, 26 Aug 2019 17:17:32 +0900
Subject: [PATCH 3/4] get Py2.7 docs to build

---
 setup.py | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/setup.py b/setup.py
index ae20cc020d..2832cd946d 100644
--- a/setup.py
+++ b/setup.py
@@ -265,6 +265,20 @@ def finalize_options(self):
 else:
     win_testenv.append('scikit-learn')
 
+docs_testenv = linux_testenv + distributed_env + [
+    'sphinx',
+    'sphinxcontrib-napoleon',
+    'plotly',
+    'pattern <= 2.6',
+    'sphinxcontrib.programoutput',
+]
+
+#
+# Get Py2.7 docs to build, see https://github.com/RaRe-Technologies/gensim/pull/2552
+#
+if sys.version_info == (2, 7):
+    docs_testenv.insert(0, 'doctools==0.14')
+
 linux_testenv = win_testenv[:]
 
 if sys.version_info < (3, 7):
@@ -385,7 +399,7 @@ def finalize_options(self):
         'distributed': distributed_env,
         'test-win': win_testenv,
         'test': linux_testenv,
-        'docs': linux_testenv + distributed_env + ['sphinx', 'sphinxcontrib-napoleon', 'plotly', 'pattern <= 2.6', 'sphinxcontrib.programoutput'],
+        'docs': docs_testenv,
     },
 
     include_package_data=True,

From f0ae8faf3f3a8a53915a6f47b03e5b3a33befd2d Mon Sep 17 00:00:00 2001
From: Michael Penkov <m@penkov.dev>
Date: Mon, 26 Aug 2019 17:23:02 +0900
Subject: [PATCH 4/4] bugfix in setup.py

---
 setup.py | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/setup.py b/setup.py
index 2832cd946d..6d2b1c02cb 100644
--- a/setup.py
+++ b/setup.py
@@ -265,6 +265,16 @@ def finalize_options(self):
 else:
     win_testenv.append('scikit-learn')
 
+
+linux_testenv = win_testenv[:]
+
+if sys.version_info < (3, 7):
+    linux_testenv.extend([
+        'tensorflow <= 1.3.0',
+        'keras >= 2.0.4, <= 2.1.4',
+        'annoy',
+    ])
+
 docs_testenv = linux_testenv + distributed_env + [
     'sphinx',
     'sphinxcontrib-napoleon',
@@ -272,22 +282,12 @@ def finalize_options(self):
     'pattern <= 2.6',
     'sphinxcontrib.programoutput',
 ]
-
 #
 # Get Py2.7 docs to build, see https://github.com/RaRe-Technologies/gensim/pull/2552
 #
 if sys.version_info == (2, 7):
     docs_testenv.insert(0, 'doctools==0.14')
 
-linux_testenv = win_testenv[:]
-
-if sys.version_info < (3, 7):
-    linux_testenv.extend([
-        'tensorflow <= 1.3.0',
-        'keras >= 2.0.4, <= 2.1.4',
-        'annoy',
-    ])
-
 ext_modules = [
     Extension('gensim.models.word2vec_inner',
         sources=['./gensim/models/word2vec_inner.c'],